2018 POCT Course Notes

Course Notes
Point of Care Patient Testing
ONPS 2425 and ONPS 2426
Topics I to XIII written by Mr Ian Farrance, FAACB

Topic XIV written by A/Prof Danilla Grando
Book Editor: A/Prof Ronda Greaves
School of Health & Biomedical Sciences (HE)
With support from
School of Vocational Health and Sciences (TAFE)

Introduction
This book provides additional ready for the POCT course. It should be used as a
reference document throughout the semester.
Index
1. Topic Ia - Introduction to POCT 3
2. Topic Ib – Market Segments 14
3. Topic II - Pre-analytical Considerations 18
4. Topic III - Analytical Considerations 41
5. Topic IV - Regulatory Requirements 76
6. Topic V - Statistics 89
7. Topic VI - IQC & EQA 116
8. Topic VII - Biological Variation & Reference Intervals 128
9. Topic VIII - MU & Fitness for Purpose 136
10. Introduction to Clinical Applications 146
11. Topic IX - Blood Gases & Acid Base 150
12. Topic X - Diabetes, Glucose & HbA1c 177
13. Topic XI – Haematology & Coagulation 192
14. Topic XII - Drugs of Abuse 212
15. Topic XIII - Cardiac Disease 237
16. Topic XIV - Microbiology 264
R. Greaves POCT Page 2

Course Code: ONPS2425 / ONPS2426
Topic 1 – Introduction to POCT
 Course learning objectives

 What is point of care testing (POCT)
 Situations in which POCT may be advantageous
 Evidence based POCT and current best practice for patient treatment
 Visually read indicator strips (“dip-stick”,“dip-stix”)
 Technology and overview with respect to types of POCT devices
 Functionality of hand-held devices and bench top devices
 Review – Policy, procedures and guidelines for POCT
 Where to obtain information
 Suggested references and supplementary reading
 Suggested reading for topic 1
1.1 Learning objectives
The aims of this course in point of care testing (POCT) are to provide participants with the
knowledge and practical skills necessary to undertake diagnostic tests using point of care devices,
assist with result interpretation and related patient management procedures. This is an on-line
distance learning programme for POCT specialists or those wishing to enhance their knowledge in
this important and continually developing area.
On completion of the course, participants will be able to:

 Understand the relevance and use of different types of biological samples which may be
used for POCT, including blood (finger prick, venous and arterial) urine and saliva.
 Interact with patients with due regard to their personal safety, comfort and wellbeing, with
due consideration of cultural, ethnic and personal privacy issues.
 Articulate the operating principals of the relevant analytical devices.
 Reliably perform test procedures using a variety of analytical devices.
 Validate and confirm the accuracy and reproducibility of test results using quality control
and proficiency testing techniques.
 Understand the relevance of measurement uncertainty and how this can assist with result
interpretation.
 Understand the meaning of fitness for purpose and how the results obtained from a given
device may be related to clinical performance requirements.
 Interpret the significance of test results.
 Communicate test results and interpretation where appropriate.
 Perform all procedures with due regard to regulatory requirements, including:
o Occupational Health and Safety
o Privacy
o Ethical practice
o NPAAC, NATA and ISO requirements as appropriate
 Trouble-shoot and maintain instrumentation and appropriate maintenance records
 Perform required inventory control and ordering
Page 1 of 11 Ian Farrance

1.2 What is point-of-care testing (POCT, PoCT) ?
Point-of-care testing (POCT) is a form of testing in which the analysis is performed where
healthcare is provided close to or near the patient. Various definitions have been provided in the
medical / scientific literature and alternate descriptions include: near patient testing (NPT), bed side
testing, physicians office testing (POL), off site testing, alternate site testing, etc.
International standard ISO 22870, Point-of-care testing (POCT) - Requirements for quality and
competence, provides what may be considered as the current internationally accepted definition:
“testing that is performed near or at the site of a patient with the result leading to possible
change in the care of the patient”.
In addition, the abbreviation used to describe point-of-care testing may be written as “POCT” or
“PoCT”, with individual authors or journals preferring one form over the other. There is no
internationally correct form of abbreviation, but “POCT” has been used throughout this course
largely due to its use within the ISO 22870 standard.
Some alternate descriptions (definitions) of POCT provided by recognised international experts are:
 “POCT is a mode of testing in which the analysis is performed at the site where health-care
is provided close to the patient.” C Price and A St John in Tietz, Fundamentals of Clinical
Chemistry, 6th edition, chapter 12.
 “POCT can (therefore) be defined as the provision of a test when the result will be used to
make a decision and take appropriate action, which will lead to an improved health
outcome.” C Price, A St John and J Hicks in Point –of-Care Testing, 2nd edition, 2004.
 POCT is “diagnostic testing performed at or near the site of patient care, outside a central
laboratory.” JH Nicholls in LA Kaplan and J Pesce, Clinical Chemistry, theory analysis
and correlation, chapter 21, 5th edition, 2010.
 The authorative guidelines “Evidence-based practice for Point-of-Care Testing”, produced
on behalf of the National Academy of Clinical Biochemistry (NACB), provides an
additional condition to the definition by adding “ … typically by clinical personnel whose
primary training is not in the clinical laboratory sciences or by patients for self testing”.
These definitions of POCT provide an accurate but rather general description of the application of
POCT. In practice, POCT may be undertaken in various physical situations including:
 home use, self testing
 pharmacy
 paramedical support, ambulance
 nursing home or aged care centre
 general medical practice, primary care (US terminology: POL, physicians office laboratory)
 rural (remote) hospital or health clinic
 critical care facility in major hospital (ED, ICU, CCU, etc)
 hospital ward or hospital clinic
 sports clinic
 work-place drug screening.
The type of tests and the manner in which a particular point-of-care (POC) device is used in each of
the above situations may differ, the testing frequency will probably be different and the testing
requirements or the fitness for purpose (refer to later course topics for discussion of this important
item) will be different.

In addition, if testing is required for diagnosis, monitoring or screening, different testing devices or
testing regimes may well apply. The required fitness for purpose may also determine the quality
management procedures and the specific requirements for quality control (QC) and external quality
assessment (EQA). Thus, when discussing the application of POCT, the particular situation in
which a given test or device will be used will also require consideration.
POCT is not a single unified entity; different circumstances may require different solutions.
From the above descriptions, it can be seen that POCT is often performed by personnel whose
primary training was not in the clinical laboratory sciences, or is performed by patients themselves
(self-monitoring). Such individuals may require additional information with regard to preanalytical
variation, quality control, instrument evaluation, instrument maintenance and the various sources of
laboratory error. In addition, numerous regulatory guidelines have been established to address these
issues. Successful completion of this course will provide this information.
1.3 Situations in which POCT may be advantageous
Due to the large variety of situations in which POCT devices may be used, management and quality
issues can be complicated by the actual operational situation, the number of different devices
available and the operators involved in testing.
In Australia, POCT testing may be provided in a central laboratory environment by skilled

laboratory staff using only a few instruments, within a hospital ward or specialist unit (such as
emergency department or intensive care) by nursing or medical staff, community based testing
facilities or doctors surgery, work-place (drug) testing, or remote (outback) regions such as
aboriginal community health clinics. In order to ascertain the quality requirements and specific
operational issues with regard to POCT in the general practice environment, the Australian
Government commissioned an extensive evaluation of selected POCT devices (POCT in general
practice trial); the final report was published in January 2009 ( http://www.health.gov.au/poct ).
The key questions asked by the trial were …
 Is it safe to perform POCT in general practice ?
 Is the effectiveness the same or better than for the same tests using pathology laboratory
testing ?
 Is it the same or more cost-effective to perform POCT compared to pathology laboratory
testing ?
 Are patients and other stakeholders more satisfied with POCT than with pathology
laboratory testing ?
 Are there differences between urban, rural and remote geographical regions in any of the
parameters measured ?
 Would the regulatory environment used for the trial meet the needs of all the stakeholders if
POCT were to be made more generally available ?
 What would be the appropriate Medical Benefit Schedule (MBS) fee for the tests selected
in the trial ?
Undoubtedly, maximum patient benefit can be obtained when diagnosis or treatment (or a change in
treatment) can be immediately determined as a consequence of test result availability. Obvious
examples are the immediate availability of test results in an emergency situation (hospital
Emergency Department) or in a remote situation where traditional laboratory testing is not easily
available. However, blood gas and selected analyte testing in a hospital intensive care, coronary
care or neonatal care setting is usually described as POCT, event though regular laboratory services
may be promptly available.

1.4 Evidence based POCT and current best practice for patient treatment
Evidence based medicine (EBM) aims to apply the best available evidence gained from using well
conducted clinical trials and experiment(s) using the scientific method, to medical decision making.
It seeks to assess the quality of evidence which describe the risks and benefits of treatments
(including lack of treatment). The generally accepted definition of evidence-based medicine is “the
conscientious, explicit and judicious use of current best evidence in making decisions about the care
of patients”. Even though there is an implicit acceptance that EBM forms the basis of current
laboratory practice, there are many instances where the evidence base is limited or procedures
flawed. This is of particular relevance to many POCT publications where outcomes are more
related to author bias than scientific evaluation.http://en.wikipedia.org/wiki/Evidence-
based_medicine - cite_note-pmid15338074-1#cite_note-pmid15338074-1
Exercise
Review the concept of evidence based medicine and its relevance to POCT.
The following references may provide useful information …
 Text books as listed (see item 1.9)
 CP Price. Evidence-based laboratory medicine: Supporting decision-making.
Clin Chem 2000; 46:1042.
 RH Christenson. Evidence-based laboratory medicine – a guide for critical
evaluation of in vitro laboratory testing. Ann Clin Biochem 2007; 44: 111.
 RC Hawkins. The evidence based medicine approach to diagnostic testing:
practicalities and limitations. Clin Biochen Revs 2005; 26:7.
 CG Fraser. Optimal quality performance for POCT. Clin Chim Acta 2001;
307:37.
 NACB guidelines. Evidence-based practice for Point-of Care Testing.
http://www.aacc.org/members/nacb/LMPG/OnlineGuide/PublishedGuidelines/poc
t/Pages/poctpdf.aspx
1.5 Visually read indicator strips (“dip-stick”)
Due to the current highly technical approach to POCT, we may be tempted to believe that POCT is
of relatively recent origin. Actually, POCT has existed for many years in the form of the doctor’s
office “laboratory”, urine testing and reagent strip (dip-stick, dip-stix) technology
Laboratory medicine began with the analysis of human urine. In ancient time this was called
uroscopy, today it is often referred to as urinalysis. Modern urine “dip-sticks” are available to test
for a wide range of constituents and clinical conditions, many with multiple test-pads which change
colour in the presence of reactive constituents in urine.
The description outlining the development of glucose oxidase based Clinistix by Alfred and Helen
Free etal at the Miles-Ames research laboratory, and published in Clinical Chemistry in 1957 still
provides interesting reading.

Glucose oxidase based test strips for measuring blood glucose have also been in use for many years.
New technology test strips for blood glucose use a variety of reaction mechanisms and are usually
designed for use with a “glucose meter” to read the reaction endpoint. Many however, may still be
used with a visual (colour reaction) endpoint.
According to one market research report, the impressive growth in point-of-care (POC) diagnostic
testing is principally due to the increased adoption of self-testing by the public. In 2008, patient
self-testing accounted for 71% (US$8.9 billion) of the POC market, and grew at a faster rate than
the professional care sector. The self-test market was dominated by blood glucose testing, and the
increasing incidence of Type II diabetes in both developed and emerging markets. This report
predicted that by 2014, blood glucose testing alone would account for 62.7% of the POC market.
Coagulation monitoring and pregnancy testing both showed positive growth. (Google internet
enquiry for “blood glucose testing market” or “blood glucose self testing market” or similar, for
example, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769893/pdf/dst-03-1219.pdf ).
Exercise
Review the role of POCT in the history and development of medicine. Review the early
forms of urine testing and the development and introduction of “dip-stick” (“dip-sitx”)
technology.
The following references may provide useful information …
 Early editions of many Clinical Chemistry texts
 JA Armstrong. Urinalysis in Western culture: A brief history. Kidney
International 2007; 71: 384.
 WI White. A new look at the role of Urinalysis in the history of diagnostic
medicine. Clin Chem 1991; 37: 119.
 AH Free etal. Simple specific test for urine glucose. Clin Chem 1957; 3: 163.
 HM Free etal. Tripple-test for urine glucose, protein and pH. Clin Chem 1960;
6: 352.
1.6 Technology and overview with respect to types of POCT devices
In general, POCT may be categorised as hand-held or bench-top procedures. Hand-held devices as

the name suggests, are essentially small, self-contained, portable units, capable of being used in
close proximity to the patient. Bench-top devices are generally not used directly at a patient’s
bedside, but in an area attached to a critical care unit, hospital emergency department, out-patient’s
department, or doctor’s office. The large range of devices being used at POC is associated with an
equally large range of technology. This breath of technology incorporates both a large range of
analytes and measurement procedures. Many of the devices use the same analytical principals as
those used in conventional laboratory analysers.
Prior to the early 1980’s, the main POCT applications were …

 simple dip-stick tests for urinalysis
 pregnancy testing
 blood glucose
These early tests relied on a visual colour change to identify the presence of the appropriate
constituent or the extent of reaction which indicated the approximate substance concentration.

At the opposite end of the technology scale, blood gas and acid-base measurement was also
available in specific situations such as critical care units.
This diversity of POCT technology and range of analytes makes it difficult to provide a simple
classification. CP Price and A St John, Point of Care testing, Chapter 12 of Tietz, Fundamentals of
Clinical Chemistry (see references), categorise POCT devices as …
 “Single-use, qualitative or semi-quantitative cartridge or strip tests
 Single-use, quantitative cartridge or strip tests with a reader device
 Multiple-use, quantitative cartridge or bench top devices”.
They also indicate the key design components for POCT technology should include …
 “the operator interface
 a barcode identification system
 sample delivery device
 reaction cell
 sensors
 control and communication system …”.
An overview of the various types of POC device and the variety of measurement procedures
employed is also given in this reference and in chapters three and four of CP Price, A St John and
LJ Kricka. Point-of-care testing: Needs, opportunity and innovation (see references).
The important role which urinalysis played in the overall development of medical practice and the
introduction of dip-stick procedures has been summarised above. Another major advance was the
development and introduction of the blood gas analyser by Poul Astrup and his colleagues in
association with the Radiometer company in Denmark. The development of the “Astrup Trolley”
and its introduction by Radiometer in 1959 was a significant event in the history of blood gas and
acid-base analysis. The “Astrup Trolley” was a portable blood gas analyser which used small
samples of arterialised capillary blood collected into two or three heparinised glass capillary tubes.
The Astrup analyser used a micro pH electrode for the measurement of blood pH, and the pH of two
sub-samples which had been equilibrated at two known pCO2 values. From these three
measurements the acid-base parameters can be calculated. A variety of nomograms were developed
to assist with calculations (for example, the Siggaard Andersen nomogram). In addition to
nomograms, various slide rules were also developed to “mechanise” calculations (in particular, the
Severinghaus slide rule developed in 1966).
Astrup trolley acid-base Siggaard Andersen acid-base

analyser nomogram

Key events in the development of POCT
1950
• Polio epidemic in Denmark
• Astrup device for pH & PCO2
1960 • PO2 electrodes

• Co-Oximetry for
Hb derivatives
1970
• Semi-automated blood • Whole blood biosensors

gas & pH analysis for glucose
1980
• Ion Selective
Electrodes
• Point-of-Care Testing
1990
• Combination analysers
• Optodes for pH, gases & ions
• Coagulation
2000 • Cardiac markers
Key events in development of POCT
Until recently, the emphasis with regard to POCT has been directed more to the technology than the
clinical application. The latter is now becoming equally important, with discussion as to the clinical
utility and fitness for purpose. That is, the performance assessment of POC devices and their ability
to provide the required clinical information.
1.7 Functionality of hand-held devices and bench top devices
There are many applications in which POCT is preferable to testing be conventional laboratory
procedures. In other circumstances however, realised outcomes may not meet predicted benefits.
These aspects and the economics of POCT will be considered at stage in this course. Never the
less, the range of testing procedures and the environments in which POCT may be used has
increased considerably. In addition to the traditional medical environment (hospital, doctor’s
office, etc), many other applications have been found for POCT …
 work-place testing, particularly for drugs of abuse and ethanol
 police road side testing for ethanol (breathalyser) and drugs of abuse
 community pharmacy
 paramedical support and ambulance
 INR testing in pathology collection centres (in addition to traditional laboratory testing)
 remote medical centres (aboriginal health service)
 self testing by patients at home
 visiting health carers (district nursing service)

 sports organisations (glucose and lactate)
 infectious disease testing (for example, HIV, influenza, streptococcus)
 breath analysis to assist with the diagnosis of metabolic disease(s).
1.8 Policies, procedures and guidelines for POCT
The policies, procedures and guidelines which govern POCT vary considerably. Except for the
basic instructions which are provided by device manufacturers, there are often no specific
procedures or regulations which govern POCT. In contrast to this, there are many government
organisations, government health departments, area health authorities or individual hospitals which
have well defined policies and procedures to regulate the use and quality management of all POCT
within their jurisdiction. The innovation, apparent simplicity and functionality of POCT also
provide many challenges for health care funding authorities. In particular, the ability to determine
the value which POCT may bring to the patient care process. In 2011, a review of existing policies,
procedures or guidelines which governed the use of POCT was commissioned by the Department of
Health and Ageing, Australian Commonwealth Government. This review provides a summary of
international and Australian standards, policy and guidelines which apply to POCT. An abridged
version of this report is available on the internet.
Formal policy and regulation generally applies in situations where service payments may be
provided; that is, where a healthcare funding agency (government) or health insurer provides a
service fee for POCT. Excluding home use and personal testing, in countries without specific
overall regulation there is little information regarding the use of POC devices in the “private”
healthcare setting or in situations where patients are prepared to pay directly for testing services.
A copy of Review - Policies, procedures and guidelines for point-of-care testing is included as
part of the POCR references or may be downloaded from
http://www.aacb.asn.au/documents/item/635
1.9 Where to obtain information
1.9.1 Journals
Peer-reviewed articles discussing POCT technology, utilisation, management and operational costs
are now available in many medical and clinical science journals. Specialist POCT texts and many
of the more comprehensive text books on clinical biochemistry contain excellent discussions and
descriptions of POCT technology.
Some useful journals are:
 Clinical Chemistry (Clin Chem), published for the American Association of Clinical
Chemistry (AACC). ). Electronic access is available through the RMIT library electronic
journal search.
 The Clinical Biochemist Reviews (Clin Biochem Rev), published for the Australasian
Association of Clinical Biochemists (AACB). Electronic access is available through the
RMIT library electronic journal search.
 Point of Care (the journal of near-patient testing and technology), a specialist “commercial”
journal http://journals.lww.com/poctjournal/Pages/default.aspx and through the RMIT
library electronic journal search.

 Medical journal of Australia (MJA), published for the Australian Medical Association.
Direct access is available for many articles at “eMJA” http://www.mja.com.au/ but you
may have to search using ‘POCT” and “point of care” descriptors.
 International Medical journals such as …
o The British Medical Journal
o The New England Medical Journal
o Lancet
o Specialist articles in Diabetes and Diabetes Care, etc
Electronic access to journals is available through RMIT library or directly as indicated above.
1.9.2 Internet sites
Like many technical disciplines today, there is a considerable amount of POCT “information”
available on the internet. As may be expected, much of this “information” may not present a
balanced view, particularly with regard to operational costs. Beware, many internet articles
contain significant biases … read with interest and a degree of scepticism!
Useful internet sites …
 http://www.pointofcare.net/ (sponsored by well recognised medical and scientific
professional associations, and reputable health care organisations)
 Australian Government, Department of Health and Ageing. Point of care testing in general
practice trial.
http://www.health.gov.au/internet/main/publishing.nsf/Content/health-pathology-poctt-
index.htm
 Association of Clinical Biochemists in Ireland. (1) Guidelines for safe and effective
management and use of Point of Care Testing. (2) Guidelines for safe and effective
management and use of Point of Care Testing in primary and community care.
http://www.acbi.ie/Articles.asp?id=1
 http://www.aacc.org/members/nacb/LMPG/OnlineGuide/PublishedGuidelines/poct/Pages/p
octpdf.aspx (National Academy of Clinical Biochemistry (associated with the AACC),
guidelines for POCT).
 http://www.guideline.gov/browse/by-topic.aspx (NGC, the National Guidelines
Clearinghouse, a comprehensive database of evidence-based clinical practice guidelines and
related documents from international sources. “Guidelines index” lists documents under
sponsoring organisation. An excellent source of international medical practice guidelines).
 Westgard QC (many useful and interesting essays, biological variation and QC information).
http://www.westgard.com/ . You will be required to register, but this is free.
 Radiometer. Acute care testing (many excellent articles covering blood gas, acid base,
glucose, metabolite, specimen collection issues and POCT).
http://www.acutecaretesting.org/ . You will be required to register, but this is free.
 Lab Tests Online (Australia). http://www.labtestsonline.org.au/
 Australian Point of Care Practitioners Network (APPN). http://www.appn.net.au/home.aspx
 Device evaluation service for the Centre for Evidence Based Purchasing (CEP) of the UK
National Health Service (NHS), has many evaluation reports and general information
covering laboratory medicine and POCT devices. Reports are free and can be downloaded
from the internet. The following two internet addresses give much useful information …

http://nhscep.useconnect.co.uk/CEPProducts/Catalogue.aspx
http://nhscep.useconnect.co.uk/CEPProducts/Default.aspx
1.9.3 Text books and general references
Many of the more comprehensive text books on Clinical Chemistry contain chapters on POCT. In
addition, specialist texts relating to POCT have been produced over the last few years and offer
excellent discussions on all aspects of POCT.
Some useful general text books and references are …

 LA Kaplan and AJ Pesce. Clinical Chemistry, theory, analysis and correlation. 5th edition,
2010.
Textbook available in RMIT library.
 Tietz, edited by CA Burtis, ER Ashwood and DE Burns. Fundamentals of Clinical
Chemistry. 6th edition, 2008.
 CP Price, A St John and LJ Kricka (editors). Point-of-care testing. Needs, opportunity and
innovation. Third edition, AACC press, 2010.
Reviewed in Clin Chem 2010; 56, 1893.
 CP Price, A St John and JM Hicks (editors). Point-of-Care Testing. Second edition, 2004.
Reviewed in Clin Chem 2005; 51, 797. See also Google books at
http://books.google.com.au/books?id=wcXrAgOVXbUC&printsec=frontcover#v=onepage
&q=&f=false
 JH Nichols. Point of Care Testing. Marcel Dekker, 2003.
Reviewed in Clin Chem 2003; 49: 1423. See also Google books at
http://books.google.com.au/books?id=v2sfsG9xBE4C&printsec=frontcover&source=gbs_si
milarbooks_s&cad=1#v=onepage&q=&f=false
 GJ Kost. Principles and Practice of Point-of-Care Testing. Lippincott, Williams & Wilkin,
2002.
Reviewed in Clin Chem 2003; 49 1424. See also Google books at
http://books.google.com.au/books?id=R-
mIg0163m4C&pg=PA25&dq=GJ+kost&cd=1#v=onepage&q=GJ%20kost&f=false
 National Academy of Clinical Biochemistry, Evidence based practice for Point-of-Care
Testing. Available at National Guidelines Clearinghouse (see above) or
http://www.aacc.org/members/nacb/LMPG/OnlineGuide/PublishedGuidelines/poct/Pages/p
octpdf.aspx
Article available in POCT course electronic library (see item 1.9.4).
1.9.4 POCT course electronic library
In addition to other information sources as outlined above, useful articles will be available as part of
this course in an “electronic library”. This library forms part of the course documentation and can
be found at the end of each of the course modules. Articles which are provided in this form give
useful additional information selected for its content and relevance to the particular topic / module.
Unless specifically indicated these articles are not mandatory reading.

1.10 Suggested references and supplementary reading
Throughout this POCT course there are suggestions for additional or supplementary reading. These
are presented as textbook references, journal references or internet references. Each topic has one
or more suggestions which extend the information already provided. Finding and acquiring
supplementary information is a key learning initiative as this is the process by which
knowledge is obtained from the scientific literature. Articles published in peer-reviewed
journals or textbooks provide the best starting point, but information obtained via the internet can be
much more variable. As indicated previously, much of the “information” available via the internet
may not be based on valid scientific evidence or may not present a balanced view. Beware, many
internet articles contain significant biases … read with interest and a degree of scepticism!
Additional reading and supplementary references have been organized as follows:

 Suggested reading: References described as suggested are presented in text boxes at
appropriate intervals within a given topic.
 Exercises: References presented as Exercise assist with integrating the various topic
components and with the review of previous information. For a more detailed
understanding of the subject their content is strongly advised but not considered a
mandatory component of the course.
 Supplementary references: These provide suggestions for obtaining more detailed
knowledge and an understanding of the breadth and depth of the current literature. They
represent a small proportion of the available articles which discuss POCT and provide
source documents for further exploration of the topic. The supplementary references
provide additional (or supplementary) information which may assist with the course
assignments or with the resolution of specific issues now or in the future. They are not
considered mandatory.
1.11 Suggested reading
Included within each topic are additional articles or textbook chapters which provide additional
information, or add important new information which is considered as part of the core course
elements and learning objectives.
Suggested reading for each topic will be clearly identified as shown in the box below (suggested
reading for topic 1).
Suggested reading for topic 1
Chapter 1; An introduction to point-of-care testing and its global scope and application. In:
A Practical Guide to Global Point-of-Care Testing. Edited by Mark Shephard. CSIRO
Publishing, 2016.
I Farrance. Review – Policies, procedures and guidelines for point-of-care testing.

http://www.aacb.asn.au/documents/item/635
Market segments for point-of-care testing devices
In vitro diagnostic devices
It is difficult to obtain consistent up-to-date information with regard to the global markets for
vitro diagnostics and point-of-care testing devices. As a general concept, the term “in vitro
diagnostic (medical) device” is usually defined as “any medical device which is a reagent,
reagent product, calibrator, control material, kit, instrument, apparatus, equipment, or
system, whether used alone or in combination, intended by the manufacturer to be used in
vitro for the examination of specimens, including blood and tissues derived from the
human body”. The IVD definition thus includes all pathology testing equipment including
POCT devices and associated consumables. The term IVD is widely used as a generic
description for medical laboratory (pathology) testing equipment and associated
consumables.
Global market segments for in vitro diagnostic and POCT devices
Point-of-care testing comprises approximately 25% of the global IVD market and is worth
approximately US$11.5 billion annually based on 2007 data. Annual growth in POCT is
around 7% to 8%, nearly twice that of the total IVD market. The main POCT applications
to date have been in those areas where demands are not well met by laboratory testing.
In particularly, diabetes control and pregnancy determination with personal (home use)
testing units are very popular. The Roche company commands approximately 20% of the
POCT market.
Some estimates of the market segmentation for IVD’s can be found at a variety of internet
sites. Most require payment of many dollars to obtain a “market report”, so current
accurate information is not readily available.

The chart below obtained from Medscape provides an estimate of the global IVD market
( http://www.medscape.com/viewarticle/584399_5 ). The diagram represents the global
IVD market of US$45 billion in 2007, including POCT at $11.5 billion (or approximately
25% of the total).
The above chart (which does not include an exact percentage distribution), does give
some appreciation of the relative proportion of the various pathology disciplines in
comparison to POCT.
In the same manner as the distribution of IVD products can be proportioned across the
traditional pathology disciplines, POCT may also be categorised or segmented on a
discipline or test type basis. Again, current data regarding the segmentation of POCT is
difficult to obtain.

One estimate which describes the POCT market segmentation for Western Europe is
available in the journal European Hospital at ( http://www.european-
hospital.com/en/article/7895-POCT_brings_values.html ). A copy of this article is also
available in the topic library. The chart below has been redrawn from the data provides in
the European Hospital article.
European POCT market segments, 2009
7% 7%
7%
9%
38%
Diabetes
Infectious
10%
Fertility
Haematology
Cardiac
22% Coagulation
Blood gas & electrolytes
Segments of the Western European POC market by discipline in 2009
A useful feature provided by this type of information, is an understanding of the relative

growth and application of POCT within the various disciplines. The relative data provides
some insight into:
 the largest proportion is related to diabetes, largely due to personal testing and
home monitoring
 infectious disease screening is becoming important for public health assessment,
HIV, influenza (various types), bacterial infections, etc
 fertility, includes home testing for pregnancy (beta hCG) and ovulation (FSH)
 blood gas and electrolytes, very important but largely applicable to professional
use within critical care centres

Supplementary reading
 JA DuBois. Point-of-care testing recognised as a distinct domain.

POC 2004; 3: 20.
 JF Dooley. Point-of-care diagnostic testing markets.

POC 2009; 8: 154.

Topic 2 – Pre-Analytical considerations
• Topic 2 learning objectives

• Topic 2 overview
• Types of specimen: blood, plasma, serum, urine, saliva
• Analytical differences between whole blood, plasma and serum
• Differences for glucose (example) – blood and plasma, capillary and venous
• Direct and Indirect potentiometry
• Specimen collection procedures
• Pre-analytical artefacts and errors
• Preservation of specimens and timing considerations
• Anticoagulants and preservatives
• Tests on urine
• Tests on saliva
• Suggested reading for topic 2
2.1 Topic 2 learning objectives

On completion of this topic, participants should be able to describe:
• The various types of specimen used for pathology and POC testing and why one
type may be preferred for a particular test.
• The relationship between whole blood, plasma and serum.
• Why different results may be obtained for a given constituent if measured in whole
blood or plasma (or serum).
• Why some results may differ if measured in arterial, venous or capillary blood.
• The difference between direct and indirect potentiometry and why some test
results may differ when comparing results obtained by these two analytical
procedures.
• The usual methods for collecting specimens in POCT.

• How pre-analytical errors may contribute to inaccurate results.
• How anticoagulants should, and should not, be used in POCT.
• The types of test which may be performed on urine, saliva and breath; and how
they are used to assist with clinical decisions.
2.2 Topic 2 overview

The correct collection, processing and storage of samples for laboratory and POC testing
are critical to the production of quality test results. Errors which may occur prior to the
analytical reaction are considered pre-analytical. Recognising and minimising these
potential errors is an important part of the overall testing process.
Many of the pre-analytical errors in pathology (and POC) testing relate to specimen
collection and correct patient identification. In addition, physiological and life-style
associated factors may increase result variation if appropriate patient preparation or
interpretative reference intervals are not considered. These physiological and life-style
pre-analytical variables include:
• Gender and age (newborn, childhood and puberty, adult, elderly)
• Diet
• Diurnal variation (circadian variation), menstrual cycle
• Time of collection, fasting, post prandial, etc
• Drugs, vitamins and herbal preparations which a patient may be taking
• Posture when specimen collected
These aspects of pre-analytical variation will be addressed in more detail within the topics
of Biological Variation and Reference Intervals.
2.3 Types of specimen available for POCT

One of the key objectives for POCT is to use a “small” sample with minimal (or no)
preparation. The use of non-invasive measurement technology, or specimens such as
urine or saliva can sometimes be appropriate. For many tests, blood obtained by finger-
prick is the preferred sample. There are both analytical and physiological reasons why a
specific specimen type may be preferred, but plasma can usually be separated from whole
blood within the POCT device if required for the measurement reaction.
Non-invasive technology is still an area of considerable research. Currently, devices for
the transcutaneous monitoring of bilirubin, the measurement of breath ethanol and blood

oxygen saturation by pulse oximetry are widely available, while non-invasive devices for
glucose monitoring are now being described.
Supplementary references
The following references provide additional useful information:
• CR Lowe. Wearable diagnostic devices. Chapter 11 in: Point-of-care

testing. Needs, opportunity and innovation. Edited by CP Price, A St John
and LJ Kricka. AACC Press 2010.
• Ryan Grueber. Transcutaneous carbon dioxide and oxygen monitoring in

critical care medicine: just for children ? Radiometer, acute care testing,
2002. http://www.acutecaretesting.org/ ).
Urine or saliva may be appropriate for some analytes, but blood is the only relevant
sample for many substances. As discussed in item 2.4 below, whole blood consists of
solid elements (cells) and plasma, which in turn consists of fibrinogen and serum. Serum
is composed of water (approximately 93%) with dissolved substances (ions, glucose,
dissolved gases, metabolites, etc) and “space occupying” compounds which are largely
proteins and lipids. These various fractions are important considerations for pathology
testing and for result interpretation (both for POCT and laboratory procedures).
Compared to urine, saliva is often easier to obtain, with less invasion of privacy (it is easier
to directly observe collection of the specimen), and provides less opportunity for
adulteration. Saliva is an ultrafiltrate of plasma, thus drug and hormone concentrations
reflect the free (or active) fraction of the given substance.
Examples of tests which may be performed on various body fluids are:

• Urine
o excretory products
o glucose (diabetes screen)
o albumin (“protein”)
o blood (haematuria), haemoglobin, myoglobin
o esterase, leucocytes (kidney inflammation)
o nitrite, nitrite forming bacteria (UTI)
o pH
o ketones (diabetes)
o bilirubin and urobilinogen (liver dysfunction)
o pregnancy test
o drugs of abuse
o some infectious disease testing
• Whole blood
o glucose
o haemoglobin
o blood gas (pCO2, pO2) and acid-base (pH, bicarbonate, lactate)
o electrolytes (Na+, K+, Ca++, Cl-)
o cardiac markers (troponin)
o cholesterol, triglycerides
o drugs
o HbA1c
o prothrombin time, INR
o enzymes
o infertility tests
• Saliva
o some drugs
o some hormones
• Breath (expired air)
o Ethanol
o Hydrogen
o Carbon dioxide
o (Breath tests that detect products of bacterial metabolism in the gut or products
of human metabolism by measuring (most commonly) hydrogen or carbon
dioxide in expired air)

2.4 Analytical differences between whole blood, plasma and serum
The best example for the requirement to understand exactly what is being measured in the
given specimen is that of blood glucose. There is a significant difference in glucose
concentrations measured in the equivalent whole blood and plasma samples, and also
between capillary and venous blood. It is essential to take this into consideration when
comparing whole blood POCT glucose with a laboratory value measured on venous
plasma (the usual venipuncture specimen).
The glucose concentration in erythrocyte (red blood cell) and plasma water is the same,
but the amount of water in erythrocytes (approximately 72 %) differs from that in plasma
(approximately 93 %). As a consequence, the total amount of glucose in plasma is higher
than in the corresponding volume of whole blood (erythrocytes plus other solid elements
plus plasma).
The World Health Organisation (WHO) preferred specimen for the diagnosis of diabetes
mellitus is venous plasma, obtained from blood collected into a tube containing a suitable
anticoagulant and an antiglycolytic agent (usually sodium fluoride), with glucose then being
measured by an accredited laboratory procedure. Reagent strips and glucose meters
should not be used for diagnosis, but play an important role in monitoring glycaemic
control in patients with established diabetes or for screening.
The published guidelines for the diagnosis of diabetes mellitus usually show the
approximate differences to be expected between the various types of blood sample. A
useful comparison is: venous plasma glucose 12 to 15 % higher than the corresponding
whole blood.

• C Higgins. Measurement of circulating glucose; The problem of inconsistent

sample and methodology. Radiometer, acute care testing, January 2008.
http://www.acutecaretesting.org/
• K Wiener. Principles and problems of blood glucose measurement.

Radiometer, acute care testing, September 2003.
• WHO. Definition and diagnosis of diabetes mellitus and intermediate

hyperglycaemia.
http://www.who.int/diabetes/publications/Definition%20and%20diagnosis%20o
f%20diabetes_new.pdf
• I Farrance. Serum vs Plasma. Clin Biochem Revs 1989; 10: 166.
Many analytes give results which are dependent on the type of specimen and/or the
particular analytical method involved. These differences are often confusing and may be
difficult to explain in all circumstances. It is not possible to review every possible
combination in a course of this type, but participants must be aware of measurement
procedures which may provide apparently “conflicting” results (particularly between a
POCT device and a traditional laboratory procedure).
POCT technology which provides results on whole blood, usually incorporates some form
of “filter” which separates plasma water (and dissolved substances) from solid elements
such as erythrocytes. For example, many test strips which contain “reaction pads”
incorporate a semi-permeable membrane which prevents erythrocytes from entering the
reaction matrix. Water and the dissolved reactant(s) diffuse through the membrane to
interact with the colour producing chemicals. In this type of reaction, the plasma water is
also an important component as it also acts as solvent for the overall reaction. Colour
producing blood glucose test strips are a good example of this process.
Two other important specimen and measurement related issues are the differences
between direct and indirect potentiometry, and the requirement for specific anticoagulants
in some situations (for example, blood gas, electrolyte and ionised calcium
measurements). These situations will be discussed in more detail below.
2.5 Differences for glucose (example) – blood and plasma, capillary and venous
Comparative values for whole blood (venous and capillary) and venous plasma glucose,
taken during a 75 gram oral glucose tolerance test for the diagnosis of diabetes mellitus
(WHO report 1999).

2.6 Direct and Indirect potentiometry
Analytical procedures which measure a substance directly on undiluted whole blood or

plasma are referred to as direct reaction procedures. Analytical procedures which measure
a substance after a preliminary dilution step are referred to as indirect reaction procedures.
In particular, ion-selective electrode measurement for electrolytes such as sodium (Na+)
may show a significant difference between the two types of procedure.
Ions such as Na+ and K+ are contained (dissolved) in the water phase of blood, even
though there are different concentrations in erythrocyte water and plasma water. This is
due to an active transport mechanism which moves these two ions in opposite directions
across the cell membrane such that sodium is at a higher concentration in plasma than in
erythrocytes (the Na+ / K+ pump http://en.wikipedia.org/wiki/Na%2B/K%2B-ATPase).
Potentiometric (ion-selective electrode) determination of ions such as Na+ measure the

active ion in the water phase which is in direct contact with the electrode surface. When
whole blood is used as sample (without pre dilution), it is the relevant ion dissolved in the
plasma water which reacts with the electrode. In this situation, there is minimal
disturbance to the proportion which is contained within the various blood cells

(erythrocytes etc), so the volume of cells (or volume of “solid elements”) relative to the
volume of whole blood is of no real concern. In fact, the volume of whole blood itself is not
relevant (provided all measurement electrode surfaces are properly covered and the
design requirements of the measuring cell are fulfilled) as the measurement is made
directly on the plasma water. This approach is described as direct potentiometric
measurement.
Traditional laboratory methods such as flame photometry and some alternate types of ion-
selective electrode measurement require pre dilution of the specimen prior to the actual
measurement procedure. This dilution is usually done using plasma, the “dissolved” ions
being largely contained within the water phase which constitutes approximately 93% of the
normal plasma volume. As outlined previously, the remaining 7% of normal plasma are
“space occupying” substances such as proteins and lipids.
When making a pre dilution for any measurement procedure, an accurately measured
volume of plasma is mixed with an accurately measured volume of appropriate diluent.
This means in effect, that only 93% of the given volume of plasma contains the substances to
be measured. In patients with an abnormal increase (or decrease) in plasma protein or lipid
(particularly triglycerides), the plasma water for a given volume of plasma may be significantly
decreased (often to as low as 80%). In traditional measurement systems, this decrease in
apparent concentration in the case of sodium is described as pseudohyponatraemia
(http://medical-dictionary.thefreedictionary.com/pseudohyponatremia). Ion-selective electrode
(potentiometric) systems which use a pre dilution of the blood or plasma specimen are
described as indirect potentiometric measurements.
In general, POCT devices which use ion-selective electrodes use direct potentiometric
procedures. For example, all blood gas and acid-base analysers use whole blood as the
primary sample. Anaerobically collected whole blood being required for pCO2 and pO2
measurement, with pH and other electrodes such as Na+, K+, Ca++ and glucose (if
precent), measuring the relevant constituent directly on the plasma water phase. Such
measurements will not show a lower sodium value as a consequence of increased protein
and/or lipids, that is they will not show a pseudohyponatraemic effect. Comparison with
alternative specimens analysed in a traditional pathology laboratory will need to be
assessed with these considerations in mind.

Exercise
Review the differences between direct and indirect potentiometry. Why are these
differences more apparent for the measurement of sodium than for potassium ?
• U Oesch etal. Ion-selective membrane electrodes for clinical use. Clin Chem
1986; 32: 1448.
• CM Loughrey et al. Sodium measurement: effects of differing sampling and
analytical methods. Ann Clin Biochem 2006; 43: 488.
• T Lang etal. Effect of low protein concentration on serum sodium
measurement: pseudohyponatraemia and pseudonormonatraemia. Ann Clin
Biochem 2002; 39: 66.
2.7 Specimen collection procedures
Blood for analysis can be obtained from arteries, veins or capillaries. Venous blood is
usually the specimen of choice for the traditional pathology laboratory, while arterial blood
is usually used for blood gas analysis. For young children and for many POC tests,
capillary blood collected by skin puncture is more likely to be used.
Suggested reading
Collection details for each of the various types of “blood specimen” are reviewed by:
DS Young etal. Specimen collection and other pre analytical variables.
Chapter 3 in: Tietz, Fundamentals of Clinical Chemistry, edited by CA Burtis, ER
Ashwood and DE Burns, 6th edition, 2008.

There are also many specialised texts on this topic (search under blood collection or
phlebotomy). Other good articles are available on the Radiometer acute care testing
internet site (for example; SB Blonshine. Arterial blood collection: sampling and storage,
part 1 (June 2005) and part 2 (October 2005), http://www.acutecaretesting.org/ ).
2.8 Pre-analytical artefacts and errors
The test results from any individual will vary over time, even without the influence of any
“disease” process. This can be ascribed to three major factors …
• Pre-analytical factors – those related directly to the individual such as posture, time
since last meal (fasting), recent exercise, time of menstrual cycle, etc; and those
influenced by the sample collection itself such as tourniquet application (if a
tourniquet is actually used), quality of skin puncture specimen (blood free-flowing,
sufficient to properly cover the reaction pad)
• Analytical errors - random error (imprecision) and possibly systematic error (bias)
due to changes in calibration
• Biological variation – inherent biological variation around a homeostatic setting

point. That is, within subject (or intra-individual biological variation).
Many of the pre-analytical variables relate to specimen collection. The quality of the
collected specimen is important for both POCT and laboratory testing. With POCT
devices, a critical operating factor is the need to cover the reaction pad completely, or to
adequately fill a reaction cell or electrode chamber. Short sampling from skin puncture
collection is a common hazard.
Whole blood as used for blood gas, electrolytes and many POC tests, may have potential
interferences hidden due to the actual nature of the specimen. Haemolysis for example,
where erythrocyte (and possibly other blood cell) integrity has been compromised,
indicated release to the surrounding fluid (plasma) of substances normally contained within
the erythrocyte cell. With whole blood specimens however, haemolysis is difficult to detect
as the released haemoglobin can not be visualised. The released haemoglobin which
produces “red” coloured plasma when blood is centrifuged in a laboratory environment
(which can be detected visually or by special detection procedures in automated
analysers), will not be obvious when used in whole blood POCT devices. Haemolysed
specimens will have artifactually increase plasma (plasma water) levels of free
haemoglobin (the visual indicator of haemolysis, which may cause spectral interference),
potassium, magnesium and some enzymes (lactate dehydrogenase (LD, LDH), aspartate
amino transferase (AST), adenolate kinase (which may interfere with creatine kinase
measurement), etc). Artefactually increased potassium in haemolysed blood gas
specimens can be of particular concern.
Increasing haemolysis (left to right) in centrifuged blood specimens.

Normal, no haemolysis >>>>>> heavy haemolysis.
Other blood constituents which may cause artifactual results in some circumstances (or in
some particular devices) are: increased bilirubin, increased lipids, pharmacologic agents
(drugs and herbal products), ascorbic acid (vitamin C).
Increased bilirubin (icterus) and increased lipids (triglycerides, turbidity)

in centrifuged blood specimens.

There are many pre-analytical factors which may have an influence on test results,
particularly results from POC testing. The following references provide additional useful
information:
• DR Dufou, Sources of control and pre analytical variation. Chapter 18 of Clinical
Chemistry, theory, analysis and correlation, edited by LA Kaplan and AJ Pesce,
5th edition 2010.
• MH Zweig. Ascorbic acid interference in reagent strip reactions for assay of
urinary glucose and haemoglobin. Clin Chem 1986; 32: 674.
• RH White-Stevens. Interference by ascorbic acid in test systems involving
peroxidise. 1, Reversible indicators and the effect of copper, iron, and mercury.
Clin Chem 1982; 28: 578.
• ML Brigden etal. High incidence of significant urinary ascorbic acid
concentrations in a West Coast population – Implications for routine urinalysis.
Clin Chem 1992; 38: 426.
• SC Kazmierczak etal. Analytical interference, more than just a laboratory
problem. Am J Clin Pathol 2000 113 9.
• JH Livesey etal. Pre-analytical requirements. Clin Biochem Rev 2008; 29:
suppl (1), S11.necke. Useful tips to avoid pre analytical errors in blood gas
testing: Electrolytes. Radiometer, acute care testing, October 2003.
• G Wennecke. Useful tips to avoid pre analytical errors in blood gas testing:
(1) Electrolytes. October 2003. (2) Metaboloitesabolites (glucose and lactate).
January 2004. Radiometer, acute care testing http://www.acutecaretesting.org/
• G Wennecke. Useful tips to avoid pre analytical errors in blood gas testing:
Neonatal total bilirubin. Radiometer, acute care testing, July 2004.
• DS Young. Pre analytical errors in neonatology. Radiometer, acute care
testing, January 2002. http://www.acutecaretesting.org/
• JO Westgard. Errors before the laboratory. .
https://www.westgard.com/essay114.htm

2.9 Preservation of specimens, specimen quality and timing considerations
Specimens for use at POC do not generally require addition of preservatives. In fact,
preservatives such as sodium fluoride which is used to maintain the integrity of blood
glucose specimens for laboratory testing interfere with most POC glucose methods and/or
testing devices.
As most POC testing (with the possible exception of blood gas and acid base as discussed
below) is performed soon after specimen collection, storage and timing issues are not
usually of concern.
Some important specimen relater pre analytical and analytical considerations are:
• Patient identification – The mechanism by which the specimen is associated with

the patient, the request form and the final result is of utmost importance. The use of
“positive specimen identification” with a unique identification label or computerised
barcode and minimisation of manual transcription, significantly decrease the risk of
pre analytical identification errors. The general literature contains clear guidance
for positive patient identification and suggestions to eliminate this type of pre
analytical error. If POC testing is truly performed “next to the patient”, initial
identification may be less of an issue than recording and storage of an adequately
identified result.
Work place or “legal” testing of body fluids for drugs of abuse requires a strict chain
of custody protocol as described by Australian Standard 4760 (AS 4760). The
National Association of Testing Authorities (NATA) accredits organisations wishing
to collect specimens for drug testing. This is discussed in more detail in Module 3,
POC testing for drugs of abuse.

• DR Dufour. Sources of control of pre analytical variation. Chapter 18 of

Clinical Chemistry, theory, analysis and correlation, edited by LA
Kaplan and AJ Pesce, 5th edition 2010.
• M Khoury etal. Error rates in Australian chemical pathology

laboratories. MJA 1996; 165: 128.
• M Plebani. Quality assurance of the pre analytical phase – complying

with ISO 15189. Radiometer, acute care testing, September 2004.
• Temperature - One of the advantages of POCT is the minimal delay between

sample collection and sample presentation to the testing device. With many
POC devices (such as blood glucose), sampling is directly from the skin
puncture site to a reaction strip. However, test strips and test meters must not
be left in environments with “extreme” temperatures. If blood is required to
remain in contact with a reaction strip (reaction or reagent pad) prior to colour
reading by a meter, the reaction strip must be handled in an appropriate manner
as described in the manufacturer’s literature. Placing the reaction strip or POC
device in direct sunlight, on a warm surface, on a window sill or near a heater,
will probably produce incorrect results. The manufacturer’s recommendations
for storage and operating temperatures should be carefully followed.
Considerable debate (particularly in some of the older literature) has arisen as to
the degree which changes in ambient temperature may have on POC test
reactivity. This particularly applies to countries (such as Australia) with large
variations in ambient temperature and where ambient temperature may often
exceed 40oC or higher.

• HM Barbour. Effect of temperature on blood/plasma glucose

measurements with a portable reflectance meter. Clin Chem 1988; 34:
1918.
• P Ridgwell and J Holmes. Effect of temperature on glucose results

obtained with the Reflolux II. Clin Chem 1990; 36: 1705.
• L Ashworth etal. HemoCue: Evaluation of a portable photometric system

for determining glucose in whole blood. Clin Chem 1992; 36: 1479.
• H Nawawi etal. Effect of temperature on analytical and clinical

performance of a blood glucose monitoring system: Omnitest sensor
glucose meter. Ann Clin Biochem 2001; 38: 676.
• Volume and quality of specimen - As described above, the specimen quality of

skin puncture specimens for POCT is an important consideration. With POCT
devices, a critical operating factor is the need to cover the reaction pad completely,
or to adequately fill a reaction cell or electrode chamber. Short sampling from skin
puncture collection is a common hazard.
• Haemolysis – Undetected haemolysis is a particular issue for whole blood POCT.

Refer to discussion above.
• Sampling using laboratory collection tubes - Laboratory blood collection tubes

are important components of sample delivery for traditional laboratory testing. With
the possible exception of blood gas and electrolyte tests in defined circumstances
(see below), laboratory collection tubes are generally not required for POCT testing.
Laboratory blood collection tubes contain a variety of substances which include clot
activators, anticoagulants, special preservatives and specimen stabilisers, in order
to retain specimen and analyte integrity during transport to the laboratory. Re-
sampling from laboratory collection tubes should be discouraged unless the nature
of the tube additive is clearly known not to interfere with the POC test reaction or
electrode activity.
• Glucose and “fluoride” preservative - A preservative is required to stabilise

glucose when blood cells are to remain in contact with plasma for more than 30 to
45 minutes. Blood cells (for example, erythrocytes) continue to metabolise glucose
and produce lactate unless inhibited by the addition of a blocking agent such as
fluoride or iodoacetate. Fluoride inhibits the enzyme enolase while iodoacetate
binds to sulphydral groups and inhibits dehydrogenase and kinase enzymes such
as phosphorfructo-kinase (both key intermediates in the glycolytic pathway of
glucose metabolism). Fluoride also inhibits many of the POC glucose test
reactions, including those associated with blood gas devices and those with
electrochemical sensors. Re-sampling from laboratory collection tubes for POC
glucose testing should only be undertaken with extreme caution. Manufacturer’s
literature will generally contain warnings with regard to specimen usage and lists of
interfering substances.
• Arterial and venous specimens for blood gas and acid base – Whole blood is
the preferred sample for gas analysis. Arterial, capillary or venous samples may all
be used depending on the clinical circumstances and clinical information being
sought. Differences in measured blood gas and acid base values between arterial
and venous are most pronounced for pO2. The main clinical reason for undertaking
the more difficult arterial collection is often only to obtain a pO2 value. All arterial or
venous specimens for blood gas and acid base are best collected anaerobically into
a syringe containing lyophilised heparin (capillary sampling and paediatric
collections are discussed below). If ionised calcium is also to be measured on this
specimen, specially prepared heparin is also required as outlines below. Special
“blood gas” syringes are available commercially. Samples collected under
anaerobic conditions should have minimal exposure to air. The pCO2 content of
room air is virtually zero (0.25 mmHg). The pCO2 of blood exposed to air will
decrease and blood pH (which is a function of pCO2) will rise. The pO2 of room air
is approximately 150 mmHg. This is approximately 60 mmHg higher than arterial
blood and approximately 120 mmHg higher than venous blood. Blood exposed to
room air will thus increase in measured pO2. Blood collected for gas and acid base
analysis should be stored at 4oC if actual measurement is to be delayed. Correctly
stored blood is usually suitable for gas and acid base analysis for up to

approximately 60 minutes. Cooling to 4oC or storing in an ice-water slurry
decreases blood cell metabolism and metabolic changes such as conversion of
glucose to lactate. Freezing of blood should be avoided as this causes cell
membrane rupture and gross haemolysis.
• SB Blonshine, Arterial blood collection: sampling and storage, part 1

(June 2005) and part 2 (October 2005). Radiometer, acute care testing.
• American Association for Respiratory Care (AARC), Clinical practice

guidelines. Sampling for arterial blood gas analysis.
http://www.rcjournal.com/cpgs/sabgacpg.html
• Capillary and paediatric specimens for blood gas and acid base – Arterialised
capillary blood is often used in the paediatric situation or when blood loss needs to
be minimised. Freely flowing cutaneous blood originates from the arterioles and in
most situations corresponds closely to arterial blood. The first drop of free flowing
blood is discarded with the remainder being collected into a heparinised capillary
tube by capillary action.

• MG Scott et al. Electrolytes and blood gases Chapter 24 of Tietz,

Fundamentals of Clinical Chemistry, edited by CA Burtis, ER Ashwood
and DE Burns, 6th edition, 2008
• C Higgins. Capillary blood gases – to arterialise or not. Radiometer,

acute care testing, July 2008. http://www.acutecaretesting.org/
• American Association for Respiratory Care (AARC), Clinical practice

guidelines. Capillary blood gas sampling for neonatal and paediatric
patients. http://www.rcjournal.com/cpgs/cbgscpg.html
• BD Vacutainer. LabNotes 2009; number 1. Capillary blood collection:

Best practices.
http://www.bd.com/vacutainer/labnotes/Volume20Number1/
• V Shah and A Taddio. Neonatal capillary blood sampling. Radiometer,

acute care testing, July 2005. http://www.acutecaretesting.org/
• Foetal capillary sampling and scalp pH – During labour, the foetus is at extreme
risk of compromised oxygen supply from the placental circulation. Compromised
oxygen supply can occur rapidly and results in severe organ damage or foetal death
if not treated promptly. Foetal capillary blood collected from the scalp during birth
allows foetal acid base monitoring.

• Medline plus.
http://www.nlm.nih.gov/medlineplus/ency/article/003408.htm
• CN Nickelson. Fetal capillary blood pH (fetal blood sampling).

Radiometer, acute care testing, June 2002.
• Sample mixing – Homogeneity of a blood sample is another important requirement

for obtaining quality results. This is particularly so for haematological parameters
and blood gases. Immediate mixing of the sample following collection insures that
the anticoagulant in the specimen container (or capillary tube) has fully dissolves
and mixed with the blood. This process also assists in producing a specimen free
of blood clots which may interfere with the testing process. (Refer to: S Narayanan.
Pre-analytical issues relating to blood sample mixing. Radiometer, acute care
testing, October 2005. http://www.acutecaretesting.org/ ).
2.10 Anticoagulants and preservatives
The use of blood which contains anticoagulants and/or preservatives requires careful
consideration. Anticoagulated blood collected for laboratory assay may not always be
appropriate for POCT, while specimens for blood gas and acid-base testing usually do
require an anticoagulant.
These issues are discussed under the relevant analytical sections. Some specific
examples are:
• Glucose, using POC meter systems, usually show interference by fluoride which is
the common preservative in laboratory collection tubes. Manufacturer’s specific
instructions should be consulted for each meter type. As a general rule however,
POC glucose measurement by re-sampling from a “fluoride” blood collection tube
should be avoided.
• Blood gas and acid-base, heparinised blood collected into a heparinised syringe is
the preferred collection procedure.
• Ionised calcium (measured on a “blood gas” analyser), blood heparinised with

special “calcium titrated heparin” is required to prevent ionised calcium binding to
heparin.
2.11 Tests on urine
There are many POC tests which are (or can be) performed on urine. In addition to the
traditional urine dip-stick tests, home testing for hCG (pregnancy test) and lutropin (as a
markers for ovulation), urine is the preferred sample for drugs of abuse testing.
• Refer to Module 1, topic 3, for additional information on urine dip-stick testing.
• Refer to Module 3, Diabetes and glucose metabolism, for a discussion on urine

creatinine and albumin in the on-going assessment of diabetes mellitus.
• Refer to Module 3, POCT testing for drugs of abuse, for a discussion of urine drug
testing.
The type of urine specimen to be collected is dictated by the test(s) to be performed. For
most POC tests, a random (untimed) specimen is suitable, even though a first morning
(fasting) specimen is usually the most concentrated. The first morning specimen is
suitable for dip-stick and pregnancy testing.
Urine testing for drugs of abuse usually requires supervision of the actual collection and
special handling considerations. “Chain of custody” procedures and documentation are
required for many situations involving drug of abuse testing. These are discussed further
in Module 3 POCT testing for drugs of abuse.
The following reference provides additional useful information:
• LA Kaplan and AJ Pesce. Clinical Chemistry, theory, analysis and correlation.

Sources and control of pre-analytical error, Chapter 18, 5th edition 2010.

2.12 Tests on saliva
Saliva (often referred to as “oral fluid”) is gaining importance as an alternate matrix for
drug monitoring and hormone analysis. The relative ease of obtaining a saliva sample and
the difficulty of adulterating this type of specimen, makes this approach attractive for
monitoring drugs of abuse in the workplace, criminal justice and police drug programmes.
As saliva contains “free” (non protein bound) constituents, it is also considered a good
matrix for the measurement of the “free” physiologically active fraction of substances
transported in blood bound to a protein carrier. Many attempts have been made to
measure the free hormone fraction of substances such as cortisol, testosterone and
oestradiol, as an index of physiological activity without the influence of protein binding.
For drugs of abuse testing where legal penalties or employment opportunities may depend
on the testing outcome, assay characterisation, assay performance and assay limitations
are important considerations.
An overview of police drug testing in Victoria and Queensland is given at

http://www.police.qld.gov.au/programs/roadsafety/drugdriving.htm and
http://www.roadsafety.vic.gov.au/key-issues/drink-driving.html
The following reference provides additional useful information:
• S George and RA Braithwaite. Use of on-site testing for drugs of abuse.
Clin Chem 2002; 48: 1639.
• OH Drummer. Drug testing in oral fluid. Clin Biochem Revs 2006; 27: 147.
• WM Bosker and MA Huestis. Oral fluid testing for drugs of abuse.
Clin Chem 2009; 55: 1910.
• M Groschl. Current status of salivary hormone analysis.
Clin Chem 2008; 54: 1759.

2.13 Suggested reading for topic 2
Suggested reading
Collection details for each of the various types of “blood specimen” are reviewed by:
• DS Young etal. Specimen collection and other pre analytical variables.
Chapter 3 in: Tietz, Fundamentals of Clinical Chemistry, edited by CA Burtis, ER
Ashwood and DE Burns, 6th edition, 2008.

Topic 3 – Analytical considerations for hand-held devices
• Learning objectives for topic 3

• Overview of topic 3
• The on-going role of “dip-stick” technology
• Glucose: instrument technology and general considerations
• Cholesterol, triglycerides and HDLC: instrument technology and general
considerations
• HbA1c: instrument technology and general considerations
• Haemoglobin and prothrombin (INR): instrument technology and general
considerations
• Urine creatinine and albumin: instrument technology and general considerations
• Troponin and markers of cardiac function
• Pregnancy testing
• Tests for other substances using POCT technology
3.1 Learning objectives for topic 3

• The types of hand-held devices available for use at POC.
• The range of test procedures and their associated clinical application(s).
• The technology and analytical aspects of the major hand-held POC devices.
• Specific issues which may limit the clinical utility (and fitness for purpose) of POCT
devices. Particularly fitness for purpose of for glucose, (cholesterol), HbA1c, INR
and troponin testing devices.

3.2 Overview of topic 3
There are a variety of hand-held POCT devices currently available from a number of
manufacturers and collectively these account for approximately one-third of the global in
vitro diagnostics market.
The majority of hand-held devices comprise a disposable reagent filled test strip, tablet,
cartridge or cassette, to which the patient sample is added. In some publications, these
single use disposable test units are described as unit use systems. The scope of hand-
held testing systems covers a very wide range of applications which include: critical care
and emergency medicine (for example; blood gas, electrolytes and troponin), risk
assessment (for example; lipids), chronic disease management (for example lipids and
glucose), anticoagulation therapy (for example; INR), infectious disease screening (for
example; hepatitis, HIV, influenza), sports medicine (for example; drug testing, lactate) and
personal health (for example; glucose, pregnancy testing). In addition to this “traditional”
type of medical use, devices have also been developed to detect chemical and biological
agents released into the environment.
Microtechnology, nanotechnology and computer electronics have all been exploited to
develop hand-held devices. The assay designs that underlie the different types of device
are many and varied: they include; agglutination, colorimetric, enzymatic,
immunochemical, electrochemical, and piezoelectrical based measurement systems.
Even though there is a wide range of test systems with wide range of operation principles,
it is critical that device operators become familiar with all test systems which they are
required to use or manage.
Suggested reading
• M Shephard. Selection and evaluation of point-of-care testing devices. In: A

Practical Guide to Global Point-of-Care Testing. Edited by Mark Shephard.
CSIRO Publishing, 2016.
OR
• LJ Kricka and GHG Thorpe. Technology of hand-held devices for POCT.
Chapter 3, in: Point-of-care testing. Needs, opportunity and innovation. Edited
by Price CP, St John A and Kricka LJ. AACC Press, 2010.

3.3 The on-going role of “dip-stick” technology
The classical “dip-stick” technology still plays a significant role in POC testing. In many
texts, urine “dip-stick” testing is often included under urinalysis. Information which can be
obtained by urinalysis includes the appearance of the (urine) specimen and qualitative
(presence or absence) or semi-quantitative measures of: pH, protein (albumin), glucose,
ketones, specific gravity, bilirubin, urobilinogen, haemoglobin and erythrocytes, white
blood cells and nitrite forming bacteria. Visually read test strips for blood glucose and
cartridge style pregnancy tests should also be considered in this category.
Urine “dipstick” testing is cheap, relatively easy and gives immediate results. It is often
used as a screening tool thus making it invaluable for clinical practice. The sample should
be mid-stream and collected into a sterile container. Contrary to some recommendations,
the “stick” should not be dipped into the container of urine, as chemicals from the strip
may leach out and contaminate the urine for further (or repeat) testing. A sub-sample
should be used for dipping test sticks or a transfer pipette should be used to distribute
urine over the test pads. The test stick should then be placed on a clean dry surface, with
readings taken at the prescribed interval(s). However with any investigation of this type,
urine “dipstick” testing is not without its limitations.
The chemistry associated with each test or reaction pad is often not fully described in the
literature provided by the manufactures(s). This is particularly so for buffer components
and “non-reactive” constituents. These items however, are possible contaminants if sticks
are dipped directly into urine as reported on several occasions. Contamination of urine by
caffeine (an undescribed component of the bilirubin reaction pad) has been reported with
possible false positive immunological tests for urine opiates.
• W Clarke and DE Palmer-Troy. Outcomes-based evaluation of dipstick urinalysis.

Chapter 11 in: JH Nichols, Point of Care Testing, 2003 (see general references).
• CG Fraser. Urine analysis: current performance and strategies for improvement.
BMJ 1985; 291: 321.
• MD Shephard et al. Urinalysis in an Australian teaching hospital. MJA 1982; 1:
300.
• JA Simerville et al. Urinalysis: A comprehensive review. Am Family Physician
2005; 31: 1153.

• JM Rattenbury. Sample contamination by test strips. Clin Chem 1983; 33: 414.
• J Kropf et al. Contamination of urine by test-strip measurements ? Clin Chem
1989; 35: 2144.
• JT Van Acker et al. Falsley increased urinary caffeine attributable to contamination
by urine test strips. Clin Chem 1999; 45: 1315.
• JP Chanoine et al. Iodine contamination of urine samples by test strips. Clin
Chem 1983; 33: 1935.
• MH Zweig. Ascorbic acid interference in reagent strip reactions for assay of urinary
glucose and haemoglobin. Clin Chem 1986; 32: 634.
• RH White-Stevens. Interference by ascorbic acid in test systems involving
peroxidise. 1, Reversible indicators and the effect of copper, iron, and mercury.
Clin Chem 1982; 28: 538.
• ML Brigden et al. High incidence of significant urinary ascorbic acid concentrations
in a West Coast population – Implications for routine urinalysis. Clin Chem 1992;
38: 426.
3.4 Glucose: instrument technology and general considerations

Patients with diabetes, especially those who require insulin, also require careful monitoring
to maintain control of blood glucose. Portable glucose meters are used in acute and
chronic care facilities, in a doctor’s office and by patients for self monitoring. A “large”
number of test strips and associated meters are available to permit rapid blood glucose
measurement with reasonable analytical performance. A good summary of POC glucose
testing is given by Wikipedia ( http://en.wikipedia.org/wiki/Glucose_meter ).
There are generally two types of measurement system, (1) enzymatic with colorimetric
detection (either visual colour comparison and/or meter reading) or, (2) enzymatic with
electrochemical detection and meter reading. Each brand or model will probably have
different reaction (glucose detection) characteristics, so manufacturer’s literature should be
consulted for specific reaction sequences.

Exercise
Review the analytical principles and performance requirements for quantitative

glucose test strips and POCT glucose meter technology.
• J Hones and P Muller (from Roche Diagnostics). The technology behind

glucose meters: Test strips. Diabetes Technology and Therapeutics 2008;
10: supplement 1, page S10.
• E-H Yoo and S-Y Lee. Glucose biosensors: An overview of use in clinical
practice. Sensors 2010; 10: 4558.
• M Shephard, J Shaw and P Zimmet. Point-of-care testing for diabetes –
glucose. In: A Practical Guide to Global Point-of-Care Testing. Edited by
Mark Shephard. CSIRO Publishing, 2016.
In common with all POCT systems however, technical performance and/or the use of
glucose measuring devices may be influenced by:
• Inter-device variability
• operator training
• technical comparisons or performance assessments which compare whole blood
to (laboratory determined) plasma glucose values (refer to the various items
outlined in topic 2)
• pre-analytical considerations as discussed in topic 2
• blood re-sampled from pathology collection tubes (fluoride may inhibit the glucose
reaction sequence)
• haemolysis (undetected haemolysis in whole blood), method and device
dependent for glucose
• high and low glucose readings; glucose meters for general use give less
satisfactory performance in the “low” glucose range (hypoglycaemic range or below
(approximately) 2.0 mmol/L), and most do not read above (approximately) 30.0

mmol/L. Test strips and meters designed for neonatal use are available for better
differentiation in the hypoglycaemic range however.
• variation of the size and direction of bias in different parts of the operating range
• impairment of vision or colour-vision
• small sample volume which does not cover the reaction pad or fill the reaction
chamber correctly
• altitude and ambient oxygen tension (?)
• haematocrit (variable, device dependent).
Example of POC glucose testing technology.

Glucose meter – electrochemical biosensor detection
• DB Sacks. Carbohydrates, chapter 22 of Tietz, Fundamentals of Clinical

Chemistry, edited by CA Burtis, ER Ashwood and DE Burns, 6th edition, 2008.
• LJ Kricka and GHG Thorpe. Technology of handheld devices for point of care
testing. Chapter 3, in: Point-of-care testing. Needs, opportunity and innovation.
Edited by Price CP, St John A and Kricka LJ. AACC Press 2010.
• WA Davis and TME Davis. Point of care testing in diabetes mellitus. Chapter 36,
in: Point-of-care testing. Needs, opportunity and innovation. Edited by Price CP,
St John A and Kricka LJ. AACC Press 2010.
• AK Aarsand et al. Diagnosis and management of diabetes mellitus. Chapter 6, in:
National Academy of Clinical Biochemistry, Evidence based practice for Point-of-
Care Testing (see references) or https://www.aacc.org/science-and-
research/practice-guidelines/point-of-care-testing
• S Coster et al. Monitoring blood glucose control in diabetes mellitus: A systematic
review. Health Technology Review 2000; 4 (12):1 to 93.
• ES Kilpatric et al. The effect of haemolysis on blood glucose measurement.
Diabet Med 1995; 12: 341.
• RF Louie et al. Point of care glucose testing, effect of critical care variables,
influence of reference instruments and a modular glucose meter design. Arch
Pathol Lab Med 2000; 124: 253.
• GJ Kost et al. Multicentre study of oxygen-intensive handheld glucose point of
care testing in critical care / hospital / ambulatory patients in the United States and
Canada. Crit Care Med 1998; 26: 581.
• R Gama et al. Glucose meter hypoglycaemia; often a non-disease (case report).
Ann Clin Biochem 2000; 33: 331.
• AC Gallagher et al. Evaluation of the Boehringer Advantage and the Bayer
Glucometer Encore QA+ blood glucose meters in a neonatal unit. Ann Clin
Biochem 2000; 33: 393.
• LMC Welschen et al. Self-monitoring of blood glucose in patients with type 2
diabetes who are not using insulin. Diabetes Care 2005; 28: 1510.
• G Lippi et al. Evaluation of four portable self-monitoring blood glucose meters.
• K Dungan et al. Glucose measurement; Confounding issues in setting targets for
inpatient management (review). Diabetes Care 2003; 30: 403.
• JH Eastham et al. Prevelence of interfering substances with point of care glucose
testing in a community hospital. Am J Health System Pharmacy 2009; 66:163.
• C Florkowski et al. Comparison of blood glucose meters in a New Zealand
diabetes centre. Ann Clin Biochem 2009; 46: 302.
• E-H Yoo and S-Y Lee. Glucose biosensors: An overview of use in clinical practice.
Sensors 2010; 10: 4558.
• J Hones and P Muller (from Roche Diagnostics). The technology behind glucose
meters: Test strips. Diabetes Technology and Therapeutics 2008; 10: supplement
1, page S10.
• Centre for Evidence based Purchasing (CEP), NHS (National Health Service, UK).
(1) Blood glucose monitoring systems, market review. October 2009, CEP 09028.
(2) Blood glucose systems, buyers guide. May 2008 CEP 08008. (3) MediSense
Optium blood glucose sensor December 2002, MDA evaluation report 0203.
(Search the CEP catalogue for these items or check the topic library for “pdf”
copies)

http://nhscep.useconnect.co.uk/CEPProducts/Catalogue.aspx?ReportType=Market
+review
• BS Karon etal. Glucose meter performance criteria for tight glycemic control
estimated by simulation modelling. Clin Chem 2010; 56: 1091.
• BN Kelly etal. Interference in a glucose dehydrogenase-based glucose meter.
Clin Chem 2010; 56:1038.
• DB Sacks etal. Diabetes; Advances and controversies (editorial).
Clin Chem 2011; 53: 143.
• S Vanavanan etal. Performance criteria of a new interference-resistant glucose
meter. Clin Biochem 2010; 43: 186.
3.5 Cholesterol, triglyceride and HDLC: instrument technology and general

considerations
Cholesterol, triglycerides and high density lipoprotein cholesterol (HDLC or HDL) are all
available on POC devices. The clinical utility and clinical application of these tests are well
described in standard Clinical Chemistry textbooks such as those suggested in the general
references.
The main issue with POC devices for lipid testing is their ability to provide an appropriate
level of analytical performance. Diagnosis and monitoring of lipid disorders require test
methods that provide both good accuracy and precision. The usual criteria used to
assess analytical performance for cholesterol and HDLC measurements are those
developed by the Centres for Disease Control (CDC) and the National Heart, Lung and
Blood Pressure Institute (NHLBI). The performance expected for cholesterol from
laboratories who do not participate in the cholesterol reference method laboratory network
is a bias of 3% or less and an overall imprecision of 3% or less. This compares to the
Australian Quality Assurance Programme (QAP) total allowable error no greater than +/-
0.50 for cholesterol values less than 10.0 mmol/L. Equivalent CDC values for HDLC are a
bias of 5% or less and an overall imprecision of 4% or less, with QAP total allowable error
no greater then +/- 0.2 for HDLC values less than 2.0 mmol/L. These performance
specifications should be compared to the desirable performance specifications based on
intra-individual biological variation as discussed in Topic 6. The concept of analytical
performance which is based on biological variation is discussed in detail by CG Fraser in
his book “Biological Variation” and by Westgard et al (see references below). The intra-

individual biological variation for plasma cholesterol in a healthy individual is approximately
5.4%.
POC lipid testing systems usually measure cholesterol, triglycerides and HDLC
simultaneously, with low density cholesterol (LDLC) being derived by calculation.
Enzymatic reaction(s) form the basis of the measurement process as described in
manufacturers’ literature and in the article by MDS Shephard et al, as outlined below. The
reaction sequences are essentially the same as those used in standard laboratory testing
and similar interfering factors should thus be expected (and in particular ascorbic acid, a
common interferent in hydrogen peroxide coupled reactions).
Exercise
Review the factors which may influence the performance of POCT lipid testing.
• Cholesterol reference method laboratory network for the notional reference

system for cholesterol, certification protocols for clinical laboratories, 2004.
http://www.cdc.gov/labstandards/pdf/crmln/CertProtocolClinLabsMay04.pdf
• CDC. National reference system for cholesterol, HDL cholesterol certification

for manufacturers, 2002.
http://www.cdc.gov/labstandards/pdf/crmln/MFRHDLNov2002final.pdf
• MDS Shephard et al. Comparative performance of two point-of-care

analysers for lipid testing. Clin Lab 2003; 53: 561.
• Westgard et al. Guest essays and biological variation database compiled by

C Ricos et al. Westgard QC at http://www.westgard.com/biodatabase1.htm
• H Halls, A Shephard and K Sikaris. Point-of-care testing for cardiovascular

disease. In: A Practical Guide to Global Point-of-Care Testing. Edited by

• RH Ng et al. Direct measurement of high-density lipoprotein cholesterol by the

Reflotron assay with no manual precipitation step. Clin Chem 1991; 33: 435.
• DA Wiebe and JO Westgard. Cholesterol – A model system to relate medical
needs with analytical performance. Clin Chem 1993; 39: 1504.
• WT Law et al. Whole blood test for total cholesterol by a self-metering, self-timing
disposable device with built-in quality control. Clin Chem 1993; 43: 384.
• C Luley et al. Point-of-care testing for triglycerides: Evaluation of the Accutrend
triglycerides system. Clin Chem 2000; 46: 283.
• M du Plessis et al. Analytical quality of near patient blood cholesterol and glucose
detarminations. Clin Chem 2000; 46: 1085.
(1) Point-of-care testing for cholesterol measurement, buyer’s guide. CEP 09020,
August 2009.
(2) Polymer Technology Systems CardioChek PA lipid system (evaluation report
number 05051).
(Search the CEP catalogue for these items or check the topic library for “pdf”
copies)
+review
3.6 HbA1c: instrument technology and general considerations

The incidence of type 1 and type 2 diabetes is gradually increasing. It has been shown in
the Diabetes Control and Complications Trial (DCCT) and the United Kingdom Prospective
Diabetes Study (UKDPS) that good control of plasma (blood) glucose in both types of
diabetes reduces the risk of both microvascular and macrovascular disease. The degree
of haemoglobin A1c glycation (HbA1c) is a good indicator of glucose control over the
previous (approximately) 120 days, weighted towards glucose control over the last 30
days. HbA1c measurement thus provides a means of assessing glycaemic control and
response to treatment. It is now widely accepted that the risk of developing complications

are minimised by maintaining HbA1c concentrations as low as possible but without
increasing hypoglycaemic episodes.
A diabetic person with good glucose control has a HbA1c level that is close to or within the
reference range of 4.0% to 6.0%. The International Diabetes Federation and the American
College of Endocrinology and the recommend HbA1c values below 6.5%, while the
American Diabetes Association and the consensus statement from the Australian Diabetes
Society, the Royal College of Pathologists of Australasia and the Australasian Association
of Clinical Biochemists recommends that HbA1c should be below 7.0% for patients under
good glycaemic control.
Exercise
Review the factors which may influence the clinical utility of HbA1c assays.
• Diabetes Control and Complications Trial (DCCT), link to DCCT information.

http://diabetes.niddk.nih.gov/dm/pubs/control/index.htm
• M Shephard, J Shaw and P Zimmet. Point-of-care testing for diabetes –
haemoglobin A1c. In: A Practical Guide to Global Point-of-Care Testing.
Edited by Mark Shephard. CSIRO Publishing, 2016.
• Diabetes in the United Kingdom, 2004.
http://www.diabetes.org.uk/Documents/Reports/in_the_UK_2004.doc
• National evidence based guideline for blood glucose control in type 2
diabetes, NHMRC 2009.
http://www.nhmrc.gov.au/_files_nhmrc/file/publications/synopses/di19-
diabetes-blood-glucose-control.pdf
• NGSP (National Glycated haemoglobin Standardisation Program),
background information on DCCT and reasons for assay standardization (and
other useful information). http://www.ngsp.org/index.asp
• Colman PG et al. Glycohaemoglobin: a crucial measurement in modern
diabetes management. Progress towards standardisation and improved
precision of measurement. MJA 1997;163:96.
Copy available in topic library.

The two principal issues with regard to HbA1c measurement are …
• analytical precision of the particular assay method
• assay standardisation and traceability to an international reference preparation and
reference method. The two competing standards are based on the “original” DCCT
procedure and the “new” IFCC reference preparation.
The measurement of HbA1c is thus an important marker for long term glucose control
provided assay procedures and conditions are appropriate and strictly monitored.
Currently in Australia, assay procedures which are calibrated with standards traceable to
the DCCT are preferred as they provide direct traceability to DCCT results and monitoring
recommendations. Which ever standard (IFCC or DCCT) is finally adopted, only assays
which demonstrate consistently high precision (low analytical CV) are recommended. As
indicated above, results based on the DCCT and endorsed by the American Diabetes
Association and supported by a consensus statement from the Australian Diabetes
Society, the Royal College of Pathologists of Australasia and the Australasian Association
of Clinical Biochemists, recommends a HbA1c level of less than 7.0% as a primary goal for
diabetes therapy. Re-evaluation of treatment in patients with a HbA1c level consistently
greater than 8.0% is also recommended. To provide a clear analytical distinction between
the HbA1c decision levels of 7.0% and 8.0%, an analytical uncertainty (method
uncertainty) expressed statistically as a coefficient of variation (CV%) of less than 3% is
required. Methods which produce analytical CV’s of 4.0% or greater are not considered
appropriate, as this degree of analytical imprecision cannot distinguish changes in
measured HbA1c of over 1% (that is, can not distinguish a HbA1c level of 7.0% from a
level of 8.0%).
An alternate approach to the above considerations is based on intra-individual biological
variation and serial assay results in the same individual. As discussed in topic 6
“measurement uncertainty and reference intervals”, the desired goal is for analytical
imprecision to be less than one half the within-person biological variation of 3.4%
(Westgard, biological variation database http://www.westgard.com/biodatabase1.htm).
Under this guideline therefore, the desired analytical variation for HbA1c would be a CV of
1.3% or less. Although this 1.3% criterion can in practice be met by several of the better
laboratory-based procedures, it is barely achievable by even the best performing POCT
device(s).
To show how the analytical CV should be interpreted in relation to the actual HbA1c
measurement, the current minimum analytical CV of 3.0% can be used as an example.

When an analytical CV of 3.0% is applied to an actual HbA1c measurement, the measured
HbA1c can be regarded as falling within a range of values defined by +/- twice the
analytical CV (or +/- 6% of the measured value in this example). Additional examples for a
nominal analytical CV of 3% (uncertainty of measurement 6%) and various measured
HbA1c levels are reproduced in the table below (from: GH White and I Farrance,
Uncertainty of measurement in quantitative medical testing. Clin Biochem Rev 2004; 25,
supplement (ii): S1. Refer to article in topic library).
Uncertainty of measurement range for

Measured HbA1c %
an analytical CV of 3%
5.0 4.3 – 5.3
6.0 5.6 – 6.4
6.5 6.1 – 6.9
3.0 6.6 – 3.4
3.5 3.1 – 8.0
8.0 3.5 – 8.5
8.5 8.0 – 9.0
9.0 8.5 – 9.5
10.0 9.4 –10.6
• GH White and I Farrance. Uncertainty of Measurement in quantitative medical

testing – A laboratory implementation guide. Clin Biochem Rev 2004; 25,
supplement (ii): S1. Article available in topic
library.http://www.aacb.asn.au/files/cbr_articles/uncertainty_of_measurement.pdf
• WG John et al. HbA1c standardisation: History, science and politics. Clin Biochem
Rev 2003; 28: 163. Article available in topic library.
• CG Fraser and PH Petersen. Analytical performance characteristics should be
judged against objective quality specifications. Clin Chem 1999; 45: 321.
• AK Aarsand et al. Diagnosis and management of diabetes mellitus. Chapter 6, in
• Consensus committee. Consensus statement on the world wide standardisation of
HbA1c measurement. Diabetes Care 2003; 30: 2399. OR Diabetologia 2003; 50:
2042.
• IFCC scientific division. Approved IFCC reference method for the measurement of
HbA1c in human blood. Clin Chem Lab Med 2002; 40: 38.
• DE Burns and JC Boyd. Few point-of-care hemoglobin A1c assays meet clinical
needs. Clin Chem 2010; 56: 4.
• E Lenters-Westra and RJ Singerland. Six of eight HbA1c POCT instruments do not
meet the generally accepted analytical performance criteria. Clin Chem 2010; 56:
44.
• JO Westgard. The January 2010 issue of Clinical Chemistry (refer to the two
references above) has an editorial plus an evaluation study of eight POCT HbA1c
instruments. The study found that most of these POC devices do not provide
sufficient analytical performance. Westgard on his internet site provides an
extensive discussion of this topic under “HbA1c in 2010”, parts I to X.
http://westgard.com/HbA1c-methods.htm
Current methods for HbA1c fall into two broad categories based on (1) charge differences
between modified and unmodified hemoglobin (ion-exchange chromatography,
electrophoresis), and (2) structural differences that recognize “glucose-modified”
haemoglobin from “nonglucose-modified” hemoglobin (boronate affinity chromatography,
immunoassay and enzymatic methods). The National Glycated haemoglobin
Standardisation Program (NGSP) provides a good overview of current methods which
have been “certified” for HbA1c measurement ( http://www.ngsp.org/index.asp ).
Boronate affinity chromatography employs a binding ligand (boronic acid) immobilized to
an inert, insoluble matrix that selectively interacts with cis-diol groups present in sugars
residues (such as glucose), to form a strong but reversible covalent bond. Glycated
hemoglobin remains bound, while nonglycated hemoglobin is eluted with wash buffer. A
soluble diol-containing agent such as sorbital is then used to displace the glycated
hemoglobin. Both species are detected photometrically at 414 nm.
Immunoassays use antibodies that recognize the glycated amino terminus of the
haemoglobin β chains. Commercial methods use antibody-mediated inhibition of latex
agglutination or immunoturbidimetry to obtain a HbA1c value. HbA1c is expressed as a

percent of total haemoglobin after spectrophotometric measurement and the ratio
HbA1c/total Hb is determined.
The National Glycated haemoglobin Standardisation Program (NGSP) website, describes
two “certified” enzymatic procedure ( http://www.ngsp.org/index.asp ).
A review of the principles of enzymatic HbA1c methods is provided by “Diazyme” in their

product information at
http://search.cosmobio.co.jp/cosmo_search_p/search_gate2/docs/DZL_/DZ168AK.200810
14.pdf
Search index at:
+review
Many interesting POCT evaluation reports, including:
o Bayer DCA+, haemoglobin A1c measurement (evaluation report MDA 03016)
o Haemoglobin A1c measurement devices (evaluation report MDA 02098)
o Buyers guide, POCT devices for the measurement of HbA1c and low
concentration albumin in urine (report CEP 08053)
• MDS Shephard and JP Gill. The analytical quality of POCT testing in the “QAAMS”
model for diabetes management in Australian aboriginal medical services.
Clin Biochem Rev 2006; 23: 185.
• RC Hawkins. Comparison of four point-of-care HbA1c analytical systems against
central laboratory analysis. Singapore Med J 2003; 44: 8.

3.7 Haemoglobin and prothrombin (INR): instrument technology and general
considerations
POCT for haemoglobin has been available for many years and is used in many different
medical settings. Haemoglobin testing is usually performed by a photometric procedure,
with various corrections being applied depending on the specific instrumentation. Dual
wavelength instruments usually correct for lipaemia, leukocytosis and other sources of
turbidity. Refer to suggested references for further details.
Patients on warfarin require daily monitoring of their response to the drug (particularly at
the start of treatment) until the prothrombin time (PT / “International normalized ratio”, INR)
has been stabilized within the therapeutic range. Thereafter, further monitoring at regular
intervals (every month or two) is necessary to check stability and compliance. Warfarin
also has a very narrow therapeutic window with a considerable patient to patient variation
(inter-individual biological variation). The drug activity is also susceptible to interaction
with other drugs and to a number of foodstuffs. Monitoring of PT / INR, is thus the only
reliable method to confirm a correct dosing schedule. POC monitoring of PT / INR is
widely used in medical clinics and for self testing.
Exercise
Review the application of POCT in haematology and anticoagulation therapy.

• P Harper and B Sheppard. Point-of-care testing for the management of oral

anticoagulants. In: A Practical Guide to Global Point-of-Care Testing. Edited
by Mark Shephard. CSIRO Publishing, 2016.
• M Williamson. Point-of-care testing for the markers of haematology
disorders. In: A Practical Guide to Global Point-of-Care Testing. Edited by
• GA Nuttall and P Santrach. Haemoglobin and coagulation. Chapter 13 in:
JH Nichols, Point of Care Testing, 2003.
• M Laposata. Point-of-care coagulation testing: Stepping gently forward.
Clin Chem 2001; 43: 801.

• Oral Anticoagulation Monitoring Study Group. Prothrombin measurement using a

patient self-testing system. Am J Clin Pathol 2001; 115: 288.
http://ajcp.ascpjournals.org/content/115/2/288.full.pdf+html
• ML Zucker et al. Coagulation. Chapter 4 in, National Academy of Clinical
Biochemistry, Evidence based practice for Point-of-Care Testing (see references)
or https://www.aacc.org/science-and-research/practice-guidelines/point-of-care-
testing
Search index at:
+review
o Buyers guide. Point of care coaguiometers for monitoring oral anticoagulation.
CEP report 03026, January 2008.
o Hemosense INRatio. Report 06003, March 2006.
o Patient self-testing using the Roche CoaguChek S. MHRA report 04002,
February 2004.
o ProTime 3 microcoagulation system. MHRA report 04082, June 2004.
o Eight D-dimer assays for the exclusion of deep vein thrombosis. MHRA report
03088, November 2003.
o Patient self-management using the CoaguChek S. MHRA report 05050,
August 2005.
o CoaguChek XS system. CEP report 06034, May 2006.
3.8 Urine creatinine and albumin: instrument technology and general

considerations
Microalbuminuria (minimal albumin excretion or minimal albuminuria) is the term used to
describe a low concentration of albumin in urine. Its presence signifies an increased
permeability of the small blood vessels within the kidney. This is a common complication
of diabetes, where high blood glucose levels damage small blood vessels, and can lead to
renal disease (nephropathy), cardiovascular disease and retinopathy.

Diabetic nephropathy affects 30% or more of cases of type 1 diabetes mellitus and 20% of
cases of type 2 diabetes. Approximately 30% to 35% of dialysis patients have diabetes.
Chronic hyperglycemia leads to nonenzymatic glycation of many of the body’s proteins.
Glycation of the glomerular basement membrane (GBM) appears to decrease its negative
charge, which impairs the selectively of the GBM to retain proteins. The GBM thus
becomes “leaky,” permitting proteinuria to develop. The earliest biochemical evidence of
glomerular injury is minimal albumin excretion (or microalbuminuria). Increased amounts of
albumin are excreted because albumin is the most abundant plasma protein and its
molecular size allows “leakage” when the GBM is compromised in this manner. The
traditional urine dipstick test for “protein” (albumin) using dye-binding procedures, is
usually negative at such low levels of protein excretion. Testing for microalbuminuria (or
microalbuminuria in the form of albumin:creatinine ratio, UACR) should be part of all
diabetic monitoring programs as increased levels indicate incipient nephropathy.
Qualitative dipsticks for creatinine and low albumin levels in urine are available. The
albumin test component may use a high affinity sulphonehthalein dye binding procedure or
an immunological procedure. The creatinine test component may involve the formation of
a copper-creatinine complex, which then catalyses the reaction of diisopropyl-benzene
dihydroperoxide and 3,3’,5,5’-tetramethylbenzidine to produce a green or blue colour.
POC devices for the quantitative determination of urine microalbumin use an
immunological (immunoturbidimetric) reaction for albumin. Urine creatinine is measured
by its reaction with 3,5-dinitrobenzoic acid at alkaline pH to form a red chromogen or an
enzymatic colorimetric procedure. The reaction with 3,5-dinitrobenzoic acid (DNBA) was
first proposed in 1936 but was not widely accepted as a reliable method for creatinine. In
1983 it was “rediscovered” and adapted for use with dry-strip technology by JF Stevens et
al. In this procedure, creatinine migrates through semi-permeable layers of the strip to
react with DNBA to form a coloured chromogen. The reaction was monitored by
reflectance spectrophotometry at 560 nm. The DNBA method has many potential
interfering substances, but actual interference depends on the properties of the semi-
permeable membrane (if one is present), reaction chemistry (end-point or kinetic
measurement) and actual strip design.

• M Shephard and T Mathew. Point-of-care testing for kidney disease. In: A

• National Institute of Health and Clinical Excellence (NICE), guidelines for the
management of Type 2 diabetes.
http://www.nice.org.uk/guidance/index.jsp?action=byID&o=11983
Search index at:
+review
o Buyers guide, point of care devices for the measurement of HbA1c and low
concentration of albumin in urine, CEP 08053.
o Point of care devices for detection and semi-quantitation of microalbuminuria,
MHRA 04098.
o Point of care devices for the quantitation of microalbuminuria, MHRA 04086.
• AK Aarsand et al. Diagnosis and management of diabetes mellitus. Chapter 6, in
• MP Parsons et al. Validation of point-of-care assay for urinary albumin:creatinine
ratio. Clin Chem 1999; 45: 414.
• TJ Justesen et al. Albumin to creatinine ratio in random urine samples might
replace 24h urine collections in screening for micro- and macro-albuminuria in
pregnant women with type 1 diabetes. Diabetes Care 2006; 29: 924.
• C Kvam et al. Development and performance of an albumin-creatinine ratio assay
on the Afinion AS100 analyzer. Point of Care J 2009; 8: 16.
• AC Parekh and C Sims. Serum creatinine assay by 3,5-dinitrobenzoates: A
critique. Clin Chem 1933; 23: 2066.
• JF Stevens et al. Measurement of bilirubin, cholesterol and creatinine in serum
and plasma, by solid-phase reflectance spectrophotometry.
J Clin Pathol 1983; 36: 598.
• S Narayanan and HD Appleton. Creatinine: A review. Clin Chem 1980; 26: 1119.
• K Spencer. Analytical reviews in clinical biochemistry: the estimation of creatinine.
• JA Weber and AP van Zanten. Interferences in current methods for measurements
of creatinine. Clin Chem 1992; 33: 659.

3.9 Troponin and markers of cardiac function
Coronary artery disease continues to be the most common cause of morbidity and
mortality among adults throughout the western world. The renewed interest in risk
stratification of patients with acute coronary syndromes (ACS) is largely due to an
improved knowledge of cardiac pathology, improved assay technology for existing
biochemical markers (and in particular the troponins), and new methods of treatment. An
important feature of this progress is a greater understanding of what is now called acute
coronary syndrome(s). This syndrome encompasses a variety of forms of unstable
ischaemic heart disease, and can be considered as a dynamic progression from unstable
angina through to myocardial infarction. This spectrum of coronary artery disease is
determined by total plaque burden, plaque stability and plaque fissuring, with presentation
dependant on the degree of plaque disruption, plaque location, the degree of stenosis and
the presence of thrombosis.
Error! Objects cannot be created from editing field codes.


In acute ischaemic heart disease, the most important biochemical tests are the cardiac
troponins (either troponin I or troponin T). These are proteins found exclusively in heart
muscle cells and released during the ischaemic episode and cell death. Increased blood
(plasma / serum) levels indicate heart muscle damage (myocardial infarction). There is
little evidence to indicate which of the troponins (troponin I or troponin T) is a better marker
of cardiac damage, but the ability of the assay to detect a “low” troponin level is an
important assay feature. In addition, as troponin T assays are patented, they are only
available through Roche Diagnostics.
Myoglobin, the oxygen binding protein of cardiac and skelet al muscle has also been
proposed a marker of cardiac muscle damage. It can certainly be detected at an early
stage following muscle damage, but as it is present in both cardiac and skelet al muscle, it
is not a specific marker for cardiac muscle damage.
Exercise
Review the current criteria for the diagnosis of myocardial infarction (MI).
• Myocardial infarction redefined – A consensus document of the joint

European Society of Cardiology / American College of Cardiology committee
for the redefinition of myocardial infarction. J Am Coll Cardiol 2000; 36: 959.
http://content.onlinejacc.org/content/vol36/issue3/index.dtl
• P Venge and H Halls. Point-of-care testing of cardiac markers. In: A




Issues to consider:
• The use of troponin T or troponin I. These are two different proteins with different
concentration levels, assay results must not be interchanged
• Different troponin I assays (assays from different manufacturers and possibly
different assays from the same manufacturer), may not be interchangeable
• For prompt diagnosis, turn-around-time (TAT) from blood collection to result is
important. Most guidelines suggest a TAT of 60 minutes or less, with 30 minutes
or less preferred
• Time from onset of chest pain is an important diagnostic factor … results on blood
collected less than 3 hours following chest pain may give a negative result
• Increase in cardiac troponin (either T or I) in conditions other than ACS
• Possible increase in troponinT in renal failure
• The detection level and “low end” sensitivity of troponin assays
• POC troponin assays as a screening (or diagnostic) procedure for AMI and
“comparison” with laboratory troponin values (which may give different troponin
values, refer to comments above).
• Why is prompt diagnosis and initiation of treatment so critical for patients
suspected of MI ?
• When should specimens for cardiac biomarkers (troponin) be collected from
patients with suspected MI ?
• How important are changing levels of cardiac biomarker (Troponin) in the diagnosis
of MI ?
Heart failure or congestive cardiac failure, is ineffective pumping by the heart. This leads
to fluid accumulation (congestion) in the lungs. Incorporated in the formal definition of
heart failure is a wide spectrum of clinical conditions which relate to impairment of pump
function. Brain (or B-type) natriuretic peptide (BNP), is a hormone originally found in
porcine brain tissue. However, its main source of release is from ventricular myocardium
in response to elevations of end-diastolic pressure and volume. The main circulating
forms are proBNP, the N-terminal fragment of proBNP (NT-proBNP), and the
physiologically active BNP. Measurement of BNP or NT-proBNP give a biochemical
assessment of ventricular function. Blood (plasma / serum) levels are increased in chronic
heart failures and correlate closely with its severity. A normal BNP virtually excludes left
ventricular systolic dysfunction (LSVD), whereas an elevated BNP (or NT-proBNP)
indicates the presence of cardiac disease and the need for an echocardiogram. While
increased BNP is indicative of LSVD, it is also elevated in other cardiovascular conditions
such as pulmonary hypertension and right ventricular overload.
There have been many recent articles describing the merits of various Troponin and BNP
assays and their role in diagnosis and clinical management. POC devices are widely
available for Troponin and / or BNP measurement. All systems utilise an immunochemical
assay procedure, the specific details of which are very much method or device related.
• European Society of Cardiology and American College of Cardiology. Myocardial

infarction redefined – A consensus document of the joint European Society of
Cardiology / American College of Cardiology committee for the redefinition of
myocardial infarction. J Am Coll Cardiol 2000; 36: 959.
http://content.onlinejacc.org/content/vol36/issue3/index.dtl
• PO Collinson et al. Impact of European Society of Cardiology / American College
of Cardiology guidelines on diagnostic classification of patients with suspected
acute coronary symdromes. Ann Clin Biochem 2003; 40: 156.
• AHB Wu et al. National Academy of Clinical Biochemistry standards of laboratory
practice: Recommendations for the use of cardiac markers in coronary artery
disease. Clin Chem 1999; 45: 1104.
• Association of Clinical Biochemists in Ireland. Guidelines for the use of
biochemical cardiac markers and risk factors, ACBI 2002.
• M Panteghini, IFCC Committee on Standardisation of Markers of Cardiac Damage.
Recommendation on use of biochemical markers in acute coronary syndrome:
IFCC proposals. ejifcc volume 14, number 2.
http://www.ifcc.org/ifccfiles/docs/1402062003014.pdf
• National Heart Foundation, Cardiac Societies of Australia and New Zealand.
Guidelines for the management of acute coronary syndromes 2006. MJA 2006;
184, number 8: supplement. http://www.heartfoundation.org.au/information-for-
professionals/Clinical-Information/Pages/acute-coronary-syndrome.aspx

• M Gorenjak. Natriuretic peptides in assessment of ventricular dysfunction. ejifcc
volume 14, number 2. http://www.ifcc.org/ifcc-communications-publications-
division-(cpd)/ifcc-publications/ejifcc-(journal)/e-journal-volumes/
• A Clerico and M Emdin. Diagnostic accuracy and prognostic relevance of the
measurement of cardiac natriuretic peptides: A review. Clin Chem 2004; 50: 33.
• AB Storrow et al. Use of cardiac biomarkers for acute coronary syndromes.
Chapter 3, in National Academy of Clinical Biochemistry, Evidence based practice
for Point-of-Care Testing (see references) or https://www.aacc.org/science-and-
• PO Collinson et al. Measurement of cardiac troponins (review). Ann Clin Biochem
2001; 38: 423.
• EJ Lamb et al. The significance of serum troponin T in patients with kidney
disease: A review of the literature. Ann Clin Biochem 2004; 41: 1.
• WE Kelley et al. Increases of cardiac troponin in conditions other than acute
coronary syndrome and heart disease (review). Clin Chem 2009; 55: 2098.
• AAA Ismail et al. Interference in immunoassay is an underestimated problem. Ann
Clin Biochem 2002; 39: 366.
• DC Gaze and PO Collinson. Multiple molecular forms of circulating cardiac
troponin: Analytical and clinical significance (review). Ann Clin Biochem 2008; 45:
349.
• JH Nichole et al. Clinical outcomes of Point-of-Care testing in the interventional
radiology and intensive cardiology setting. Clin Chem 2000; 46: 543.
• TJ Clark et al. The Triage cardiac panel. Point of care J 2002; 1: 42.
http://journals.lww.com/poctjournal/Fulltext/2002/03000/The_Triage_Cardiac_Pane
l__Cardiac_Markers_for_the.11.aspx
• PO Collinson et al. A randomised controlled trial of point-of-care testing on the
coronary care unit. Ann Clin Biochem 2004; 41:393.
• FS Apple et al. Future biomarkers for detection of ischemia and risk stratification in
acute coronary syndrome. Clin Chem 2005; 51:810.
• G Wilcox et al. Measurement of cardiac troponin I levels in the Emergency
Department: Predictive value for cardiac and all-cause mortality. MJA 2001; 134:
130.
• AHB Wu. Risk stratification of cardiac troponin in ischaemic and non-ischaemic
cardiac diseases and procedures. Clin Biochem Revs 2000; 21: 39.

• P Tiderman. Risk stratification of cardiac patients: Point of care testing, progress
and practicalities. Clinical Biochemist Newsletter June 2004, page 18.
• M Panteghini et al, IFCC committee. Quality specifications for cardiac troponin
assays. Clin Chem Lab Med 2001; 39: 134.
Search index at:
+review
o Nine troponin assays (CEP report 05085, December 2005).
o Three point of care devices for Troponin measurement (CEP report 06020, May
2006).
3.10 Pregnancy testing

Chorionic gonadotropin (human chorionic gonadotropin, hCG) is a glycoprotein hormone
produced by trophoblastic tissue. This is found in the placenta in normal pregnancy, in
choriocarcinoma and in trophoblastic elements in germ cell tumours. hCG secretion may
also occur in a range of other tumours of the reproductive system, lung and
gastrointestinal tract. Chorionic gonadotropin consists of two subunits, the alpha subunit
being identical to the alpha subunit of the pituitary glycoprotein hormones (lutrophin (LH),
follitropin (FSH) and thyrotropin (TSH)). The beta subunit, even though similar to that in
lutropin, is larger and “unique” to hCG. The beta hCG subunit can be measured
“specifically” by monoclonal antibody procedures (beta-hCG assays).
Chorionic gonadotropin is found in maternal serum and urine in several forms, including
whole unmodified hCG, and forms modified by various degrees of degradation. Different
assay systems react in differently ways with these various forms of hCG. In particular,
tumours may produce a variety of hCG’s, so great care is required in assessing results.
Current urine POC pregnancy tests (home tests or POC tests used in a doctor’s office, etc)
or POC blood pregnancy tests are based on immunoassay using a monoclonal antibody
directed to epitopes on the beta subunit of hCG. Depending on assay sensitivity, a
http://en.wikipedia.org/wiki/Human_chorionic_gonadotropinpositive urine test may be obtained
6 to 12 days after fertilization. Sensitive quantitative blood test(s) (serum beta-hCG), can
detect hCG levels as low as 2 IU/L, while urine tests (depending on the brand) have
published detection thresholds between approximately 20 IU/L to 50 IU/L. Qualitative

blood tests generally have a detection limit of around 25 IU/L. Almost all of the current
POC urine pregnancy tests use a form of immuno-chromatography called lateral flow
immunoassay (see supplementary references below). Both false negative and false
positive tests can occur and test results must always be interpreted in relation to the
patient’s clinical situation. Analytical artifacts relating to the “Hook effect” (where a very
high result actually gives a negative result) and heterophilic antibodies affect POC devices
in a similar manner to laboratory tests. Similarly, unusual physiological or pathological
situations may give positive results which are not indicative of pregnancy.
• KM Koczula and A Gallotta. Lateral flow assays. Essays in Biochemistry 2016;

60: 111. (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4986465/ )
• EB Bahadir and MK Sezginturk. Lateral flow assays: Principles, designs and
labels. Trends in Analytical Chemistry 2016; 82: 286.
• AM Gronowski et al. Reproductive testing, chapter 13 in National Academy of
Clinical Biochemistry, Evidence based practice for Point-of-Care Testing (see
references) or https://www.aacc.org/science-and-research/practice-
guidelines/point-of-care-testing
• LM Boscato and MC Stuart. Heterophilic antibodies: a problem for all
immunoassays. Clin Chem 1988; 34: 23.
• LA Cole and A Kardana. Discordant results in human chorionic gonadotropin
assays. Clin Chem 1992; 38: 263.
• RJ Norman et al. Simple quantitative measurement of serum chorionic
gonadotropin compared with immunoradiometric, immunoenzyme, and
chemiluminescent assays. Clin Chem 1992; 38: 144.
• J Daviaud et al. Reliability and feasibility of pregnancy home-use tests: Laboratory
validation and diagnostic evaluation by 638 volunteers. Clin Chem 1993; 39: 53.
• LA Cole. Immunoassay of human chorionic gonadotropin, its free subunits and
metabolites. Clin Chem 1993; 43: 2233.
• LA Cole et al. False positive hCH assay results leading to unnecessary surgery
and chemotherapy and needless occurrences of diabetes and coma.
Clin Chem 1999; 45 313.

• SA Butler et al. Detection of early pregnancy forms of human chorionic
gonadotropin by home pregnancy test devices. Clin Chem 2001; 43: 2131.
• LA Cole. “Background” human chorionic gonadotropin in healthy, nonpregnant
women. Clin Chem 2005; 51:1365.
• TK Er et al. False negative pregnancy test in hydatiform mole.
Clin Chem 2006; 52: 1616.
• CR Mc Cudden et al. Persistant low concentration of human chorionic
gonadotropin in a nonpregnant woman. Clin Chem 2008; 54 209.
• AM Gronowski et al. False negative results in point-of-care qualitative human
chorionic gonadotropin (hCG) devices due to excess hCG beta core fragment.
Clin Chem 2009; 55: 1389.
• Device Evaluation Service (DES) and Centre for Evidence Based Purchasing
(CEP) of the UK National Health Service (NHS).
(1) A buyers guide to pregnancy tests. December 2006, CEP report 06053.
(2) Over the counter pregnancy tests. December 2006, CES report 06051.
Search index at:
+review
3.11 Tests for other substances using POCT technology

There are many substances which can be measured in a point of care situation. The
medical requirement for assisting with diagnosis or the importance for monitoring drug
therapy usually determined which actual analytes are measured. In addition to medical
needs however, POC testing by Police for ethanol, work-place testing for drugs of abuse
and self-testing for specific substances such as hCG and glucose are also common
practice. Microbiological testing for infectious substances is also becoming an important
application for POC testing as is the detection of chemical and biological agents released
into the environment. The increasing range of POC testing is clearly outlined in texts such
as Point-of-care testing. Needs, opportunity and innovation (Editors CP Price, A St John
and LJ Kricka. Third edition, AACC Press, 2010) or A Practical Guide to Global Point-of-
Care Testing (Edited by Mark Shephard. CSIRO Publishing, 2016) .
Additional important POCT applications include:

• Acid base, blood gas and electrolytes (Na+, K+, Ca++, Cl-, etc), see topic 4

• Co-oximitry
• Drugs of abuse (amphetamines, cannabinoids, cocaine, opiates, …), see later
topic in this course.
• Ethanol
• Lithium
• Lactate
• Intraoperative parathyroid hormone
• Occult blood
• Transcutaneous bilirubin
• Ovulation testing (lutropin, LH)
• Foetal fibronectin
• Infectious disease testing (bioterrorism, Chlamydia, HIV, infectious
mononucleosis, group A streptococcal antigen, group B streptococci,
Helicobacter pylori, influenza, Trichomonas, Candida, …)
• Various chapters in: A Practical Guide to Global Point-of-Care Testing. Edited by

Particularly the chapters on :
27. POCT in general practice.
28. POCT in rural, remote and indigenous settings.
30. POCT in paramedic services.
31. POCT in disaster management.
32. POCT – the Australian Antarctic Division.
33. POCT in sports science.
• Various chapters in: Point-of-care testing. Needs, opportunity and innovation.
Editors CP Price, A St John and LJ Kricka. Third edition, AACC Press, 2010.
• JH Nichols. Point of Care Testing, 2003.
• National Academy of Clinical Biochemistry, Evidence based practice for Point-of-
• D Rowland. Diagnosing infectious disease at point of care. Clinical Biochemist
Newsletter, June 2004. Article available in topic library.
• KN Hale etal. POCT for disaster and pandemic preparedness. Chapter 27 in:
Point-of-care testing. Needs, opportunity and innovation. Edited by Price CP, St
John A and Kricka LJ. AACC Press 2010.
Suggested reading
• M Shephard. Selection and evaluation of point-of-care testing devices. In: A

And / Or
• LJ Kricka and GHG Thorpe. Technology of hand-held devices for POCT.

Chapter 3, in: Point-of-care testing. Needs, opportunity and innovation. Edited
by Price CP, St John A and Kricka LJ. AACC Press, 2010.

Topic 4 – Regulatory Requirements, Standards, Guidelines and

Documentation of Procedures
 Learning objectives for topic 4

 Overview of topic 4 (overview of regulatory requirements)
 International standards
 International Organisation for Standardisation (ISO) and Standards Australia
 “Minimum” or “appropriate” standards
 Pathology regulations and funding in Australia
 National Pathology Accreditation Advisory Council (NPAAC)
 National Association of Testing Authorities (NATA)
 Workplace drug testing and specimen collection accreditation for drugs of abuse testing
 Review – Policy, procedures and guidelines for POCT
 POCT guidelines from professional associations
o AACB
o AACC
o ACBI
o NACB
 Accreditation
 Overview of American (USA) regulations
 Current issues with POCT quality in USA
 Occupational Health and Safety (OH&S)
 Documentation of procedures: regulatory requirements or good laboratory practice
 Mandatory reading for topic 4
On completion of this topic, participants should:

 Understand the differences between regulatory requirements (mandated by legislation /
regulation), standards such as those produced by ISO and NPAAC (which, in countries
such as Australia, may also be mandated by regulation), and professional and clinical
practice guidelines.
 Be able to describe the procedures required by regulation, the relevant standards or
guidelines.
 Understand how standards and professional guidelines may assist with the quality
management of POCT.
 Be able to describe the role which various agencies play in the regulatory, quality
assessment and accreditation process.
 Be able to find relevant regulatory information, standards and guidelines when required.

4.2 Overview of topic 4 (overview of regulatory requirements)
Formal policy and regulation generally applies to situations where service payments may be
service fee for pathology or POC testing. In return for payment, the funding agency expects quality
management procedures to have been implemented, including quality control and external quality
assessment. The advantages and clinical utility of POCT are largely dependent on proof of
acceptable analytical performance. This concept dominates policy statements from government
agencies of many jurisdictions, articles published in peer reviewed journals, and statements made
by international professional commentators on POCT. The paper by Tirimacco (Design,
implementation and outcomes for POCT: cost implications. POC 2008; 7(3): 128 ) provides a
good summary of these quality concepts:
“Although POCT appears to be deceptively simple, if incorrectly performed it may present a
risk to patient care and, if used inappropriately or overused can lead to significant increases in
the cost of patient care. To ensure results obtained are comparable to the traditional
pathology laboratory, POCT should be implemented within a framework of quality
standards.” … “This quality framework should include: operator education, training and
competency, quality control, proficiency testing and accreditation.”
Evidence-based medicine has now become the international standard for improving health care.
Evaluation of new technologies and clinical procedures should therefore be evaluated using
evidence-based medicine processes which include technical, cost-efficiency and patient benefit
perspectives. At an elementary level, patients as consumers of health services assume quality of
care to be synonymous with freedom from medical error, which includes freedom from errors in
laboratory and POC testing. For patient safety to be achieved within the current context (of
laboratory and POC testing), quality goals for both analytical and clinical management must be
given a high priority.
The innovation and functionality of POCT brings many challenges for health care funding
authorities. In particular, the ability to determine the value which POCT may bring to the patient
care process. The penultimate sentence of chapter 1 in Point-of-care testing: Needs, opportunity
and innovation, by CP Price, A St John and LJ Kricka (AACC Press 2010), states that:
“The four main challenges are, (a) producing the evidence to demonstrate that POCT
improves outcomes, (b) changing clinical practice to deliver the benefit, (c) maintaining
clinical governance for a more distributed laboratory medicine service, and (d) adjusting the
resource allocation to reflect the likely increase in investment in the POCT technology, while
recouping the resources from the point in the pathway at which the benefits are made.”
Many countries have developed their own “standards” for laboratory practice and POCT. Most
have been developed by expert professional committees in association with the relevant
professional colleges and associations. Adherence to these “standards” is often a requirement for
government funding, but compliance is also related to the desire for quality testing and good patient
care. The threat of litigation in some countries for poor quality testing may also be an incentive to
comply with a local standard or professional guideline.
4.3 International standards (see also item 4.4 below)
The international standard for POCT is ISO 22870, Point-of-care testing – requirements for quality
and competence. This standard, produced by the International Organisation for Standardisation
(ISO), gives specific requirements applicable to POCT and is intended to be used in conjunction
with ISO 15189, Medical laboratories – particular requirements for quality and competence. The

requirements of ISO 22870 “apply when POCT is carried out in a hospital, a clinic, or healthcare
organisation providing ambulatory care”.
Patient self-testing is excluded from ISO 22870, but is covered specifically in ISO 15197 In vitro
diagnostic test systems – requirements for blood glucose monitoring systems for self-testing in
managing diabetes mellitus, and ISO 17593 Clinical laboratory testing and in vitro medical devices
– requirements for in vitro monitoring systems for self-testing of oral anticoagulant therapy.
The term standard is itself defined by ISO as a document describing … “a set of rules that control
how people develop and manage materials, products, services, technologies, processes, and
systems.” ISO standards are developed by technical committees whose members represent many
countries and professional groups and generally have very broad international support. Australia's
representative (member organization) to ISO is Standards Australia (see item below, National
standards and guidelines), which also provides a similar definition for the term standard
( http://www.standards.org.au/DevelopingStandards/WhatisaStandard.aspx ).
4.4 International Organisation for Standardisation (ISO) and Standards Australia
International Organisation for Standardisation (ISO) is a network of national standards institutes

from 163 countries, one member per country, with a central secretariat in Geneva, Switzerland
(Australia is represented by “Standards Australia”).
ISO is a non-government organization that forms a link between the public and private sectors.
Many of ISO’s member institutes are part of the governmental structure of their respective
countries, while others may be national partnerships of industry (professional) associations.
Standards Australia coordinates standardisation activities within Australia. It develops

internationally aligned Australian Standards and facilitates the accreditation of other standards
development organisations. Standards Australia is not related to NPAAC, which as described
above is a separate government advisory group specialising in pathology matters and pathology
standards. However, as Standards Australia is the official representative to ISO (the developer
of ISO 15189 and ISO 223780), ISO standards are distributed in Australia through Standards
Australia.
 International Organisation for Standardisation, ISO. http://www.iso.org/iso/home.htm

 ISO 15189. Medical laboratories -- Particular requirements for quality and competence.
http://infostore.saiglobal.com/store/results2.aspx?searchType=simple&publisher=all&keyw
ord=15189
 ISO/FDIS 22870. Point-of-care testing (POCT) -- Requirements for quality and
competence
http://infostore.saiglobal.com/store/Details.aspx?ProductID=276065
4.5 “Minimum” or “appropriate” standards ?
The term minimum standard is often used to describe the least permissible condition or procedure
required to demonstrate a basic level of performance (or provision of test results in the current
context). The term minimum is often used by government agencies which require adherence to
some basic procedure, but in practice this terminology actually encourages the development of low
quality testing. In circumstances where quality is not considered of primary importance, minimum
usually becomes synonymous with actual. When discussing any testing situation where quality is

an important component of the expected outcome, appropriate standard(s) which provide the
desired level of measurement uncertainty are required.
For POC and quantitative pathology testing in general, appropriate analytical tolerance levels are
defined by quality goals based on biological variation or internationally recognized clinical decision
values. The desirable analytical variability which is required to insure that a test is fit for purpose is
discussed in later topics of this course. Failure to meet appropriate standards and failure to
demonstrate that quality once attained is also maintained, may jeopardize patient safety and
contribute to a poor health outcome. Incorrect or misleading test results which initiate inappropriate
treatment may cause more harm than no results at all. Appropriate standards which support quality
testing should be the desired approach, not minimum standards which provide the opportunity for
quality failures.
4.6 Pathology regulations and funding in Australia
Regulations governing the performance of pathology testing in Australia are designed to insure a
“minimum standard” (re-read previous section !) and to assure the public (the consumers of
pathology testing) that appropriate testing standards are being met. In a similar manner, many
countries (for example, the USA with its CLIA regulations) regulate to insure a “minimum testing
standard”.
These regulatory requirements are often associated with laboratory accreditation, which in itself
defines the standard which is required. Such regulations or the requirement for a testing laboratory
to become an accredited laboratory, are usually linked to government fee payments. The
requirement for accreditation is in itself a form of regulation, with the procedures and “standards”
being defined as part of the accreditation process.
In Australia, the government agency responsible for defining and setting standards for the operation
of pathology laboratories is the National Pathology Accreditation Advisory Council (NPAAC). The
rules defined by NPAAC and then approved by the Minister for Health and Ageing (on behalf of
the Australian Federal Government), are used by the agency contracted to undertake the actual
laboratory inspections and recommend accreditation. The inspection agency in Australia is the
National Association of Testing Authorities (NATA). The government agency responsible for fee
payments is Medicare. In order to claim pathology testing fees, a laboratory must be accredited to
perform the particular test(s) in a relevant laboratory category (haematology, chemistry,
microbiology, immunology, tissue pathology, cytology, cytogenetics, infertility and pregnancy
tests). A separate category within the pathology section of the Medicare Benefits Schedule (MBS),
allows a medical practitioner (general practitioner, community doctor, family doctor, etc) to
perform a limited range of “simple basic pathology tests” without the requirement for accreditation.
These “simple basic pathology tests” are defined in “group P9” of the pathology section of the fee
schedule.
Regulation for POCT is largely incorporated into the framework of pathology laboratory regulation,
associated accreditation and government fee payment(s). As yet, there are no specific regulatory
requirements for POCT. The following internet references may be of assistance (even though they
relate to general pathology testing):
 Medicare benefits schedule (MBS) and test categorisation. To search the MBS, use
“advanced search” and select “category” and “group” from drop down list.
http://www.health.gov.au/internet/mbsonline/publishing.nsf/Content/Downloads-2015-04
 MBS “chemistry” section, item description 66500:
http://www9.health.gov.au/mbs/search.cfm?cat1=147&cat2=151&cat3=&adv
Item 66500 is the first item in the “chemistry” schedule, scroll down for other items.
With the exception of dip-stick tests, general biochemical tests performed by reagent strip (with or
without a reflectance meter), and other tests within group P9 of the MBS (see above), operators of
POCT devices wishing to claim a testing fee through Medicare (a Medicare benefit) are required to
be part of an accredited laboratory service operating within one of the NPAAC categories (see
below, requirements for the supervision of pathology laboratories). There are no MBS fees for dip-
stick tests or general biochemical tests (MBS section P2, test item 66500) which are measured by
“reagent strip (with or without reflectance meter)”. However, if there is no requirement to claim a
Medicare benefit (as POCT used within a public hospital) or if the patient is prepared to pay a “fee
for service”, then no accreditation is required (that is, there are no regulations covering testing).
As suggested above, the real role of regulation is to insure that testing is performed to a “minimum
standard” by a trained operator, that appropriate quality control (QC) is undertaken, that
participation in an approved proficiency testing (PT) program (often called external quality
assessment, EQA or QA) which demonstrates on-going performance quality and that clinically
relevant results are provided to the patient.
The real issue here in Australia (and is part of a continuing debate), is whether the Australian
Federal Government (through its Department of Health and Ageing) mandate in some way that QC
and PT be required for POCT if Government fees (Australian Medicare) are to be claimed for non-
accredited laboratories (essentially general medical practitioners). The "more complex" types of
test (those traditionally performed in a central pathology laboratory) such as HbA1c, urine albumin,
urine creatinine, iSTAT tests, lipids, INR and clotting factor testing etc, are the real items of debate.
Most of the technical and clinical issues with regard to these tests will be discussed in some detail in
forthcoming sections of this course.
The Australian Federal Government (Department of Health and Ageing) however, does provide
paid POCT within a special funding agreement for the Quality Assurance for Aboriginal and Torres
Strait Islander Medical Services (QAAMS) program. Funding for this program requires regular QC
and the participation in EQA. More details with regard to QAAMS can be found at:
 http://www.qaams.org.au/ and
 http://www.health.gov.au/internet/main/publishing.nsf/Content/health-pathology-qaams-
index.htm
The Department of Health and Ageing have also conducted an extensive trial of the use of POCT in
general medical practice, see references or report details available at:
 http://www.health.gov.au/internet/main/publishing.nsf/Content/health-pathology-poctt-
index.htm
The executive summary of this trial contains useful information which is relevant to this POCT
course.
An important component of the trial was the development of Interim standards for point-of-care
testing in general practice. These standards formed the basic quality management manual used by
the trial participants and were prepared by the Royal Australian College of General Practitioners in
association with the Department of Health and Ageing general practice trials’ technical advisory
committee. The introduction to these interim standards provide a succinct overview of the
requirement for appropriate POCT standards, which incorporate training and competency, clinical
governance, quality control and the requirement for participation in an external proficiency testing
program.

Mandatory reading
 Tirimacco R. Evolution of point-of-care testing in Australia. Clin Biochem Rev

2010; 31: 75. Article available in topic library.
4.7 National Pathology Accreditation Advisory Council (NPAAC)
In Australia, the government agency responsible for defining and setting standards for the operation
of pathology laboratories is the National Pathology Accreditation Advisory Council (NPAAC). The
rules defined by NPAAC and then approved by the Minister for Health and Ageing (on behalf of
the Australian Federal Government), are used by the agency contracted to undertake laboratory
inspections and recommend accreditation.
With the introduction of an international standard for pathology laboratory testing by the
International Standards Organisation (ISO), NPAAC agreed to use this approach as the primary
standard for pathology testing laboratories in Australia. Implementation of this standard (referred to
as ISO 15189, see “ISO” below) on behalf of NPAAC commenced in 2005.
 NPAAC home: http://www.health.gov.au/npaac
 NPAAC publications (regulations and guidelines):
http://www.health.gov.au/internet/main/publishing.nsf/Content/health-npaac-
publication.htm
 NPAAC: Requirements for the supervision of pathology laboratories.
http://www.health.gov.au/internet/main/Publishing.nsf/Content/health-npaac-docs-
supervision2007.htm
4.8 National Association of Testing Authorities (NATA)
The agency contracted to undertake actual laboratory inspections and to recommend accreditation in
Australia is the National Association of Testing Authorities (NATA).
 NATA home: http://www.nata.com.au
 NATA publications: http://www.nata.com.au/nata/accreditation-publication
4.9 Workplace drug testing and specimen collection; accreditation for drugs of
abuse testing
In Australia and many countries, drug testing which is required for legal purposes (for example;
workplace testing, where a positive test involves a penalty or dismissal), requires confirmatory
testing to be performed. In Australia this would require compliance with Australian Standard AS
4308. The use of POCT for drugs of abuse in both medical and non-medical settings raises many
issues, including; operator training, quality control, quality assurance, accuracy of results,
limitations of the technology and economics. In particular, POC drug testing is generally designed
to screen for the presence of designated drugs or groups of drugs. None are intended to serve as a
definitive or confirmation test.

Both AS/NZS 4308:2008 Procedures for specimen collection and the detection and quantitation of
drugs of abuse in urine and AS 4760:2006 Procedures for specimen collection and the detection
and quantitation of drugs in oral fluid outline the procedures required for a laboratory to
demonstrate its technical competency. They are prescriptive in the procedures required to ensure
appropriate specimen collection, handling, storage, transport, on-site screening, the use of quality
control and proficiency testing, initial testing and confirmatory testing. A number of these issues
are further discussed the topic relating to drugs of abuse.
The procedures are intended for (but not limited to) medico-legal, workplace, correctional services
or court directed testing, of any or all of the following classes of drugs:
 amphetamine type substances.
 benzodiazepines.
 cannabis and metabolites.
 cocaine and metabolites.
 opiates.
The Standards cover all aspects of specimen collection and testing, and defines procedures which
must be followed to allow proper legal scrutiny. These procedures include:
 specimen collection, storage, handling and dispatch (chain of custody)
 integrity and identity of the specimen
 general laboratory requirements
 laboratory screening procedures and test cut-off levels (cut-off drug concentrations),
recording of results and reporting
 laboratory security, specimen reception, specimen integrity and storage
 quality control and proficiency testing (external quality assessment)
 laboratory confirmatory procedures
 laboratory personnel and training
 Mighani RA. Current status of point-of-care testing in clinical toxicology. Point of Care
2008; 7: 266.
 Watson ID, etal. Chapter 7, Drugs and ethanol. In: National Academy of Clinical
Biochemistry (NACB). Laboratory medicine practice guidelines: Evidence-based practice
for point-of-care testing. 2006. https://www.aacc.org/science-and-research/practice-
 AS/NZS 4308. Procedures for specimen collection and the detection and quantitation of
drugs of abuse in urine.
http://infostore.saiglobal.com/store/results2.aspx?searchType=simple&publisher=all&keyw
ord=drugs+of+abuse
 NATA drug testing information relating to AS/NZS 4308
http://www.nata.com.au/nata/news/latest-news/728-nata-guide-to-industry-workplace-drug-
testing

4.10 Review – Policy, procedures and guidelines for POCT
The policies, procedures and guidelines which govern POCT vary considerably. Except for the
basic instructions which are provided by device manufacturers, there are often no specific
procedures or regulations which govern POCT. In contrast to this, there are many government
organisations, government health departments, area health authorities or individual hospitals which
have well defined policies and procedures to regulate the use and quality management of all POCT
within their jurisdiction. The innovation, apparent simplicity and functionality of POCT also
provide many challenges for health care funding authorities. In particular, the ability to determine
the value which POCT may bring to the patient care process. In 2011, a review of existing policies,
procedures or guidelines which governed the use of POCT was commissioned by the Department of
Health and Ageing, Australian Commonwealth Government. This review provides a summary of
international and Australian standards, policy and guidelines which apply to POCT. An abridged
version of this report is available on the internet.
Formal policy and regulation generally applies in situations where service payments may be
service fee for POCT. Excluding home use and personal testing, in countries without specific
overall regulation there is little information regarding the use of POC devices in the “private”
healthcare setting or in situations where patients are prepared to pay directly for testing services.
A copy of Review - Policies, procedures and guidelines for point-of-care testing is available in
the POCT references or may be downloaded from http://www.aacb.asn.au/documents/item/635
4.11 POCT guidelines from professional associations
The term guideline as used in medicine (or more often clinical practice guideline) is usually
defined as … “systematically developed statements to assist practitioners with decisions about
appropriate health care for patients in specific circumstances”. Guidelines are designed to support
the decision-making processes in patient care. Clinical practice guidelines are based on the best
available evidence derived from a systematic evidence-based review of available data.
Standards and guidelines developed for POCT using evidence-based procedures are designed to
assist with the implementation, management, operation and on-going quality assessment of the
selected technology. With adherence to appropriate standards and guidelines, POCT provides
significant benefits for both patients and healthcare providers. As stated in the introduction to ISO
22870 “Advances in technology have resulted in compact, easy-to-use in vitro diagnostic medical
devices that make it possible to carry out some examinations at, or close to, the location of the
patient. Point-of-care or near patient testing may benefit the patient as well as healthcare facilities.”
Provided that … “risk to the patient and to the facility can be managed by a well-designed, fully
implemented quality management system that facilitates:
 evaluation of new or alternate POCT instruments and systems
 evaluation and approval of end-user proposals and protocols
 purchase and installation of equipment
 maintenance of consumable supplies and reagents
 training, certification and recertification of POCT systems operators
 quality control and quality assurance.”
In addition to ISO, NPAAC or government standards, professional organisations such as the

Australasian Association of Clinical Biochemists (AACB) and the American Association for
Clinical Chemistry (AACC) publish recommendations (or guidelines) on the preferred approach to
managing and operating POC devices. These are professional opinions aimed at outlining “best
practice”. They are not regulations and generally have no legal standing. They may however, be
used to support arguments as to what should be considered as good analytical practice.
 Australasian Association of Clinical Biochemists (AACB)

Refer to documents in topic library.
 The Association of Clinical Biochemists in Ireland (ACBI)
 National Academy of Clinical Biochemistry, Evidence based practice for POCT.
https://www.aacc.org/science-and-research/practice-guidelines/point-of-care-testing
 Point of care.net. http://www.pointofcare.net
Mandatory reading
 E Jacobs etal. POCT: Standards, guidelines and governance. Chapter 16 in: Point-
of-care testing. Needs, opportunity and innovation. Edited by Price CP, St John A
and Kricka LJ. AACC Press 2010
4.12 Accreditation
In Australia, the USA and United Kingdom, accreditation of pathology services are required for
government funding (however this payment may be made: by “block grant” or fee reimbursement).
http://www.nata.com.au/nata/accreditation-publication/nata-field-and-program-accreditation-
criteria/category/64-iso-15189-medical-testing
4.13 Overview of American (USA) regulations
In 1988, the US government passed the Clinical Laboratory Improvement Amendments act (CLIA
88 or CLIA), which specified regulations and standards for all facilities in the United States that
perform laboratory testing on human specimens for health assessment or the diagnosis, prevention,
or treatment of disease. CLIA classifies tests by a complexity model rather than testing location or
intended use and tests are defined as waived, of moderate complexity or high complexity. Waived
tests include test systems cleared by the Food and Drug Administration (FDA) for home use and
those tests approved for waiver under the CLIA criteria. The only criteria for non-accredited
laboratories to perform waived tests is for the device to have been cleared by the FDA, for operators
to follow manufacturer’s instructions as a minimum requirement (including perform quality control
(QC) at the frequency defined in the manufacturers package insert), and to allow random inspection
of the testing facility to establish compliance. Of note however, is that all laboratories which
perform testing on human samples (with the exception of defined research laboratories and personal
home use testing) are required to be certified (registered) by the Centers for Medicare and Medicaid
Services (see below) in order to perform any testing procedure. In this regard, “laboratory”
essentially means any site where testing occurs, including a hospital ward, a doctors office,
community testing facility or ambulatory testing service.
The CLIA requirements are administered through the Centers for Medicare and Medicaid Services
(CMS), which regulate all laboratory testing (except research) performed on humans. CMS is the
agency which provides testing certificates; a laboratory which performs moderate complexity or
high complexity tests requires a certificate of compliance and a laboratory which performs only
waived tests may apply for a certificate of waiver. In total, CLIA covers approximately 200,000
laboratory entities. The Division of Laboratory Services, within the Survey and Certification
Group, under the Center for Medicaid and State Operations has the responsibility for implementing
the CLIA program.
In addition to CMS which administers the CLIA program, the Secretary of Health and Human
Services under whose authority and oversight the CLIA program operates, is required to develop
appropriate standards to assure consistent, accurate, and reliable test results by all clinical
laboratories. To this end, the Secretary is authorized to establish advisory committee(s) to assist
with this process. The Clinical Laboratory Improvement Advisory Committee (CLIAC) was
established in 1992 “to provide scientific and technical advice and guidance to the Secretary and the
Assistant Secretary for Health regarding the need for, and the nature of, revisions to the standards
under which clinical laboratories are regulated; the impact on medical and laboratory practice of
proposed revisions to the standards; and the modification of the standards to accommodate
technological advances”. [Authors comment: The US CLIAC advisory committee appears very
similar to the Australian NPAAC].
A twenty year review of CLIA regulations with comments from the principal agencies involved
with laboratory regulation and accreditation is provided by the CLIAC summary report for the
meeting of February 2008, “Recognising 20 years of CLIA” (Clinical Laboratory Improvement
Advisory Committee (CLIAC). Summary report of meeting, “Recognising 20 years of CLIA”.
February 2008. http://wwwn.cdc.gov/cliac/pdf/CLIAC0208.pdf ). In addition, the complete set of
CLIAC meeting summaries is available through the Centers for Disease Control and Prevention
(CDC) (CDC information. Clinical Laboratory Improvement Advisory Committee (CLIAC).
Summary reports of meetings (scroll down) http://wwwn.cdc.gov/cliac/default.aspx ).
Supplementary reference
The following reference provides additional useful information
 SS Erhmeyer. Regulatory issues regarding POCT. Chapter 19 in: Point-of-care testing.

Needs, opportunity and innovation. Edited by Price CP, St John A and Kricka LJ. AACC
Press 2010.
4.14 Current issues with POCT quality in USA

The original list of waived tests was relatively small and consisted mainly of “dipstick” tests for
urine and some monitoring devices cleared by the FDA for home-use (for example blood glucose).
However, the list of wavered tests has grown considerably over the years and was revised in 2008 to
include a large number of analytes, including many which can be performed at POC.
(Clinical Laboratory Improvement Amendments - Currently waived analytes.
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfClia/analyteswaived.cfm ). Tests which
remain in the moderate complexity category (and thus require full accreditation) include blood gas
instruments and all activated clotting time tests.
With the increased use of waived tests outside of the accredited laboratory environment, there has
also been increasing concern as to the quality of results and the potential harm to patients. The
identification of quality issues in waived testing sites has been compiled from on-site inspection
reports obtained by CMS during the period 1999 to 2003 and summarized by the CDC in their
report and recommendations paper Good Laboratory Practices for Waived Testing Sites, 2005
(Centers for Disease Control and Prevention (CDC). Good laboratory practices for waived testing
sites. http://www.cdc.gov/mmwr/PDF/rr/rr5413.pdf ). In addition to a review of quality failures,
this CDC document provides recommendations developed by CLIAC for improving the quality of
waived tests.

In addition to the quality issues identified through CMS inspections, there is currently an ongoing
debate within the scientific community and within FDA with regard to the quality of blood glucose
and haemoglobin A1c (HbA1c) testing devices. Queries have been raised with regard to the correct
and intended use of glucose and HbA1c testing devices and their fitness for purpose.
Many laboratories performing only waived tests however, do actually seek laboratory accreditation,
follow one of the many professionally based POCT operational guideline documents, perform
quality control, and participate in external quality assurance. Laboratories which perform moderate
and high complexity tests which require full accreditation often have all testing procedures
(including any waived tests which they may perform) included within the accreditation.
4.15 Occupational Health and Safety (OH&S)
Occupational Health and Safety (OH&S) can be considered a regulatory requirement and is
identified here for this reason. Patient and operator safety are important issues and are discussed as
a separate topic.
4.16 Documentation of procedures: regulatory requirements or good laboratory practice
Like any medical, scientific or pathology testing procedure, full documentation of all steps is
required. This is described in ISO 15189 and mandated for accreditation of pathology testing. It
applies to POCT in the same manner as it applies to central laboratory testing. This includes correct
patient identification, documented specimen collection and testing procedures, recording of quality
control (QC) and external quality assessment or proficiency testing (EQA or PT) results, and
recording of patient results in an indexed and easy retrievable format.
Documentation which describes all aspects of a given method or procedure is generally referred to a
standard operating procedure or SOP. Such documentation is required by ISO and NPAAC
standards, clinical practice guidelines and for accreditation.
Standard operating procedure documentation should include:

 Patient procedures and specimen collection
 Instrument operation and performance specification
 Quality procedures (QC and PT or EQA)
 QC monitoring, performance documentation and error-logging
 Trouble-shooting
 “Non-conformance” and “corrective action” reporting
ISO 15189 provides a clear statement of the requirement for good documentation. An extract from
this standard is provided below for information.
Management procedures for POCT, which includes the requirement for documentation, are
discussed in general Clinical Biochemistry texts and speciality POCT texts as described in the
course introduction (topic 1, references).
The agreed procedures or actual regulatory requirements by which the laboratory must operate, are
usually incorporated into a laboratory quality system and documented within a laboratory quality
manual.

The following reference provides additional useful information
 JH Nichols. Point-of-care (near-patient) testing. Chapter 21 in: LA Kaplan and AJ Pesce.

Clinical Chemistry, theory, analysis and correlation. 5th edition, 2010.
 CP Price and A St John. Point-of-care testing. Chapter 12 in: Tietz, edited by CA Burtis,
ER Ashwood and DE Burns. Fundamentals of Clinical Chemistry. 6th edition, 2008.
 CP Price, A St John and JM Hicks. Point-of-Care Testing. 2nd edition, AACC press, 2004.
See Google books at:
http://books.google.com.au/books?id=wcXrAgOVXbUC&printsec=frontcover#v=onepage
&q=&f=false
4.17 Mandatory reading for topic 4
Mandatory reading
 Tirimacco R. Evolution of point-of-care testing in Australia. Clin Biochem Rev

2010; 31: 75. Article available in topic library.
 E Jacobs etal. POCT: Standards, guidelines and governance. Chapter 16 in: Point-
of-care testing. Needs, opportunity and innovation. Edited by Price CP, St John A
and Kricka LJ. AACC Press 2010

Topic 5 – Statistics for POCT and Laboratory Applications
Learning objectives for topic 5
Topic 5A - Statistics for POCT and laboratory applications

 the nature of statistical information and numerical data
 the difference between a “population” and a “sample”
 various methods for the presentation of data and the use of tables, frequency distributions ,
histograms and frequency curves
 the terms “measures of central tendency” and “measures of dispersion” and provide
examples of each
 the key role of the “normal curve” in statistical procedures
 the meaning of probability and its relationship to the normal curve
 the use of linear regression procedures and their role in method comparison studies
 in general terms which statistical procedures (which statistical tools) are appropriate for
laboratory applications including:
o quality control
o external quality assessment or proficiency testing
o method comparison
o reference interval determination
o comparison of patient data
o critical difference calculations
o uncertainty of measurement applications
o fitness for purpose assessment
Topic 5B – Method comparison

 the various procedures which may be used to statistically compare analytical methods
 the use of linear regression procedures and difference plots (Bland and Altman) in method
comparison studies
 the importance of alternate regression procedures which do not rely on the independent
variable (usually the x-axis variable) being error free (that is, the importance of Deeming
regression and Passing and Bablok regression)
 the issues which need to be addressed in method comparison studies

Statistics, section headings
1. Statistics Overview
2. Difference, Change and Association
3. Presentation of data
4. Frequency curve(s)
5. Averages and Measures of Central Tendency
6. Dispersion
7. Variance and Standard Deviation
8. Degrees of Freedom
9. Calculation of Variance and Standard Deviation
10. Summation of Variances
11. Notation used in Statistics
12. Normal Curve or Gaussian Curve
13. Probability and Sampling
14. Student’s t-Test and t-Distribution
15. Standard Error of the Mean
16. Standard Error
17. Application of the t-Test
18. Fisher’s F-Test and F-Distribution
19. Confidence Intervals
20. Parametric and Non-Parametric Tests and Data
21. Association, Correlation and Regression
22. Simple Linear Regression
23. Deming, Passing and Bablok, Regression
24. Standard Error of the Estimate for a Regression Line
25. Correlation Coefficient (Pearson correlation coefficient)
26. Non-Linear Relationships
27. References

1. Statistics Overview
“There are three kinds of lies: lies, damned lies, and statistics” … This well-known saying is part of
a phrase attributed to Benjamin Disraeli (British Prime Minister 1874 to1880), and popularized by
the writer Mark Twain.
Statistics however, is concerned with scientific methods for collecting, organising, summarising,
presenting and analysing data. It is also the principal procedure for drawing valid conclusions and
making reasoned decisions based on this analysis. The misuse of statistics can of course lead to
incorrect conclusions. This misuse can be either deliberate or unintentional. It is thus important to
understand how to apply statistical procedures correctly and to recognise when they are being used
inappropriately.
The ubiquitous nature of statistics … the basis for many pathology quality procedures (BV,
biological variation; RR, reference ranges (reference intervals); QA, quality assurance (proficiency
testing); QC, quality control; MU, uncertainty of measurement).
BV
Statistics
QC MU RR
Error Theory
QA
There are essentially two sorts of statistics, descriptive statistics and inductive statistics.
 Descriptive statistics … includes the presentation of data in tables and diagrams, calculating
percentages, averages, measures of dispersion and correlation. These procedures are done in
order to display the salient features of the data and to reduce large data sets to manageable
proportions.
 Inductive statistics … involve methods for inferring the properties of a population on the basis
of known samples. These methods are based directly on probability theory.

2. Difference, Change and Association
Statistics are used to answer three types of question …

 Do two or more groups of observations differ from each other ?
 Have observations changed when repeated on a subsequent occasion ?
 Is there an underlying association between different observations (or measurements) which may
have been made at the same time ?
1. Is there a difference ? 2. Has there been a change ?
3. Is there an association ?
3. Presentation of data (data, plural; datum, singular)
In nearly all forms of publication (scientific, business, government, magazines, newspapers,

internet, etc), data of all sorts are presented in the form of tables, diagrams and pictures. This
approach provides a visual representation in which the key features can be more easily seen.

Types of data presentation include…
 Tables, data arranged in rows and columns (for example, Excel spreadsheet).
 Graphs and charts, many different styles and types (bar graph, pie chart, line graph, etc).
 Frequency distributions and their tabular and graphical representations.
In general, before raw data can be tabulated, interpreted, and presented, it must be classified.
Classification is the process of relating the separate items within a set of data to a defined category.
Numerical data can also be described as discrete or continuous variables …
 Discrete variables are measured in single units (that is, they are countable objects such as
houses, people, cars, etc). However, some statements made about discrete variables may appear
to contradict this; for example, “the average family has 2.4 children”. Even though it is difficult
to give meaning to a fraction of a child, when figures are averaged this type of result inevitably
occurs.
 Continuous variables are in units of measurement which can have any value whatsoever
between certain limits, or measurements which can be broken down into definite graduations.
Examples include, height and length in decimals (or fractions) of a meter, mass measurement in
decimals of a gram and volume measurement in decimals of a litre.
For large data sets, the items are best presented in a summarised form to more easily see their most
significant features. This is most often achieved by constructing a frequency table, which classifies
the set of observations according to the number of items falling within defined limits.
Definitions
 Population, the entire group of individuals, items or measurements, all possessing a common
specified characteristic.
 Sample, a set of individuals, items or measurements, selected from a population so that the
properties or parameters of the population may be estimated. A sample is a set of measurements
which constitute part of a population.
 Parameter, a numerical characteristic of a population such as its mean or standard deviation.
Such items may be referred to as a population parameter, or simply a parameter (see additional
information below). Parameters are rarely known and are usually estimated by statistics
computed on samples.
 Statistic, a quantity calculated from a sample such as its mean or standard deviation is a sample
statistic, or simply a statistic (see further information below).
 Frequency, the number of items or values falling within a given class (or defined range of
values), or having a specific value.
 Distribution, an arrangement (table, chart or graph) of values that shows the number of times
(frequency) a given score (or value) or group of scores (or values) occurs.
 Frequency distribution, the set of frequencies, together with their respective classes or specific
values.
 Cumulative frequency for a given class, is the sum of the frequencies for all (preceding) classes
up to and including the given class (assuming that the class with the smallest class boundaries is
written first).
 Class, a collection of values (data items) sharing a common characteristic.
 Class boundary, the boundaries for the given classes when data are grouped.
 Class interval, the interval between class boundaries for grouped data.

The following examples show the manner in which data are presented as a frequency histogram and
as a frequency table.
Frequency histogram (the number of observations (Y-axis) which fall within a stated range or
class (mean of range shown on X-axis).
Frequency table (for same data as shown in frequency histogram above)
Class interval Absolute frequency Cumulative frequency
0.0 to 9.9 1 1
10.0 to 19.9 3 4
20.0 to 29.9 8 12
30.0 to 39.9 18 30
40.0 to 49.9 24 54
50.0 to 59.9 22 76
60.0 to 69.9 15 91
70.0 to 79.9 8 99
80.0 to 89.9 0 99
90.0 to 99.9 1 100

4. Frequency curve(s)
A frequency polygon is a “curve” produced by joining the mid-points at the tops of the histogram
blocks (or bars). Such a polygon is usually drawn with straight lines, and the resulting diagram is a
“many-sided” figure or polygon (see diagram below).
Frequency
Class interval
Histograms and frequency polygons enable the properties of distributions to be examined, and
various forms of distributions can be compared in a general manner. However, neither the
histogram or frequency polygon gives the most accurate view of a frequency distribution. The
histogram implies that the frequencies are the same throughout the class interval, when this may not
actually be the case. A more accurate view of the distribution would be obtained if there were more
observations and smaller class intervals.
As the number of observations (and number of items within a given class or frequency) and the
number of class intervals are increased, the histogram and frequency polygon move more closely
towards a smooth curve.
Frequency
As the number of observations increase, the number of classes can also be

increased. When “many” observations are available, the frequency histogram and
frequency polygon move towards a smooth curve (frequency curve).

5. Averages and Measures of Central Tendency
In order to reduce a mass of data to a simpler form, a frequency table and/or frequency histogram is
often used. It is also useful to further simplify the presentation with measures which describe the
scatter of data about a central point and whether there are a number of extreme cases. Any measure
which indicates a “central feature” of a distribution is called a measure of central tendency. A
numerical value indicating the amount of scatter about a central point is called a measure of
dispersion.
An average is a measure of central tendency and also a measure of location. The three most
commonly used averages are the arithmetic average, the median and the mode.
 The arithmetic mean (mean, or the “average”), is defined as the total value of all items divided
by the number of items.
 The median, is defined as the middle value in a distribution which is set out in value order. That
is, there are an equal number of items above and below the median value.
 The mode, is defined as the value which occurs most frequently.
 Quartile, the median divides an ordered distribution into half; in a similar manner the quartiles
divide the distribution into four equal parts with an equal number of observations in each part
(each quarter of the distribution). There are thus three quartile boundaries, the lower quartile
(Q1), the middle quartile (Q2) which is the same as the median, the upper quartile (Q3). The
method for calculating quartiles is similar to the method for calculating the median.
 Additional comments regarding mean, median and mode, see section below … Parametric and
Non-Parametric Tests and Data.
The most widely used measure of central tendency is the arithmetic mean. It is easy to compute,
easy to define and takes all measurements into consideration. One of the principal advantages of
the mean is its reliability in sampling. The mean of a sample from a given population will, in
general, show close agreement with the mean of another sample taken from the same population
(and closer agreement than two medians).
6. Dispersion
Even though the mean is a satisfactory measure of central tendency, it does not give any indication
as to the spread or distribution of values or measurements. Measures which show the spread or
distribution of values are called measures of dispersion. Measures of dispersion include …
 Range, the highest value in a distribution minus the lowest value.
 Interquartile range, the difference between the upper quartile and the lower quartile (see above
for definition of quartile).
 Deviation and mean deviation, (to be discussed further below).
 Variance, (to be discussed further below).
 Standard Deviation, (to be discussed further below).
 Coefficient of Variation, (to be discussed further below).

7. Variance and Standard Deviation
The procedure for calculating variance and standard deviation from raw data forms the basis for
many statistical calculations. It can easily be done using spreadsheets such as Excel. For “manual”
calculations, additional columns can be added which control the flow of any calculation(s) done
sequentially. In addition to any sequential “manual” calculation, there are “automatic” functions
which provide many of the basic statistical calculations directly on the data provided. The Excel
“help” feature describes how such automatic functions can be utilised.
However, understanding the basic sequence of calculations provides a better understanding of the
mathematical processes involved.
The standard deviation is the “standard” measure of dispersion because it is very useful both
practically and mathematically. The standard deviation can be used as a measure of dispersion in
all symmetrical unimodal distributions. It may also be used in some distributions which are
moderately skewed (where data can be “transformed” to give a more symmetrical distribution).
The standard deviation shows the dispersion of values around the mean, the greater the dispersion,
the larger the standard deviation.
8. Degrees of Freedom
Degrees of freedom is defined as the number of ways in which a group of numbers can vary
independently. This is not an easy concept to explain, understand intuitively, or show
mathematically. It is best explained by way of an example … if we have 20 independently
measured serum sodium results there are 20 degrees of freedom because no single measurement
affected the measurement of any other. However, if we calculate the mean of the 20 results the
series now has 20-1 (or n-1) degrees of freedom, that is 19 degrees of freedom, because the
twentieth value in the series cannot be changed (whatever value it is) without altering the mean.
Alternately, by knowing the mean plus 19 of the measured sodium values, the twentieth value
becomes fixed and can be calculated from the information in-hand, it is no longer independent.
In statistics and probability, the degrees of freedom are often used when deriving sample statistics
as approximations for population parameters (see definitions above).
In statistical analysis, calculations on data obtained from measurements derived from samples
(statistical definition) is used as “the best estimate” or as an approximation to a corresponding
population (statistical definition). In this situation, many of the calculations (such as standard
deviation and variance) use degrees of freedom (n-1) instead of the actual number of observations
(n).
9. Calculation of Variance and Standard Deviation
The standard deviation is calculated by taking the deviations (the differences) of each value from
the mean, and then calculating an average deviation for the number of values involved. However, if
the actual deviations from the mean are added together, the result will always be zero as the positive
deviations and the negative deviations will cancel. To obtain a measure of dispersion around the
mean, the negative signs must be “removed”.

There are two ways of doing this …
 By ignoring the signs and taking the absolute values of the deviations. This process will give
the mean deviation which is the arithmetic average of the absolute differences of each value
from the mean. The mean deviation itself is seldom used as a measure of dispersion, but forms
an intermediate part of the calculation of variance, standard deviation and other statistics.
 By squaring. A common method of converting positive and negative numbers into quantities
which are all positive (or in obtaining the absolute value of a quantity arithmetically), is by the
process of squaring and then taking the square root (see also below sum of squares).
The average of the squared deviations from the mean is called the variance (V). It is often used
directly as a measure of dispersion. An alternative to the variance is the standard deviation. The
standard deviation (S or SD) is the square root of the variance.
An advantage of the standard deviation is that it is expressed in the same units as the original
values. It also has some very convenient properties in relation to continuous symmetrical
distributions such as the normal curve or Gaussian curve (to be discussed below). The standard
deviation is a much used statistic / parameter in calculations and in the presentation of data.
V = ∑ (xi – mean)2 / (n-1)
S=V½
or
S = ﴾ ∑ (xi – mean)2 / (n-1) ﴿ ½
CV = S / mean
Where:
 V = (sample) variance
 ∑ = symbol used to indicate “the sum of”
 xi = an individual value, measurement or observation
 n = the number of individual values, measurements or observations
 S = (sample) standard deviation (or sometimes written as SD)
 CV= coefficient of variation
 CV%= coefficient of variation expressed as a percentage (CV * 100)
 Mean … the arithmetic mean … mean = ∑(xi) / n
 n-1 degrees of freedom … used instead of “n” as described above for sample statistic
as compared to population parameter.
 Sum of squares … a concept which is often used in statistics and is the term used for the
sum of the squared deviations. The “distance” of any value in a collection of data to the
mean is the deviation, usually written as (xi – mean) … if all such deviations are squared
and then summed (∑ (xi – mean)2), this term is referred to as the sum of squares.

10. Summation of Variances
Statisticians often prefer to work with variances instead of standard deviations or standard errors, as
variances are additive. This is particularly true for the average of the squared deviations from the
mean (the variance (V), as described above) and is often used directly as a measure of dispersion.
The advantage of the variance is that independent sources of variability (each with its own variance
(V1, V2, etc)) which contribute to the total variation can be described the equation:
Vtotal = V1 + V2 + …etc
or
Vtotal = S total = S12 + S22 + …
2
The importance of this procedure is that only variances are additive (V or S2) not standard
deviations or standard errors. Similarly coefficients of variation (CV’s) are not additive unless
they are first squared and their mean values are the same or similar (which is often the case in
medical laboratory applications).
This procedure for the summation of variances has several useful applications in laboratory
medicine. These include:
 A component in the procedure for calculating t for a t-test (as described in section 17 below)
 Examining the effect of intra-individual, population variation and analytical imprecision on
laboratory result data. For calculation of the variability of individual patient results;
Stotal = √(S2intra-individual +S2analytical)
 The assessment of critical difference or reference change values (RCF) for interpreting a change
in test results (refer to separate topic notes or text books of Clinical Biochemistry).
11. Notation used in Statistics
The notation used in statistics (and mathematics) is often confusing if not misleading. Many books
on statistics use two notational forms, one for describing populations and population parameters,
and one for describing approximations derived from samples or sample statistics. Even the terms
population and sample have multiple meanings depending on the context in which they are used.
Many books (often those describing a more simplified approach to statistics) do not clearly
distinguish calculations made on samples (particularly with low sample numbers), from calculations
relating to populations (or samples with large sample numbers). From a practical perspective in
laboratory medicine, such differences may not actually be all that significant.
As the form of some calculations may appear different (and inconsistent) from one text to another,
it is important to fully appreciate the differences in notation used to describe population parameters
and sample statistics.
As described earlier, degrees of freedom is generally considered more appropriate for calculations
relating to samples than the actual number of sample values (or “n”). For example, the formula for
standard deviation can be found with either “n” or “n-1” in the denominator without any clear
explanation as to the difference. Both formulae are correct … the one with “n” only applies to
calculations based on populations, the one containing “n-1” relates to calculations based on samples
which are being used as an approximation to the corresponding population. However, if “n” is large
(with n greater than 100), then the use of “n-1” or “n” give essentially the same result.

However, many texts do correctly distinguish the difference between population and sample data by
the symbols used in calculations. For example:
Measure Sample Population
Individual value x or xi X or Xi
Number of values n N
Arithmetic mean x (with a – above) μ or X (with a – above)

Standard deviation s or S or SD σ
Variance v or V or s2 V or σ2
12. Normal Curve or Gaussian Curve
The normal or Gaussian curve is a bell-shaped symmetrical distribution. It is probably the most
important continuous distribution, and one on which much statistical theory is based. Its shape is
defined by a (complex) mathematical equation which includes the mean and standard deviation. In
general, data which conform to a normal distribution (or where a normal distribution can be used as
a good approximation to the distribution of the data) include (for example)…
 Distributions (and data) which arise in sampling theory
 Data which is obtained from intelligence tests
 Data from examination results
 Biological data and measurements relating to biological variables such as height and weight
 Biological reference interval calculations
 Error theory
 Repeat (replicate) measurements of a given property on the same substance
 Data from Quality Control and Quality Assurance.
The relationship between the mean and standard deviation for the normal distribution (curve),
provides a valuable procedure for summarizing and comparing biological data. It also provides a
tool for predicting possible outcomes relating to “large” groups, from data obtained on relatively
“small” samples (statistical definition). The fact that 68% of the total area under a normal curve lies
between one standard deviation either side of the mean makes it possible to say that two items out
of three (approximately) lie within one standard deviation of the mean. Thus if one item is selected
at random from a distribution, there would be a two in three chance of it being one of these items
lying within one standard deviation of the mean. This would be correct two times out of three, or it
would be likely with a 68% “level of confidence”.
Similarly, since approximately 95% of the area under the curve lies within two standard deviations of
the mean, one item selected at random would lie within these limits 95 times out of 100 (or 19 times
out of 20). These statements are actually statements on the probability of an event occurring. In this
manner, data collected from statistical samples can be used to make predictions about larger groups or
populations (see probability, below).

Frequency
Normal or Gaussian curve showing mean (μ), standard deviation (σ) and areas under curve.
13. Probability and Sampling
Statistics and probability are so fundamentally interrelated it is difficult to discuss statistics without
discussing probability. An understanding of probability theory makes it possible to interpret
statistical results, since many statistical procedures involve conclusions based on samples which are
always affected by random variation. It is by the use of probability that we can express numerically
the inevitable uncertainties in scientific measurements, biological variation, etc.
Providing always that experimental data do actually follow (or at least approximate) a normal
distribution, calculations based on mean, standard deviation, and other statistics can be used to
determine the probability of specified events occurring. If an item is selected at random from a
population, there is a good chance that this item will actually represent the population from which it
is drawn. Describing a population from observing a small sample requires …
 The sample to be taken at random
 It is understood that there is always some degree of error in sampling.
A sample will not always be an exact replica of the population because chance plays an important
part in random selection. It is likely that the sample mean and sample standard deviation will differ
slightly from the population mean and population standard deviation (see also degrees of freedom
above). The larger the sample, the less chance of selecting a group of items that is unrepresentative
of the population.
Even though probability has both common and mathematical definitions, in practice it is often taken
as the proportion of times which an event occurs out of a large number of trials ( or the proportion
of the items with the largest frequency to the total number of items). Stated mathematically …
 If an event can happen in “s” (for success) ways and fail to happen in “f” (for fail) ways, and in
each of these “s” and “f” ways is equally likely to occur, then
 the probability (“p”) of success (the event happening) in a single trial is: p = s / (s+f)
 the probability (“q”) of failure (the event not happening) is: q = f / (s+f)
 where (s+f) is the total number of ways the event can occur.

14. Student’s t-Test and t-Distribution
As there is always some degree of error in sampling, it is useful to have a measure which provides
an indication of the extent to which a sample mean and its associated standard deviation deviate
from their corresponding population parameters. For further comments regarding the variation of
the standard deviation, see below under F-distribution and F-test.
Since the error of the sample mean derived from a set of values (measurements or observations) is
smaller than that for a single value, the more times a measurement is made the more certain is its
true value. Sample means which are derived from a large number of values (with 100 or more
values) taken from a normal distribution, are themselves also normally distributed. For smaller
samples, WS Gossett (working under the pen-name of “Student”) showed that a “flatter”
distribution was required to describe the distribution of sample means. In a true normal
distribution, the proportion of values falling within multiples of the standard deviation from the
mean can be accurately predicted. In a similar manner, the probability of the true mean falling
within multiples of the standard error of the mean (see below) within Student’s “t” distribution can
be calculated.
The t-distribution has a similar shape to the normal distribution but is “flatter” than the normal
distribution and becomes “flatter” as the number of values which contribute to the sample mean
decrease. The exact shape of the t-distribution curve is determined by the degrees of freedom. As
the degrees of freedom (the number of values) increase, the t-distribution approaches the normal
distribution (see below). Probabilities are determined from the t distribution by calculating the area
under the curve for a given number of degrees of freedom. In practice, this information is obtained
from pre-calculated statistical tables.
The application of the t-distribution to determine the probability of a given (calculated) t-value is
called a t-test. It is used to ascertain whether the means of two samples are likely to differ
significantly (two sample or paired t-test) or if the mean of a sample differs from a given number
(one sample or unpaired t-test). Refer also to the discussion regarding the application of the t-test
in section 17 below. Most books on statistics and Clinical Biochemistry laboratory texts describe
the appropriate application of paired and unpaired t-tests. A good summary of the t-test,
t-distribution and shape of the distribution curves, is given by the introduction to Student’s
t-distribution on Wikipedia. ( http://en.wikipedia.org/wiki/Student's_t-distribution ).
Comparison of t distribution (green and red curves) with the normal distribution.
(blue curve = normal curve; red curve = t distribution curve with 10 degrees of freedom;
green curve = t distribution curve with 5 degrees of freedom).
See chart on next page.

Refer also on the discussion regarding the application of the t-test in section 17, below.
15. Standard Error of the Mean
Multiple sample means which are derived from a large number of values (see t distribution above)
obtained from the same general population, are distributed normally about the population mean. In
a similar manner, sample means which are derived from a lower number of values are distributed
about the true (population) mean according to the t distribution. The random variation in this group
of sample means is described by the standard error or the standard error of the mean.
The “average” of all sample means is equal to (or very close to, or the best estimate of) the true
population mean.
The standard error of the sample mean is often written as SEM or SE, and is calculated from the
standard deviation of the sample (S, SD) and the number of values (n).
SEM = standard deviation of sample / √ (sample size)

SEM = S / √ n
Some comments with regard to standard error of the mean and its interpretation may be useful …
 As outlined previously, 95% of the items within a normal distribution lie within two standard
deviations of the distribution mean
 The distribution of sample means derived from large samples (> 100 values) is normal, with an
“average mean” equal to the population mean
 The distribution of sample means derived from smaller samples follows the Student t
distribution
 As the normal distribution forms the limiting distribution of the t distribution, the t distribution
can be used whenever sample means are being compared

 95% of all sample means must lie within two standard errors (or two standard errors of the
mean) from the “average mean” (or true population mean)
 The accuracy of the estimate is independent of the population size … but does depend on the
sample size and the variability of the characteristic(s) being measured.
16. Standard Error
In a more general sense, the term standard error of any statistic applies to the standard deviation of
the sampling distribution of that statistic. In addition to the standard error of the mean (as outlined
above), there are standard errors for standard deviation, regression lines (see below), etc.
17. Application of the t-Test
The t-distribution and the application of the t-test may be used to address three types of problem:
1. Given the mean and standard deviation of a sample, what is the probability that the
population mean (which is unknown) lies within a certain range of the sample mean ?
This can be considered as hypothesis testing. That is, testing between an experimental
mean and a perceived or stated value.
2. Given the means and standard deviations of two samples, how significant is the
difference between the two means ?
Often referred to as the unpaired t-test. It is used when two independent samples are
being compared. For example; comparison of the means of patients’ glucose values
from two wards measured by a POCT device.
3. If paired observations (measurements) are made on two sample sets (or in succession on
the same set of samples), what is the significance of the difference between the means of
the two sets of observations (measurements) ?
Often referred to as the paired t-test. It is used when paired samples from a series of
observations (measurements) are being compared. For example when comparing two
analytical methods; blood specimens from patients’ are split and analysed for glucose by
a POCT device and by a laboratory procedure which also measures whole blood glucose.
In each of the above situations, the problem is essentially the same. What is the significance of the
difference between two means ? The approach to answering this question is to calculate how many
times the difference (the difference between the means) can be divided by its standard error. With
“large samples”, 1.96 times its standard error has a probability of 5% (or less) based on the normal
or Gaussian distribution. With “small” samples these multiples of standard error become larger and
the t-distribution should be used to assess probability.
Examples for unpaired and paired t-tests will be given. For further details and for information with
regard to hypothesis testing, the supplementary reading (section 25, references) should be
consulted.
17.1.1 Comparison of two means – Unpaired t-Test
The requirement to compare the means obtained from two groups of “similar” but independent
samples (or measurements) is a common situation in laboratory medicine and in method
comparison studies. This comparison may also be restated as “is the mean of one set of results
significantly different from the mean of another set ?”. One of the issues with regard to this
comparison is that the value of the mean is influenced by the number of observations (results) in
each set of measurements. That is, each set of observations is characterized by its own mean, its

own standard deviation and its own sample size (with the mean of each set having its own standard
error of S / √ n ).
In order to ascertain if there is a significant difference between the two means, another estimate of
variability is required. This is an estimate of the variability in the actual difference (between the
two means), as this has itself been calculated from two statistics which have inherent variability as
described by their respective standard errors. To obtain the required variability information, the
procedure described in section 10 for the summation of variances is applied. Individual variances
(V, S2, standard deviation squared) which contribute to overall variability, may be summed to
obtain the total variance. In the case of means and their respective standard errors, the standard
error may be squared and treated mathematically in the same manner as a standard deviation in
order to allow summation as a variance.
In evaluating the difference between two individual samples (x1 - x2) or between two means
(m1 – m2), it is usual to consider the problem in the form of “the probability that no difference
exists”. In this situation, the difference would ideally be zero with a variability determined by the
number of observations in each of the two data sets. That is, the difference would be zero +/- its
own standard error (or standard deviation) with its own probability distribution.
17.1.2 Difference for “large” samples
When the number of observations is “large” (certainly greater than 100, even though some
authorities suggest that 30 may be a reasonable cut-off value), the question “is there a difference
between two means” may be answered by the following procedure:
 Obtain the difference between the two means by subtraction, that is D=(m1 – m2)
 Calculate the variance of “D” by summation of the individual variance components (as outlined
in section 10)
 The variability of “D” will depend on the inherent variability of the both m1 and m2, the larger
the variability in those items which contribute to the value of “D”, the larger the inherent
variability of “D” itself
 Thus, VD = Vm1 + Vm2 = (SEM1)2 + (SEM2)2
 Thus, SD = √VD (where SD is the standard error for the difference “D”)
 If the difference “D” is more than 1.96 (or 2, as it is often specified) times its standard error
“SD”, it is likely to occur at a frequency of only 1 in 20 or less by chance alone. That is, there is
only a 5% probability that the two means originate from the same population (or a 95%
probability that the two means do not originate from the same population)
 Depending on local usage or the requirement for precise definition of the statistics involved;
o the use of 1.96 standard deviations provides a probability of 95.0%
o the use of 2.0 standard deviations provides a probability of 95.4%
 The more direct mathematical (statistical) approach is to divide the difference “D” by its
standard error “SD” and compare the value obtained to 1.96 (or 2.0, as the case may be). That
is, D / SD should be less than 1.96 for the two means to be considered from the same
population. This procedure for determining probability assumes that the two means and their
difference follow normal (Gaussian) probability distribution characteristics as described in item
12 above.
17.1.3 Difference for “small” samples
When samples are “small” a different approach is required. The definition of a “large” or “small”
sample is somewhat controversial. Sample numbers less than 30 are usually considered “small”,
while samples greater than 100 are usually considered “large”. So what about the 30 to 100 sample
size ?. It depends essentially on the purpose for the comparison and the degree of “correctness”
required. If in doubt, always use the “small” sample number method. As the number of
observations increase and the data become more robust, the “small” sample t-test method (to be
described below) merges into the “large” sample method described above.
With “small” samples, more chance variation must be allowed. This additional chance variability is
the basis for the “t” distribution as described in section 14 above. Sample means, which are derived
from a large number of individual values (with 100 or more values ?) which are themselves taken
from a normal distribution, are also normally distributed. For smaller samples, WS Gossett
(working under the pen-name of “Student”) showed that a “flatter” distribution was required to
describe the distribution of sample means. In a true normal distribution, the proportion of values
falling within multiples of the standard deviation from the mean can be accurately predicted. In a
similar manner, the probability of the true mean falling within multiples of the standard error of the
mean (see section 15 above) for the Student’s “t” distribution can also be calculated.
The t-distribution has a similar shape to the normal distribution but is “flatter” than the normal
distribution and becomes “flatter” as the number of values which contribute to the sample mean
decrease. The exact shape of the t-distribution curve is determined by the degrees of freedom
(number of observations). As the degrees of freedom increase, the t-distribution approaches the
normal distribution (see above). That is, the dispersion of the t-distribution varies according to the
size of the sample. Probabilities are determined from the t-distribution by calculating the area under
the curve for a given number of degrees of freedom. In practice, this information is obtained from
pre-calculated statistical tables.
The procedure for “small” samples does not differ greatly to that for “large” samples. To answer
the same question as before, “is the mean of one set of results significantly different from the mean
of another set ?”, the difference between the two means and the standard error of the difference is
calculated. The following procedure can be applied:
 Obtain the difference between the two means by subtraction, that is D=(m1 – m2)
 Calculate the variance of “D” by summation of the individual variance components (as outlined
in section 10)
 The variability of “D” will depend on the inherent variability of the both m1 and m2, the larger
the variability in those items which contribute to the value of “D”, the larger the inherent
variability of “D” itself:
Thus, VD = Vm1 + Vm2 = (SEM1)2 + (SEM2)2
Thus, SD = √VD (where SD is the standard error for the difference “D”)
 So far, this is the same as for the “large” sample comparison as described above
 In a similar manner to that used for the “large” sample procedure, the number of standard errors
“SD” represented by the difference “D” are calculated. This number is the “t” value, and is
calculated as t = D / SD.
 When the number of samples contributing to either m1 or m2 is “low” (less than 100), the
probability value is obtained from “t” tables with the degrees of freedom calculated by the
rather complex formula:
Df = (Vm1 + Vm2)2 / [(Vm1)2 /n1-1)+ (Vm2)2 / n2-1)]
Where: Df = degrees of freedom for t-test

Vm1 = variance of mean 1 (m1)
Vm2 = variance of mean 2 (m2)
n1 = number of samples comprising m1
n2 = number of samples comprising m2
 The degrees of freedom (Df) obtained from this formula is rarely a whole number. The number
obtained should be rounded down to the nearest integer.

 If the calculated t value does not exceed the critical t value (as provided in the t table, given by
the statistics calculator or “Excel” spreadsheet, etc) a significant difference between the two
distributions (means) does not exist (at the probability level chosen).
When it can be shown that the standard errors of the two sample means are the same (or very
similar), a simplified approach may be used for the comparison (as a number of the terms in the
various equations cancel). However, it is first necessary to demonstrate that the standard errors of
each of the means are actually the same (or very similar) by application of the F-test or variance
ratio test as described below. The above procedures may be applied in all situations, irrespective of
the relative magnitudes of the individual standard errors (variances).
17.2 Comparison of two means – Paired t-Test
When individual observations (measurements) are made on a series of samples in order to

determine if the two sets of observations (measurements) differ, a paired t-test is the procedure of
choice. For example; when a series of patient samples (each with different analyte values) are
measured by two different methods, using an unpaired t-test to compare the two method means is
inefficient. In this situation there are two principal types of variation:
1. Variation in results due to the specimens having different analyte values
2. Variation in results due to the two different methods
In this example, the variation in results due to the specimens having different analyte values is
much greater than the difference due to the two methods. The paired t-test provides a procedure to
exclude the variability due the different analyte values.
The procedure for paired t-test is:

 For each pair of observations or results (x1 method A (x1A) and x1 method B (x1B)), the values
from one set of measurements are subtracted from the other to give a difference “d”.
That is; d = (x1A – x1B)
 If there are “n” pairs of observations (“n” pairs of results), there will be “n” values of “d”
 Calculate the mean (md) and standard deviation (Sd) of the “d” values
 If there is no difference between the mean values for the two sets of results, then the mean value
of “d” will be zero (that is; md will be zero, or a small number near zero). To test whether there
is a statistical difference between md and zero, a t value is calculated assuming that the values
for md are normally distributed with a standard error of Sd / √n
 Calculate: t = md / (Sd / √n)
Where: md = Σd / n
Sd = √[ (Σ(d – md)2 / (n – 1)]
 In this case, the degrees of freedom is one less than the number of paired observations.
That is; Df = n-1. This value of Df should also be used in association with the calculated t value
for determining probability.
 If the calculated t value does not exceed the critical t value (as provided in the t table, given by
the statistics calculator or “Excel” spreadsheet, etc) a significant difference between the two
distributions (in this case, the results of measurements by two methods) does not exist (at the
probability level chosen).

18. Fisher’s F-Test and F-Distribution
In a similar manner for requiring a test to determine if the means of two samples differ, it is also
necessary to determine whether observed variances (or standard deviations) vary from one sample
to another.
As described above, the t-test is used to compare the means of two groups of observations in order
to test whether there is a significant difference between the two group means. Two assumptions are
made with the t-test … that the groups are normally distributed and that there is no significant
difference between the group variances (standard deviations). It is not possible to tell simply be
inspecting the data how different the variances in the two groups must be before the t-test cannot be
used. However, the hypothesis that there is no significant difference between the two variances can
be ascertained from the F-test.
An F-test (or variance ratio test) should be applied before a t-test is used, to check if the equal
variance assumption for applying a t-test is valid. The F-test statistic is calculated by dividing the
larger variance by the smaller variance …
F = (V1) / (V2) = (S1)2 / (S2)2
The calculated “F-value” is compared with a critical F-value obtained from an F-table by using the
number of degrees of freedom from each group at a specified probability level (usually 95% or p =
0.05). If the calculated F-value is less than the table value (critical value), the probability is such
that the two variance would be observed in samples taken from the same population.
The F-test is an integral part of the more general analysis of variance (ANOV). The test is named
after RA Fisher, a very distinguished statistician. It is derived from another important probability
distribution, the F-distribution. A statistical textbook should be consulted for further discussion
regarding the F-distribution and analysis of variance procedures.
19. Confidence Intervals
Statistical procedures (such as the mean and standard deviation) are important tools which are used
to assess:
 Quality Control (QC)
 Quality Assurance (QA) or proficiency testing
 The determination of biological variation
 The determination of reference intervals (reference ranges)
 Comparison of analytical methods.
When the data is normally distributed (follows a normal distribution), the area under the curve from
+1S to -1S represents 68.3% of the values, from +2s to -2S represents 95.4% of the values, and
from +3S to -3S represents 99.7% of the values. Intervals of this type which contain a stated
percentage of the data are called confidence intervals.
For samples of QC material, calculations using the mean and standard deviation form the basis of
statistical quality control and are assessed in accordance with specific acceptance or rejection rules
for the analytes concerned. In laboratory quality control procedures, the 95.4% (or 95% as it is
usually abbreviated) confidence interval is most often used. This indicates a probability that 95% of
the values will fall within +2S or -2S from the mean.

20. Parametric and Non-Parametric Tests and Data
Samples that have come from (or are likely to have come from) a normal distribution (or a similar
symmetrical distribution) are described as parametric. Statistical procedures applied to data which
are known (or are assumed) to be normally distributed are called parametric procedures or
parametric statistics. Many of the procedures described above (such as standard deviation and t-
test) rely on the data being (approximately) normally distributed.
For a normal distribution, the three measures of central tendency (the mean, the median and the
mode) will all be the same or very similar. When these three measures differ significantly, the
corresponding data are not normally distributed and most likely are not drawn from a normal
population.
Frequency distributions showing the relative positions for three measures of central tendency
(mean, median and mode). The x-axis, values of variable: y-axis, frequency. Graph (a),
symmetrical (normal) distribution (mean, median and mode all equal). Graph (b) bimodal
distribution with two equal modal frequencies. Graph (c), asymmetric positively skewed
distribution. Graph (d), asymmetric negatively skewed distribution. In an asymmetric unimodal
distribution, the mean lies approximately one-third of the distance between the mean and the mode.
Skewness indicates the degree of asymmetry of a distribution. A skewness statistic can be

calculated which gives a numeric indication of the degree of asymmetry. For a normal distribution
the value of skewness equals zero A “right hand” or positive skew (graph (c) above) gives a skew
value greater than zero, while a “left hand” or negative skew (graph (d) above) gives a negative
skew value (less than zero).

Statistical analyses which do not depend on knowledge of the distribution or on assumptions as to
the shape of the population are called non-parametric or distribution-free methods or procedures.
Confirmation that a distribution (experimental data, observations, etc) does actually conform to a
normal distribution can be obtained mathematically. It is usually prudent to perform such a check,
but if numbers are small (say less than 30) and the histogram plot looks reasonably symmetrical
(and / or a plot of cumulative frequency against data values on “probability” graph paper gives a
straight line), standard parametric statistics may give approximate information which is still useful.
In general, statistical procedures may be classified as parametric or non-parametric depending on

the shape (or assumed shape) of the distribution. For normally distributed data (parametric data),
parametric procedures should be used. If data is not normally distributed, or the form of the
distribution is not known, non-parametric procedures are the most appropriate methods to use.
21. Association, Correlation and Regression
The previous discussions (above) related to populations with only one variable. This section
relating to correlation and regression considers populations made up from pairs of measurements
(or pairs of variables). If, for every measurement of a variable “x” we know a corresponding value
of a second variable “y”, the resulting set of pairs is called a bivariate population.
The main questions which need to be considered are …

 Is there a relationship between the two variables x and y ?
 If such a relationship exists, how can it be expressed in the form of an equation ?
A graphical representation of a set of paired values in a two dimensional coordinate system for x
and y is called a scatter graph. A scatter graph or scatter diagram, provides a useful method for
deciding if there is any association between the two variables. The basic purpose is to see if there
is any pattern among the points. If there appears to be an association between the points, a line of
best fit can be drawn with the same number of points or coordinates on each side of the line. From
the pattern of values and the line of best fit, it is often possible to …
 Determine if any association exists,
 Determine if there is a “strong” or “weak” association,
 Determine the direction of the relationship,
 Predict the value of one variable from the other.
For a wide variety of laboratory data, it is useful to determine the relationship between two
variables, x and y. It is usual to consider x as a “fixed” or independent variable with y as “not
fixed” or as a dependent variable.
Definitions
 Association, the tendency for two or more events to occur together.
 Correlation, the interdependence between two or more random variables. If two variables are
correlated, then changes in the value of one are associated with changes in the value of the
other.
 Regression (regression analysis), refers to techniques for the modeling and analysis of
numerical data which consist of values for a dependent variable and one or more independent
variables.

 Regression line, the line of best fit drawn through a set of data points on a graph. The “y” axis
(and y variable) is generally the dependent variable, the “x” axis (and x variable) is the
independent variable.
For many applications within Clinical Biochemistry, the relationship between two variables is linear
(that is, based on a straight line). Linear regression analysis is the mathematical procedure used to
define the linear relationship between the x variable and the y variable according to the straight line
equation y = mx + c (where “m” is the slope of the line and “c” is the intercept on the y-axis).
When a relationship exists between two variables, these variables are considered to have a
correlation with each other. The “product moment” coefficient is a measure of linear correlation
devised by Karl Pearson and often called Pearson’s correlation coefficient. The correlation
coefficient attempts to show the strength of the relationship between the two variables. Not only is
this an attempt to show whether a correlation exists (which is also the aim of a scatter graph), but to
show how closely it exists. A correlation coefficient is usually calculated as part of any regression
calculations.
There are essentially three linear regression procedures used for calculating a straight line
relationship between paired variables …
 Simple linear regression (simple least-squares linear regression)
 Deeming regression
 Passing and Bablok regression
22. Simple Linear Regression
Simple linear regression will provide a correct estimate of y for a given x, provided that the
following assumptions apply …
 The “x” variables are considered fixed with zero error in their measurement (or if any random
error in their measurement it is considered negligible),
 For every value of “x”, there is a normal distribution of “y” values,
 The standard deviation of “y” values is similar at all values of “x” (that is, the variance around
the line is independent of “x”),
 The estimated straight line is not horizontal (that is, slope “m” is not zero).
The procedure for calculating the line of best fit is by least squares regression. This procedure
minimises the sum of squares of the distance between the observed data points and the theoretical
line. For simple least squares linear regression, the sum of squares is minimized in the vertical
direction only (y values and those predicted by the line), thus not allowing for any variability of the
x values.
In Clinical Biochemistry, a common application of regression and correlation is the comparison of

two methods. In method comparison studies, paired measurements are made using the two
procedures on the same clinical samples. It is usual to consider the x values as those from the
“existing” method, the “reference” method or the method which is being used as the “reference”
method, with the y values those from the test method. This type of comparison also highlights a
disadvantage in using the simple linear regression approach. With method comparison, even with
the best method available as the “reference method”, there is still an error component which
invalidates the first assumption for the use of simple linear regression (as outlined above). For this
reason, the alternate procedures of Deeming or Passing and Bablock are often used. Both of these
do not incorporate a requirement that the x values are error free.

23. Deming Regression, Passing and Bablok Regression
When applying simple linear regression to a set of paired variables, different regression equations
are obtained depending which group of data is taken as x and which is taken as y. The regression
line of x on y is not the same as the regression line of y on x unless all points of the given sample
actually lie on a straight line. This is very unlikely to occur in the usual type of method comparison
study. The procedures of Deming, and Passing and Bablok, both only give one regression line
irrespective of which data set is used for x and which used for y. The Deming method, which
allows both variables to be measured with error, is relatively popular and can be applied easily
using a computer spreadsheet. The method of Passing and Bablok is less widely used as the
calculations are somewhat more involved.
The procedure for calculating the line of best fit by Deming regression, minimises the sum of
squares of the distance between the observed data points and the theoretical line in both the vertical
and horizontal directions simultaneously (both x and y values and those predicted by the line). This
procedure allows for variability in both the x and y values. It is thus a more appropriate procedure
for method comparison studies. (These procedures will be discussed further in the lecture on
statistics).
Simple linear regression of x on y and y on x showing two different regression lines.
Unlike simple linear regression, Deming regression always results in one line, whether x or y is
used as the independent variable.

When a linear relationship can be shown to exist between a “test” and a “reference” method, the
slope and intercept of the line can provide important estimates of the proportional and constant
errors between the two methods.
24. Standard Error of the Estimate for a Regression Line
In a similar manner to all statistics derived from values which contain measurement error, the slope
and intercept of a regression line have an inherent variability. As indicated above, the general term
for variability in a calculated statistic is the standard error. The variability of the data points about
the regression line is given by the standard deviation of the difference between the actual
(observed) value and these predicted by the regression equation. This variability is called the
standard error of regression (or the standard error of the estimate).
The standard error of the regression can be interpreted as the standard deviation of the mean value
expected for y for a given value of x (which is close to mean x). It is a measure of the scatter of
points about the calculated regression line.
25. Correlation Coefficient (Pearson correlation coefficient)
Another measure (in addition to the standard error of the estimate described above) which attempts
to show the closeness of association between two variables is the correlation coefficient (r). The
product moment coefficient of linear correlation devised by Karl Pearson (Pearson’s correlation
coefficient, or correlation coefficient), is a numerical way of expressing how closely two
independent (bivariant) groups of measurements agree on a scale of -1 to +1. When r = 1,
agreement is perfect (as one variable increases so does the other variable); when r = 0, values are
random with respect to each other (no correlation); when r = -1, agreement is perfect but with a
negative association (as one variable increases, the other variable decreases). Intermediate values
indicate mixtures of “agreement and randomness”.
Correlation and the correlation coefficient are frequently used incorrectly in Clinical Biochemistry
and medical sciences. Practically anything can be correlated with anything. A correlation
coefficient by itself does not imply cause and effect.
The Pearson correlation coefficient gives no useful information in method comparison studies
because r can be highly significant even when there is an obvious bias between the two methods.
An r value near 1.0 will be obtained when data points lie close to a straight line even if the slope of
the line varies significantly from 45 degrees (equality). It would actually be very surprising if a
significant r value were not obtained when comparing paired results from two methods which
purport to measure the same analyte.
26. Non-Linear Relationships
There are of course many non-linear relationships. Calculations and curve fitting procedures for
non-linear relationships are much more complicated and are available in specialized statistical texts.

27. References
There are many references describing statistics, quality control, quality assurance, proficiency
testing and related laboratory quality issues (too many to conveniently read in any reasonable
time !). The Westgard internet site is particularly good even though it may take some practice in
getting around. It covers all aspects of laboratory quality, much of which is equally applicable to
POCT. The ten short essays on HbA1c testing are particularly appropriate. There are also book
chapters available for view which cover relevant topics on Quality Control, Quality Assessment and
method validation. There are also many “guest essays” which are very good and very relevant to
this POCT course. You will be required to register for full access to this site … but this is free.
Some specific items from Westgard are shown below.
Westgard home page …

http://www.westgard.com
Westgard “basic QC practices” … list of available statistical information, essays, book chapters and
useful tools. See end of page for list of “guest essays”…
http://www.westgard.com/lessons.htm
Westgard, basis statistics and statistical procedures …

http://www.westgard.com/z-stats-basic-statistics.htm
 Z-Stats (Basic Statistics) lessons:
 Aligning attitudes through purpose
 An organizer of statistical terms, part I
 An organizer of statistical terms, part II
 Mean, standard deviation and coefficient of variation
 Sum of squares, variance, and the standard error of the mean
 Probability and the standard error of the mean
 Hypothesis testing, tests of significance, and confidence intervals
 Two-Sample and hypothesis testing
 Errors, power , and computerized testing
 ANOVA - The analysis of variance
 Confidence Intervals
 Correlation and least squares regression
 Least-squares regression model
 More on regression
 Applying it all
JO Westgard and SA Westgard. First, do no harm. A safety net to catch analytic errors.
http://www.westgard.com/essay125.htm
JO Westgard and SA Westgard. The risk management process.

http://www.westgard.com/essay127.htm#whatRM
CG Fraser. Biological variation data for setting quality specifications in laboratory medicine.
http://www.westgard.com/guest12.htm

 PJ Cornbleet and N Gochman. Incorrect least squares regression coefficients in method
comparison studies. (Deming regression).
Clin Chem 1979; 25; 432.
 JO Westgard. Internal quality control: planning and implementation strategies.
 JM Bland and DG Altman. Statistical methods for assessing agreement between two
methods of clinical measurement.
Lancet 1986; i; 307 to 310.
 National Pathology Accreditation Advisory Council (NPAAC). Requirements for the

development and use of in-house in vitro diagnostic devices 2014.
http://www.health.gov.au/internet/main/publishing.nsf/Content/health-npaac-dhaivd.htm
 National Pathology Accreditation Advisory Council (NPAAC) Requirements for the

estimation of measurement uncertainty 2007.
http://www.health.gov.au/internet/main/publishing.nsf/Content/health-npaac-emu.htm
 Gill JP and Shephard MDS. The conduct of quality control and quality assurance testing
for POCT outside the laboratory. Clin Biochem Rev 2010; 31: 81
Article available in topic library.
 Gill JP and Watkinson L. Chapter 18, Quality control and quality assurance in point-of-
care testing. In: Point-of-care testing. Needs, opportunity and innovation. Edited by Price
CP, St John A and Kricka LJ. AACC Press 2010.
 The series of twenty articles entitled Statistics at Square One, written by TDV Swinscow
and published in the British Medical Journal (BMJ) in 1976 provide a good basic overview
of “medical” statistics (statistics of this nature do not go out of date – modern computers
(PC, Apple, etc) just make the calculations very easy !). They are easily accessible at
http://www.bmj.com/ (you will be required to register, but this is free).
The Statistics at Square One series following articles are:

o I Tabulation of data. BMJ 1976; 1: 1262.
o II Mean and frequency distribution. BMJ 1976; 1: 1325.
o III Standard deviation. BMJ 1976; 1: 1393.
o IV Standard deviation (concluded). BMJ 1976; 1: 1458.
o V Populations and samples. BMJ 1976; 1: 1513.
o VI Variation between samples. BMJ 1976; 1: 1585.
o VII Statements of probability. BMJ 1976; 2: 33.
o VIII Difference between means. BMJ 1976; 2: 94.
o IX Percentages. BMJ 1976; 2: 166.
o X Paired alternatives. BMJ 1976; 2: 225.
o XI The t tests. BMJ 1976; 2: 291.
o XII The t tests (continued). BMJ 1976; 2: 358.
o XIII The t tests (concluded). BMJ 1976; 2: 408.
o XIV The χ2 tests. BMJ 1976; 2: 462.
o XV The χ2 tests (continued). BMJ 1976; 2: 513.
o XVI The χ2 tests (concluded). BMJ 1976; 2: 573.
o XVII Some non-parametric tests. BMJ 1976; 2: 632.
o XVIII Correlation. BMJ 1976; 2: 680.
o XIX Correlation (continued). BMJ 1976; 2: 747.
o XX Correlation (concluded). BMJ 1976; 2: 802.
Topic 6 – Quality Control and External Quality Assessment

 Definitions – Quality management procedures
 Quality management program
 Accuracy and precision
 Types of error (random error, systematic error, trend and shift)
 Internal quality control (QC)
 Statistical quality control
 Westgard rules and establishing control limits
 Limitations of internal QC
 Medically important changes
 External quality assessment
 PowerPoint presentations - Procedures for QC and EQA
 Supplementary references
 Issues with regard to standardisation of POCT technology
 Problem documentation and resolution
 “Non-conformance” and “corrective action” reporting

Refer also to Quality Management in POCT, Overview for Module 2.
On completion of this topic, participants should be able to:

 define the terms; quality management, quality system, quality assurance, quality
control (internal quality control, QC or IQC) quality assessment (external quality
assessment, EQA) and proficiency testing PT)
 describe the terms precision (imprecision), accuracy (inaccuracy) and bias
 describe the use of control materials for POCT systems and how these assist with
maintaining testing quality
 describe the construction and use of control charts and how quality control data is
plotted and interpreted
 describe the Westgard rules for interpretation of laboratory and POCT quality
control data
 apply the Westgard rules to actual control data and determine what actions
should be undertaken to correct out-of-control values
 describe the use of external quality assessment (proficiency testing, EQA), how
this process assists with maintaining testing quality, and provides confirmation that
test results are representative of the “true” value and consistent with other testing
facilities

6.2 Definitions – Quality management procedures
 Quality system: the organisational structure, responsibilities, procedures,

processes and resources needed to implement quality management.
 Quality management: All activities of the laboratory (and laboratory
management), that determine the quality policy, quality objectives and
responsibilities, the implementation of quality procedures such as quality control,
external quality assessment or proficiency testing, measurement uncertainty,
method validation, assessment of reference intervals, and continuing improvement
within the quality system.
 Quality assurance (QA): that aspect of quality management focused on providing
confidence that quality requirements (quality goals, fitness for purpose) are being
fulfilled.
The terminology with regard to quality assurance may at times seem inconsistent.
Depending on the context or particular publication, quality assurance may also be
used to describe proficiency testing, external quality assurance (EQA) or external
quality assessment (also designated as EQA).
 External quality assessment (EQA): often referred to as proficiency testing or
PT); a program in which multiple samples are periodically sent to members of a
group of laboratories for analysis and / or identification, in which each laboratory’s
results are compared with those of other laboratories in the group and / or with an
assigned value, and reported to the participating laboratory and others. Such a
program may also compare an individual’s results with their peer group.
 Quality control (QC) often referred to as internal quality control or (IQC): the
procedure or set of procedures that assess the performance of a testing system in
real time and compares it with previously accepted performance criteria.
For consistency within this course, the terms quality control (QC) and external quality
assessment (EQA) will be used in accord with the above definitions.
6.3 Quality management program
The elements of a good quality management program include initially establishing the
accuracy and precision of the tests which are to be performed. This initial assessment
includes establishing the performance criteria based on medical requirements (that is,
fitness for purpose). The day-to-day usefulness of any POC test depends on maintaining
the accuracy and precision and showing that results are consistent over time. Physicians
make clinical decisions using test results, assuming result variability (uncertainty of
measurement) is maintained at a low level day-to-day, month-to-month and year-to-year.
Patient results which change over time (or appear to change due to testing errors and poor
quality management procedures) may significantly influence their clinical interpretation.

6.4 Accuracy and precision
Accuracy (inaccuracy) is the closeness of a result to the true value. The issue for
laboratory medicine (and POCT in particular) is actually knowing the true value. For many
substances measured in laboratory medicine the true value is not known and for others the
value obtained is method dependent. In the first instance, the material used for assay
calibration should have a defined traceability to an appropriate international standard. In
addition to the many technical issues with regard to the traceability process, substances
measured in biological fluids may show bias due to the biological matrix or actual method
of measurement.
Precision (imprecision) is the closeness of the results of repeated analyses performed

on the same material. Even if results for a given material (patient sample, QC material)
can be consistently reproduced, poor calibration or other technical failure may produce
inadequate results due to bias (inaccuracy).
The “classical” pictorial representation of accuracy (inaccuracy) and precision

(imprecision) is provided by the distribution of target “hits” as shown below.
Accuracy vs Precision
Overall accuracy, but not precise Precise, but not accurate
6.5 Types of error

 Random error (imprecision)
Random errors create a characteristic spread of results for any test method. When
a large number of measurements are made under essentially “identical” conditions,
random errors cause variation which is unpredictable in magnitude and sign. They
cannot be corrected by applying adjustments or “corrections” to the data. Random
errors are difficult to eliminate but repetition reduces the influences of random
errors (refer also to “replicate measurement” in statistics lecture). Examples of
random errors include: errors and variations in pipetting and changes in incubation
period. Random errors can be minimized by training, supervision and adherence
to standard operating procedures.
 Systematic error (bias)

Systematic errors may be constant or proportional. That is, a constant systematic
error or constant bias describes a constant difference between the true analyte
value and the measured (or observed) value, regardless of the concentration level.
Proportional systematic error or proportional bias describes a difference between
the true value and the measured value which changes in proportion to the
concentration value. Systematic errors may be induced by factors such as:
deviation of standard or calibrator from the “true” value or change in the reagent
batch. These “errors” create a characteristic bias in the test results which can be
corrected by applying an adjustment or mathematical “correction” if an accurate
relationship between the variables is known.
 Trend (drift)
A gradual change over time which suggest a progressive problem with the testing
system, a problem with the control material, or other cause(s) which are
responsible for a drift in results.
 Shift
An abrupt and sustained change in one direction usually indicates a significant
performance issue (change) with the analytical system or the control material.
6.6 Internal quality control (QC)
The control of quality has many facets as discussed in textbooks of Clinical Biochemistry
and in the many excellent essays on the Westgard QC internet site (refer to references
below). QC (or IQC) is the procedure or set of procedures that assess the performance of
a testing system in real time and compares it with previously accepted performance criteria
in order to verify that the test (or POC device) is working correctly.
Samples with known values are tested at pre-defined intervals and the results compared
with those obtained on previous occasions. Inspection of the results and comparison with
those obtained previously allows the operator to assess testing performance. The results
are recorded (tabulated) and usually graphed to allow easies inspection of the values. The
“correct” result must be obtained before further testing can proceed. Determining the
acceptable range of values requires knowledge of both the analytical characteristics of the
testing procedure and clinical requirements of the test. Ideally at lease two levels of QC
should be used.
For POCT procedures, the QC material will probably be obtained commercially or as part
of the reagent pack for a given device. Documentation of all operational procedures,
maintenance, incident reporting, identification of non-conformance, corrective actions,
calibrations and QC procedure is required as part of the overall quality system or as
required by accreditation. Recording of QC results, assessment with regard to defined
action limits and the ongoing display of results in a user-friendly and informative manner,

also provides the “hands-on” operator with confidence in the stability of the testing process
and reliability of the results provided.
There are no specific rules for how much or how often to perform QC. The requirements
for accreditation may provide guidance, but in general, if a method (or POCT device)
appears analytically stable and is performing well, it may be acceptable to extend the QC
interval but this must be supported by (historical) records and documentation. Conversely,
if a method (or POCT device) appears to be providing results with more variability, closer
QC monitoring should be considered. Many issues such as this are discussed at
Westgard QC.
Manufacturers of POCT devices will usually suggest a minimum QC monitoring interval,

often using commercial QC material with a broad acceptance range. Both the minimal
testing interval and/or the stated acceptance range may be inadequate for the testing
which is being undertaken. Careful assessment is required prior to the commencement of
operational testing to verify method (or POCT device) performance and the appropriate
number and concentration levels of QC. The National Association of Testing Authorities
(NATA) field application document for ISO standard 15189 states “many factors will
influence the frequency with which QC is performed. The quality control protocol must
take into account these factors and be such that the laboratory has confidence in the
patients’ results”.
Separate recommendations may also apply to cartridge base instruments such as a blood
gas or iSTAT analysers. The NATA field application document provides a good overview
of these alternate procedures (NATA AS 4633 (ISO 15189) Field Application Document,
July 2009 (refer to topic library). The mandatory references for this topic also provide a
useful discussion of these issues.
Mandatory reading
 Gill JP and Shephard MDS. The conduct of quality control and quality
assurance testing for POCT outside the laboratory. Clin Biochem Rev 2010;
31: 81. Article available in topic library.
http://www.aacb.asn.au/files/cbr_articles/August_CBR_2010_Gill_&_Shephard.pdf
 Gill JP and Watkinson L. Chapter 18, Quality control and quality assurance
in point-of-care testing. In: Point-of-care testing. Needs, opportunity and
innovation. Edited by Price CP, St John A and Ericka LJ. AACC Press 2010.

6.7 Statistical quality control
Statistical quality control applies principles from process control – the production of
“widgets” in factories. The procedure allows us to identify whether today’s results are
consistent with previous results. In order to apply statistical QC we use the mean and
standard deviation (SD) of a series of observations to define the variability of the process.
Then, assuming that the errors are Gaussian in distribution, 95.4% of the results will fall
within the mean +/- 2SD, and 99.7% within the mean +/- 3SD. Using this concept, a new
QC result should fall within these limits also. If the new result is outside these limits, it is
likely that some error has been introduced and therefore detected. This process merely
compares the result with previous performance and does not assess whether the
method is inherently satisfactory or fit for purpose (this is required prior to introduction
of the test or device as discussed in a later topic). It must also be noted however, that a
result will fall outside of the +/- 2SD range by chance alone 4.6% of the time (100% -
95.4% = 4.6%). For convenience, the 2SD range is usually taken as 95% confidence
(rounded out from 95.4%).
A very useful form of QC chart was introduced by Shewhart (originally as an industrial

quality check) and later by Levey and Jennings (for clinical laboratory applications). It
uses statistical limits (as SD’s) plotted on a graph to display and compare results.
6.8 Westgard rules and establishing control limits
The Westgard multirule system was developed by James Westgard as a method for
evaluating quality control data ( http://www.westgard.com/westgard-rules ). The rules are
available as part of many QC computer programs or can be applied manually. They are
equally applicable to POCT as to central laboratory testing and provide an excellent
method for assessing QC data for POCT.
The Westgard multirule system may be established by the following procedure:

 Analyse the same specimen at least 20 times (20 + results, that is 20 + replicates)
and calculate mean and standard deviation. Plot mean +/- 2SD and +/- 3SD.
 Westgard rules:
As the mean +/- 2SD limits are often too narrow (limits too tight) or not sufficiently
“sensitive” to gradual change, and mean +/-3 SD too loose, Westgard developed a
multi-rule concept. This multi-rule approach applies a series of rules (usually 6) in
sequence to the interpretation.
 In QC evaluation, it is usual practice to run at least two different controls (at two
different levels), usually chosen to represent values at critical clinical decision
points (such as the upper limit of the reference interval). The following guidelines
(Westgard rules) should be used when two (2) different controls are measured.
In the Westgard information below “s” = SD.
6.8.1 The rules:
 1 2s (1 2s or 1,2SD rule). One result greater than 2s but less than 3s. A warning
rule. Requires careful inspection of the control data using the following rejection
rule(s). Does not require test rejection.

 1 3s Testing should stop when a single control measurement exceeds the mean
plus 3s or the mean minus 3s control limit. Detects primarily random error. May be
an indication of the beginning of large systematic error.
 2 2s Reject when 2 consecutive control measurements exceed the mean plus 2s

or the mean minus 2s control limit. Detects systematic error or bias.
 R 4s Reject when 1 control measurement in a group exceeds the mean plus 2s

and another exceeds the mean minus 2s. Detects random error.
 4 1s Reject when 4 consecutive control measurements exceed the same mean

plus 1s or the same mean minus 1s control limit. Indicates a systematic error or
bias in the test system.
 10x Reject when 10 consecutive control measurements fall on the same side of
the mean. Indicates a systematic error or bias.
Triggering (failing) any of the last 5 rules requires testing to stop until the reason for the
error(s) has been determined and resolved. Further testing of QC must give acceptable
results consistent with previous good performance before recommencing testing of
patients.

6.9 Limitations of internal QC
Internal QC only detects change in performance between current and “stable” operation. If
the original mean as determination contained a systematic error(s), this will not be
detected. The original method evaluation (method validation) needs to establish accuracy
(as well as recovery and interference). Ongoing comparison studies are needed to ensure
systematic errors do not increase slowly and are undetected (this can be largely achieved
by external quality assessment (EQA) or proficiency testing).
6.10 Medically important changes
A variety of approaches have been used to establish acceptance criteria based on the
clinical use of the test results. Performance goals based on biological variation will be
discussed in the lecture on biological variation and reference intervals.
6.11 External quality assessment
Internal quality control (QC) is essential for detecting problems that arise during the routine
operation of POCT methods. These problems are caused by instability in the measuring
process and on-going changes in performance compared to what was previously
observed. Internal QC compares laboratory performance to itself over time, assuming that
the performance observed earlier represents correct or appropriate performance.
Appropriate analytical performance, should be validated before the method or procedure is
first introduced into routine operation. This initial confirmation is achieved by method
validation experiments and must be confirmed (or revalidated) on a continuing basis by
external quality assessment (EQA) or proficiency testing (PT).
In an EQA program, a group of laboratories analyze the same specimen(s) (usually the
same control materials) and submit their results to a central facility where the data are
examined for consistency and comparative statistics are produced. These statistics
usually include:
 a check for outliers
 the overall mean and standard deviation of result data
 specific method related means and standard deviations
 other statistics which characterize the performance of the group
 reports to participants which compare the performance of an individual laboratory
to the peer group, and often to target values established by reference methods or
reference laboratories.
It is much more difficult to determine the accuracy of a method on an on-going basis. The
laboratory or POCT facility needs to compare its test results with the true or correct values.
“True values” are often approximated by using results from other established laboratory
methods. For example, in the initial validation studies of a method a group of patient
specimens are usually analyzed by the new method and an established method. The
mean of results by the new method is compared to the mean of results by the established
method to estimate the bias (average difference) between methods. The essential feature
of an EQA program is the availability to compare results from other methods in other
facilities. The unique features of an EQA program are the ability to monitor the
comparative accuracy of methods, to provide an assessment of on-going performance

stability, and to provide reassurance that performance of your method(s) are aligned with
the “true” or correct values.
6.11.1 External quality assessment programs
To be eligible for Medicare benefits in Australia, a laboratory (including a POCT service or

laboratory utilising POCT devices) must fulfil a number of specified requirements, one of
which is to demonstrate active participation in an external quality assessment (EQA) or
proficiency testing (PT) programme.
The NATA Field Application Document as discussed previously provides guidance in how
participation should be conducted … “(i) Where available, a program which includes other
laboratories in the country / region is preferred as this optimises the opportunity to reduce
between-laboratory variation such as may be seen in a patient using multiple laboratories.
(ii) Where analysers (e.g. blood gas analysers) are located outside the laboratory these
must be enrolled separately in a QA unless they meet the requirements of Appendix B (QA
enrolment requirements for multiple same-analyte instruments …). (iii) Regular
submission of results to the program organisers is required whether or not the timing
coincides with the testing of patients’ samples. (iv) Participants must test QA samples in a
manner similar to patient samples. …”
In Australia, the most popular and well recognised EQA programme is that provided by the
Royal College of Pathologists of Australasia in association with the Australasian
Association of Clinical Biochemists. Information regarding this program be found at :
https://www.rcpaqap.com.au
6.12 PowerPoint presentations – Procedures for QC and EQA
The associated PowerPoint presentation also forms part of this topic. It provides an
overview of accuracy and precision, internal QC and EQA.
The quality of test results is a major concern with POCT. While POCT may look
deceptively simple, there are many opportunities for pre-analytical, analytical, and post-
analytical errors to be introduced. POCT devices usually incorporate built-in quality
systems that automatically check instrument function, reagent quality, and test
performance. These automatic functions aim to keep faulty instruments (and operator
procedures ?) from being used to produce erroneous results. In-built functions such as
these are an adjunct to good quality management and should not be used to replace an
effective QC and EQA programmes.

 AACB. Point of care testing implementation guide, 2008. Article available in topic
library.
 NATA Accreditation publications. AS 4633 (ISO 15189), Field Application
Document, medical testing, July 2009.
 RCPA/AACB Chemical Pathology QAP. https://www.rcpaqap.com.au
 ISO. Point-of care testing (POCT) – Requirements for quality and competence.
International standard ISO/DIS 22870.
 Westgard QC. (Much useful information on all aspects of QC, QA, methods for
presentation of data, preparation of worksheets, statistics and general
calculations). http://www.westgard.com/
 KE Blick and RB Passey. Quality control for the clinical chemistry laboratory.
Chapter 25 in: Clinical Chemistry, theory, analysis and correlation. Edited by LA
Kaplan and AJ Pesce, 5th edition, 2010.
 GG Klee etal. Quality management. Chapter 16 in: Tietz, Fundamentals of
Clinical Chemistry. Edited by CA Burtis, ER Ashwood and DE Burns, 6th edition,
2008.
 CG Fraser. Analytical requirements for point-of-care testing. Chapter 8 in; CP
Price, A St John and JM Hicks. Point-of-Care Testing, 2nd edition, 2004.
http://books.google.com.au/books?id=wcXrAgOVXbUC&printsec=frontcover#v=on
epage&q=&f=false
 DG Bullock. Quality control and quality assurance in point-of-care testing. Chapter
13 in; CP Price, A St John and JM Hicks. Point-of-Care Testing, 2nd edition, 2004.
http://books.google.com.au/books?id=wcXrAgOVXbUC&printsec=frontcover#v=on
epage&q=&f=false
 CG Fraser. Optimal analytical performance for point of care testing. Clin Chim
Acta 2001; 307: 37.
 CL Martin. Quality control issues in point of care testing. Clin Biochem Rev 2008;
29, supplement (i): S79. Article available in topic library.
 M Shephard et.al. Design, implementation and results of the quality control
program for the Australian government’s point of care testing in general practice
trial. Ann Clin Biochem 2009; 46: 413.
 Radiometer, acute care testing. http://www.acutecaretesting.org/
 M Shephard. How to set up and manage a POCT service. Clinical Biochemist
Newsletter. Article available in topic library.

6.13 Issues with regard to standardisation of POCT technology
With the possible exception of blood gas and acid base, POC testing equipment invariably
differs from that of a central Pathology laboratory. Also as discussed previously, results
from POC devices often vary from those obtained from a Pathology laboratory service.
Understanding these differences is the critical issue. Blood gas and acid base analysers
are a special case in this regard, as the same type of instrument may be used in both a
POC situation and a Pathology laboratory. Even so, different brands of blood gas analyser
and different models within the same brand, often give “different” results which are clearly
shown on proficiency testing reports (such as the end of period reports from QAP).
As many POCT devices measure substances directly on whole blood, significant matrix
effects may be introduced if calibration is performed using a “standard” contained in an
alternate matrix such as plasma (serum) or aqueous solution. Such “standards” are often
assigned values which represent the level which is considered appropriate for a whole
blood specimen on the specific POC device. In this regard, the specific calibrator can be
considered as a “contrived” standard, the purpose of which is to ensure that the POC
device is set at the correct analytical measurement level. In a similar manner, aqueous
“control material” (often provided with replacement reagent packs or test cartridges) or
“control material” with a different matrix base to the recommended patient sample, may
show consistent and reproducible results even though it does not truly represent the actual
variations which may occur in patient samples.
Other aspects of this difficult issue with regard to calibration of POCT devices and
traceability of results have been discussed previously in topic 2 (refer to item 2.4,
analytical differences between whole blood, plasma and serum; and item 2.6, direct and
Indirect potentiometry). The general issue of traceability is discussed in the topic of
measurement uncertainty.
6.14 Problem documentation and resolution
The International Standard (ISO 15189) specifies in detail the type of documentation
required for medical testing “laboratories” as discussed in topic 4. Documentation of
operational procedures is a basic component of quality management and applies equally
to POCT. As indicated previously, documentation should include information about a
methods’ (POCT device) performance under both stable and unstable conditions, including
regular maintenance, calibrations, change of reagent lot number, any inconsistency in
performance, QC and EQA. It should include a description of initial validation, past
operational performance and current performance. This information will assist with
trouble-shooting and corrective action. The documentation (and the various records) may
be either manual or electronic.
6.15 “Non-conformance” and “corrective action” reporting
Non-conformance and corrective action are quality management terms which are now
commonly used in pathology testing and accreditation procedures.
 Corrective actions; steps taken to remove the causes of an existing non-

conformity or undesirable situation. The corrective action process is designed to
prevent the recurrence of non-conformities or undesirable situations by

eliminating causes or situations which produced the (undesirable) action. Because
of this, the corrective action process can be regarded as a problem solving
process.
 Non-conformance (or non-conformity); refers to a failure to comply with
requirements. There are many types of requirements and include … quality
requirements, customer requirements, management requirements, procedural
requirements, and legal requirements. Whenever an organization or process fails
to meet one of these requirements, nonconformity occurs.
 Preventive action(s); steps that are taken to remove the causes of potential non-
conformities or potential situations that are undesirable. The preventive action
process is designed to prevent the occurrence of errors (non-conformance) or
situations that do not yet exist. While corrective actions prevent recurrence,
preventive actions prevent occurrence. Both types of actions are intended to
prevent non-conformities. In general, the preventive action process can be
thought of as a risk analysis process.
Mandatory reading
innovation. Edited by Price CP, St John A and Ericka LJ. AACC Press 2010.

Topic 7 – Biological Variation and Reference Intervals

 Overview of topic 7
 Uncertainty of measurement and reference intervals
 Pre-analytical factors which influence patient results
 The difference between population and sample statistics
 Biological variation
 Intra-individual and inter-individual variation
 Reference intervals and decision values
 PowerPoint presentations
 Relationship between reference intervals and uncertainty of measurement
 Interpretation of serial measurements (how good must serial results be to
effectively monitor significant change ?)
 Clinical utility of uncertainty of measurement data

 the pre-analytical and patient related factors which may influence test results
 procedures for minimising pre-analytical interference
 the difference between the statistical terms of “population” and “sample”
 “biological distributions” and factors which affect statistical sampling
 “inter-individual” and “intra-individual” biological variation
 the statistical procedure(s) used to quantify a reference interval
 the effect of analytical variability on reference intervals
 “critical difference” and the interpretation of a change in a patient’s test results.
 The relationship between uncertainty of measurement, reference intervals and
clinical interpretation of patient results

The clinical interpretation of laboratory and POCT data is a comparative process in which
a test result for an individual is compared with a reference interval derived from
reference values. It is therefore critical that reliable reference intervals are available.
Understanding the process used to establish reference intervals also provides a better
understanding of the limitations inherent in the procedure.
In addition to reference intervals, clinical interpretation for some tests may be based on
comparison to defined limits or clinical decision values. Examples of clinical decision
values are glucose, HbA1c and cholesterol.
 Diagnostic criteria for the diagnosis of diabetes mellitus are based on measured
glucose values exceeding defined limits.
 The use of HbA1c to monitor glucose control in a diabetic patient, requires the
measured HbA1c level to be less than 53 mmol/mol (or 7.0% in “old” NGSP units)
for “good control” to be obtained.
 The National Heart Foundation and the Cardiac Society of Australia and New
Zealand provide recommended target level for patients being treated with
cholesterol lowering drugs (Total cholesterol < 4.0 mmol/L and LDL-cholesterol <
2.5 mmol/L).
A patients’ test result(s) may thus be compared to a reference interval derived from
“healthy” individuals or compared to a clinically defined decision value. In both of these
circumstances, an appreciation of testing imprecision and inaccuracy assists in answering
the two fundamental questions relating to result interpretation:
 What is the probability that a result, which is subject to error, is significantly
different to a previous result for the same substance ?
 What is the probability that a result, which is subject to error, differs from a fixed
value or decision value ?
Detailed discussions regarding reference intervals, their development and utilisation, are
provided in standard test books on Clinical Chemistry (Clinical Biochemistry).
7.3 Uncertainty of measurement and reference intervals

The correct interpretation of test results relies on many factors, including:
 a good understanding of the underlying physiology and pathophysiology

 any artefacts (or errors) which may be introduced as pre-analytical, analytical or
post-analytical factors
 the accuracy and precision of the testing process (uncertainty of measurement)
 the relative deviation of a particular result(s) from those usually found in healthy
persons
 the actual range of results usually found in a group of healthy persons (reference
range or reference interval)
 the natural variation of serial results within one individual over a period of time
(intra-individual variation)
 the difference required for consecutive results to actually be different (for serial
results to be significantly different, the difference in numerical values must be
greater than the combined variation inherent in the two results, the critical
difference or reference change value)
 the relationship between the above items, in particular the relationship between the
biological reference range, the intra-individual biological variation and
measurement uncertainty.
7.4 Pre-analytical factors which influence patient results
The correct collection, processing and storage of samples for laboratory and POC testing
are critical to the production of quality test results. Errors which may occur prior to the
analytical reaction are considered pre-analytical. Recognising and minimising these
potential errors is an important part of the overall testing process.
Many of the pre-analytical errors in pathology (and POC) testing relate to specimen
collection and correct patient identification. In addition, physiological and life-style
associated factors may increase result variation if appropriate patient preparation or
interpretative reference intervals are not considered. These physiological and life-style
pre-analytical variables include:
 Gender and age (newborn, childhood and puberty, adult, elderly)
 Diet
 Diurnal variation (circadian variation), menstrual cycle
 Time of collection, fasting, post prandial, etc
 Drugs, vitamins and herbal preparations which a patient may be taking
 Posture when specimen collected

Exercise
Review the information presented in Topic 2 – Pre-Analytical variables.
7.5 The difference between population and sample statistics
Definitions:
Sample, a set of individuals or items selected from a population so that the properties or
parameters of the population may be estimated.
Sampling error is the difference between the true value of a parameter of a population
and that estimated from the sample. This error is due to the calculation being based on
the sample (or sub-set) rather than the whole population (refer also to topic 5 – Statistics).
Sample, pathology. One or more parts taken from a system and intended to provide
information on the system (definition from AS4633 / ISO 15189).
7.6 Biological variation
If we could measure a particular blood constituent from the same sample a large number
of times, we may expect (ideally) to obtain the same result on each occasion. In practice
however, the result is not always the same due to analytical imprecision. The results
obtained usually vary by a small amount but cluster around a particular value. This central
value is not necessarily the true value due to possible systematic error or bias. Similarly, if
samples were collected from the same (normal) individual on several occasions, the
results for the measured constituent would cluster around a central value but the spread
would be greater (wider) than multiple measurements on the same sample. This greater
variation is due to intra-individual variation being added to the analytical variation
(analytical imprecision). To extend this process further, if samples were collected from
many different individuals (not just the one individual as before), the spread of results
would be even greater due to inter-individual variation. These concepts of analytical, intra-
individual and inter-individual variability underlie many of the issues discussed in this
module. Their importance to the interpretation of pathology results should not be
underestimated.

Many analytes also vary over an individual's lifetime, simply because of natural biological
factors involved in the aging process. These variations may occur rapidly at critical points
in the life cycle, such as during the neonatal period, childhood, puberty, menopause, or old
age (age related factors). In addition, some analytes have predictable biological rhythms
or cycles. These cycles may be daily, monthly, or seasonal. Knowledge of these cycles is
vital for test result interpretation. For example, a patient sample must be collected at the
time in the cycle that is appropriate for the clinical purpose to which the test result will be
applied (for example cyclical changes in female hormones). Since developing good
reference values is complex and time consuming, it is important to generate these values
correctly and at clinically important decision-making points.
7.7 Intra-individual and inter-individual variation
Many (most) analytes however, do not have cyclical rhythms as described above. Any
variability or fluctuation which can be observed is principally due to random fluctuation
around a homeostatic setting point. This variability within a given individual is known as
the intra-individual biological variation. The setting point for different individuals will
probably be similar but is unlikely to be exactly the same. The overall variation between
individuals is known as between-subject or inter-individual biological variation.
This biological variation, or the natural fluctuation of constituents around their homeostatic
setting point can therefore be classed as within-subject and between-subject variation
(intra-individual or inter-individual variation).
In laboratory practice, the application of biological variation can be used to:

 set quality specifications for analytical performance
 evaluate the clinical significance of any change in consecutive results from a given
patient
 evaluate the clinical significance of a result in relation to a clinical decision value
 validate new procedures, devices or test methods and their fitness for purpose
These applications are discussed further as part of the PowerPoint presentations.
An extensive list of intra-individual and inter-individual biological variation is given by

C Ricos on the Westgard internet site at http://www.westgard.com/biodatabase1.htm

7.8 Reference intervals and decision values
Discussed as part of the PowerPoint presentations.
The several PowerPoint presentations are provided as part of module 2. These integrate
many of the statistical concepts which are required for an understanding of result
interpretation and the correct utilisation of POCT technology. They include:
 POCT and laboratory statistics
 uncertainty of measurement
 biological variation and reference intervals
 critical difference
 fitness for purpose
7.9 Relationship between reference intervals and uncertainty of measurement
7.10 Interpretation of serial measurements (how good must serial results be to

7.11 Clinical utility of uncertainty of measurement data

Mandatory reading for module 2
 GH White and I Farrance. Uncertainty of Measurement in quantitative

medical testing – A laboratory implementation guide.
Clin Biochem Reviews 2004; 25: Supplement (ii) S1 to S24. Article available
in topic library.
innovation. Edited by Price CP, St John A and Kricka LJ. AACC Press 2010.
 KE Blick and RB Passey. Quality control for the clinical chemistry laboratory.
Chapter 25 in: Clinical Chemistry, theory, analysis and correlation. Edited by
LA Kaplan and AJ Pesce, 5th edition, 2010.
 Fraser CG. Optimal analytical performance for point of care testing.

Clin Chim Acta 2001; 307: 37.
 Fraser CG and Scott MG. Chapter 12, Analytical performance requirements

for point-of-care testing: Glucose as an example. In: Point-of-care testing.
Needs, opportunity and innovation. Edited by Price CP, St John A and Kricka
LJ. AACC Press 2010.
 Jones GRD etal. Change of HbA1c reporting to the new SI units.

Medical J Australia 2011; 195: 45. Article available in topic library.

 CG Fraser. Biological variation, from principles to practice. AACC Press, 2001.
 NPAAC. Requirements for the estimation of measurement uncertainty, 2007.

emu.htm
 PS Horne. Reference intervals and clinical decision limits. Chapter 24 in: Clinical
Chemistry, theory, analysis and correlation. Edited by LA Kaplan and AJ Pesce,
5th edition, 2010.
 HE Solberg. Establishment and use of reference values. Chapter 14 in: Tietz,

Fundamentals of Clinical Chemistry. Edited by CA Burtis, ER Ashwood and DE
Burns, 6th edition, 2008.

Topic 8 – Uncertainty of Measurement and Fitness for Purpose

 Overview of topic 8
 PowerPoint presentations
 Uncertainty of Measurement or Measurement Uncertainty (UM, MU)
 Relationship between reference intervals and UM
 Fitness for purpose
 Interpretation of serial measurements (how good must serial results be to
 Clinical utility of MU data
 the concept of Uncertainty of Measurement and laboratory error

 how “repeatability” and “traceability” are the basic elements which underlie the
concept of Uncertainty of Measurement
 what factors contribute to “laboratory error” and Uncertainty of Measurement
 the concept and practical use of “numerical precision” in reporting laboratory results
 the clinical requirements and their relationship to analytical accuracy and precision
 how “fitness for purpose” may be used to assess the relevance of analytical
performance to clinical interpretation
 how specific POCT devices may be assessed as “fit for purpose”

Uncertainty of measurement (also referred to as Measurement Uncertainty, MU),

traceability and numerical significance are separate but closely related concepts that affect
both the format and the information conveyed by a quantitative test result. In addition, the
use of SI units provides a consistent basis for the reporting of clinical laboratory data.
 Uncertainty: “A parameter associated with the result of a measurement that
characterises the dispersion of the values that could reasonably be attributed to the
measurand” (GUM).
 Traceability: “Property of the result of a measurement or the value of a standard,
whereby it can be related to stated references, usually national or international
standards, through an unbroken chain of comparisons all having stated
uncertainties” (ISO 15189).
 Numerical Significance: The significant figures of a number are those that have
some practical meaning. The significant figures of a number express its magnitude
to a specified degree of accuracy.
 Systeme Internationale (SI) units: The system of metric units which has been
adopted by agreement in all major countries for use in science, medicine, industry
and commerce. SI is a coherent system based on the seven basic quantities of
length (meter, m), mass (kilogram, kg), time interval (second, s), electric current
(ampere, A), thermodynamic temperature (degree Kelvin, K), luminous intensity
(candela, cd) and amount of substance (mole, mol).
As every measurement has an error, it is often stated that a measurement result is only
complete when accompanied by a quantitative statement of its uncertainty. The
uncertainty assessment is required in order to decide if the result is adequate for its
intended purpose (fit for purpose) and to ascertain if it is consistent with other similar
results. Both of these concepts will be explored further as part of this topic.
8.3 PowerPoint presentations
The several PowerPoint presentations are provided as part of module 2. These integrate
many of the statistical concepts which are required for an understanding of result
interpretation and the correct utilisation of POCT technology.

They include:
 POCT and laboratory statistics
 uncertainty of measurement
 biological variation and reference intervals
 critical difference
 fitness for purpose
8.4 Uncertainty of Measurement (Measurement Uncertainty)
There are two major sources of uncertainty which contribute to the total uncertainty of
measurement of a routine quantitative diagnostic (POCT) method. Firstly, there is
uncertainty associated with the numerical value assigned to the calibrator used in the
standardization process. This uncertainty should be estimated by the commercial supplier
of the calibrator or by the laboratory if the calibrator has been prepared in-house. The
method for estimating the uncertainty of calibrator value(s) will depend on how the value is
determined (for example, gravimetric, definitive method, etc). For most POCT methods,
calibration (or calibrator value assignment) by the manufacturer will be by cross
comparison of patient specimens against a laboratory reference procedure. At the present
time only some commercial manufacturers of calibration materials provide the necessary
uncertainty estimates of assigned values (if specifically requested however, this
information should be made available). Secondly, there is uncertainty associated with the
value of a test result due to the random errors that normally occur when conducting the
testing procedure. This uncertainty component is demonstrated by the dispersion of
values observed when a measurand (analyte) in the same specimen is repeatedly
measured by a properly conducted test method. In the medical and POCT laboratory this
dispersion is termed imprecision, and has long been used as the basic quantitative
estimate of the confidence that can be placed in a result.
Actual method performance data derived from internal quality control (QC) also frequently
include the effect of several sources of uncertainty simultaneously. Information on test
method performance is typically obtained from:
 data accumulated during validation and verification of a test method prior to its
application in the testing environment
 inter-laboratory studies, split-specimen comparisons, etc
 accumulated quality control and external quality assessment data
For practical purposes, imprecision data obtained from the routine application of internal
quality control is recommended as the quantitative estimate of the uncertainty of
measurement. For clinicians who use and interpret POCT information, the dispersion of
test results around a clinical decision value is the major uncertainty that has the potential
to affect interpretation and clinical management.
Where the estimate of uncertainty is known for both the calibrator and the routine
analytical imprecision of a test procedure, the total estimate of uncertainty can be
calculated by summing the two variance estimates as described in topic 5 (Statistics for
POCT and laboratory applications).
8.5 Relationship between reference intervals and UM
The correct interpretation of test results relies on many factors, including:
 a good understanding of the underlying physiology and pathophysiology
 any artefacts (or errors) which may be introduced as pre-analytical or post-

analytical factors
 the accuracy and precision of the testing process (uncertainty of measurement)
 the relative deviation of a particular result(s) from those usually found in healthy
persons
 the actual range of results usually found in a group of healthy persons (reference
range or reference interval)
 the natural variation of serial results within one individual over a period of time
(intra-individual variation)
 the difference required for consecutive results to actually be different (for serial
results to be significantly different, the difference in numerical values must be
greater than the combined variation inherent in the two results, the critical
difference or reference change value).
 relationship between the above items, in particular the relationship between the
biological reference range, the intra-individual biological variation and uncertainty
of measurement (UM or MU).

Exercise
As much of this topic is mathematically and statistically based, review the basic
statistical procedures provided as topic 5.
Mandatory reading
 GH White and I Farrance. Uncertainty of measurement in quantitative

medical testing. Clin Biochem Rev 2004; 25, supplement (ii): S1.
8.6 Fitness for purpose
The fundamental role of medical testing laboratories and POCT providers is to produce
test results that are fit for their purpose. A test result must have appropriate analytical
accuracy and precision in order for it to be considered fit for purpose: that is, suitable for
the clinical purpose for which the test is being used. In order to determine whether a
method is routinely producing results which are fit for purpose, there needs to be a
relevant analytical goal against which the estimated uncertainty of measurement can be
compared. Some methods have internationally agreed analytical goals (for example;
glucose, cholesterol and haemoglobin A1c), but in their absence various approaches have
been used to set goals for both inaccuracy (bias) and imprecision. A widely used and
internationally recommended concept is to define the upper acceptable limit for
imprecision as a proportion of the intra-individual biological variation of the measurand
(analyte). With correct choice of the proportionality factor, analytical imprecision should
not contribute significant additional variation to the test result when compared with the
natural variation of the substance being measured. A similar approach to goal-setting can
be used for total analytical error (inaccuracy plus imprecision). These concepts are clearly

described in the NPAAC document Requirements for the estimation of measurement
uncertainty and the article by White and Farrance Uncertainty of measurement in
quantitative medical testing – A laboratory implementation guide.
The practical problems associated with selecting appropriate quality goals (quality
specifications) are exemplified by the most widely used POCT device, the glucose meter.
The American Diabetes Association (ADA) and the National Academy of Clinical
Biochemistry (NACB) have documented this dilemma in considerable detail (National
Academy of Clinical Biochemistry (NACB). Guidelines and recommendations for
laboratory analysis in the diagnosis and management of diabetes mellitus, 2011.
https://www.aacc.org/science-and-research/practice-guidelines/diabetes-mellitus and
Sacks DB, etal. Guidelines and recommendations for laboratory analysis in the diagnosis
and management of diabetes mellitus. (Clin Chem 2002; 48: 436). They reinforce the
WHO recommendation that a laboratory measurement of plasma glucose be used for the
diagnosis of diabetes mellitus, and not whole blood glucose measured on a glucose meter
with the statement “Portable meters are used by healthcare workers in acute and chronic
care facilities, in physicians’ offices, and by patients. Because of the imprecision and
variability among meters, they should not be used to diagnose diabetes and have limited
value in screening.” One of the issues which should also be considered as part of this
argument, particularly for hospital in-patients whose insulin infusion (and consequent
insulin dosage) is adjusted in association with POC glucose testing, is that relatively small
changes in measured glucose are associated with relatively large changes in insulin
infusion rate. Blood glucose monitoring is used to guide therapy in two distinct situations,
one to adjust insulin dosage in diabetic patients, and one for adjusting insulin requirement
in acutely ill patients on “tight glucose control”. The quality specifications for these two
situations have been reviewed by Price (Price CP. Glucose monitoring and patient safety.
Point of Care 2008; 7: 253). In a recent evaluation study by Freckmann etal (Freckmann
G, etal. System accuracy evaluation of 27 blood glucose monitoring systems according to
DIN EN ISO 15197. Diabetes Technology and Therapeutics 2010; 12: 221.), eleven of
twenty seven home-use blood glucose monitoring systems (40%) did not fulfill the rather
generous minimum accuracy requirements of ISO 15197 (that is; +/- 0.83 mmol/L at blood
glucose less than 4.2 mmol/L, or +/- 20% at blood glucose levels 4.2 or greater). From
their study, Freckmann etal concluded that inaccurate results which lead to false treatment
decisions by diabetic patients may cause severe health injury.

In addition to portable POC glucose meters, the fitness for purpose of portable devices
which measure haemoglobin A1c (HbA1c) has also been questioned recently (Burns DE
and Boyd JC. Few point-of-care haemoglobin A1c assay methods meet clinical needs
(editorial). Clin Chem 2010; 56: 4. and Lenters-Westra E and Slingerland RJ. Six of
eight haemoglobin A1c point-of-care instruments do not meet the general accepted
analytical performance criteria. Clin Chem 2010; 56: 44.). The main reason for this
concern is the questionable ability of some HbA1c devices to provide a clear analytical
distinction between the recommended HbA1c treatment levels of 53 mmol/mol (or 7.0% in
“old” NGSP units) and 64 mmol/mol (or 8.0% in “old” NGSP units). To clearly distinguish
between HbA1c values of 53 mmol/mol and 64 mmol/mol, an analytical (method)
uncertainty (expressed as coefficient of variation, CV%) of less than 3.0% is required.
Methods which produce analytical CV’s of 4.0% or greater are not considered appropriate,
as this degree of analytical imprecision cannot distinguish changes in patient HbA1c levels
of under approximately 10 mmol/mol (that is, can not distinguish a patient HbA1c result of
53 mmol/mol from 64 mmol/mol). A more detailed discussion of these concepts and of the
relationship between HbA1c measurement and analytical performance is given by White
and Farrance (Uncertainty of measurement in quantitative medical testing. Clin Biochem
Rev 2004; 25, supplement (ii): S1).
The direct influence of analytical imprecision and systematic error (bias) on laboratory and
POC measurements are critical issues for the interpretation of patient results and the
fitness for purpose of testing devices.
Mandatory reading

medical testing. Clin Biochem Rev 2004; 25, supplement (ii): S1. Article
available in topic library.
http://www.aacb.asn.au/files/cbr_articles/uncertainty_of_measurement.pdf
 National Academy of Clinical Biochemistry (NACB). Guidelines and

recommendations for laboratory analysis in the diagnosis and management
of diabetes mellitus, 2011. https://www.aacc.org/science-and-
research/practice-guidelines/diabetes-mellitus

8.7 Interpretation of serial measurements

Refer also to reference:
 K Linnet and JC Boyd. Selection and analytical evaluation of methods – with
statistical techniques. Chapter 13 in: Tietz, Fundamentals of Clinical Chemistry.
Edited by CA Burtis, ER Ashwood and DE Burns, 6th edition, 2008.
8.8 Clinical utility of UM data
When presenting test result and uncertainty data, the person responsible for the POCT
facility should review the application of SI units and the relevant number of significant
figures used for reporting both the numerical result and any uncertainty estimate. The
number of significant figures used in reporting a result has the capacity to impart an
incorrect impression of the uncertainty of the test measurement if appropriate rounding
does not occur.
For some specific methods or clinical applications, the provision of uncertainty data
together with the test result may reduce the potential for significant clinical
misinterpretation (for example, immunological-based methods, where antibody specificity,
cross-reactivity with closely related species or clinically significant interfering substances
are probably unknown to the requester).
In addition to the clinical application of a test result, there are two important aspects which
also need to be considered. The most important of these is the in vivo biological variability
of the measurand, as this is the signal that may differentiate health from disease. The
second is the imperfection in the analytical method that may lead to different results on
different occasions. It is vitally important that variation due to imperfect analysis (the
analytical uncertainty) is less than the measurement signal we are trying to discriminate.
As a general principle, it has been widely suggested that the analytical goal for imprecision
of a test method remain below half the intra-individual biological variation (CVA < 0.5 CVI).
If this condition is satisfied and the analytical variability is appropriately less than the
biological variability, the test can be confidently used for clinical diagnosis and monitoring.
The impact of uncertainty does not end here however, as diagnostic decisions may be
made by comparison to a reference population (reference interval or limit) or compared to
a diagnostic cut-off. The methods used to establish these diagnostic decision points have
their own imperfections, but once established they become set values without variation.
Analytical uncertainty will change the “distance” between the test result and the particular
cut-off used for comparison. If the “distance” between the test result and the diagnostic
cut-off point is less than 1.96 SD (2.0 SD), then it cannot be stated (at the usual 95%
confidence level) that a repeat analysis would not produce an analytically valid result on
the other side of that diagnostic cut-off. This analytical uncertainty should be conveyed to
the clinician who might otherwise see the result in more absolute terms.
These concepts are further discussed as part of the PowerPoint presentations.
Mandatory reading

 National Academy of Clinical Biochemistry (NACB). Guidelines and

recommendations for laboratory analysis in the diagnosis and management
of diabetes mellitus, 2011. https://www.aacc.org/science-and-
research/practice-guidelines/diabetes-mellitus
 Fraser CG. Optimal analytical performance for point of care testing.

Clin Chim Acta 2001; 307: 37.
 Fraser CG and Scott MG. Chapter 12, Analytical performance requirements

for point-of-care testing: Glucose as an example. In: Point-of-care testing.
LJ. AACC Press 2010.

8.10 Supplementary reading
 CG Fraser. Biological variation, from principles to practice. AACC Press 2001.
 NPAAC. Requirements for the estimation of measurement uncertainty, 2007.

emu.htm
 CC Gruber and RN Carey. Evaluation of methods. Chapter 26, in: Clinical

Chemistry, theory, analysis and correlation. Edited by LA Kaplan and AJ Pesce,
5th edition, 2010.
 K Linnet and JC Boyd. Selection and analytical evaluation of methods – with

statistical techniques. Chapter 13 in: Tietz, Fundamentals of Clinical Chemistry.
Edited by CA Burtis, ER Ashwood and DE Burns, 6th edition, 2008.
 Price CP. Glucose monitoring and patient safety. Point of Care 2008; 7: 253.
 Freckmann G, etal. System accuracy evaluation of 27 blood glucose monitoring

systems according to DIN EN ISO 15197. Diabetes Technology and Therapeutics
2010; 12: 221.
 Sacks DB, etal. Guidelines and recommendations for laboratory analysis in the
diagnosis and management of diabetes mellitus. Clin Chem 2002; 48: 436.
 Burns DE and Boyd JC. Few point-of-care haemoglobin A1c assay methods meet
clinical needs (editorial). Clin Chem 2010; 56: 4.
 Lenters-Westra E and Slingerland RJ. Six of eight haemoglobin A1c point-of-care

instruments do not meet the general accepted analytical performance criteria. Clin
Chem 2010; 56: 44.
 Little RR. Performance of HbA1c assay methods: Good enough ? Clin Chem
2014; 60: 1031.
 Lenters-Westra E and Slingerland RJ. Three of seven HbA1c POCT instruments

do not meet generally accepted analytical performance criteria. Clin Chem 2014;
60: 1062.
Module 3 – Clinical applications

Market segments for point-of-care testing devices
In vitro diagnostic devices
It is difficult to obtain consistent up-to-date information with regard to the global markets for
in vitro diagnostics and devices (IVD’s) for point-of-care testing. As a general concept, the
term in vitro diagnostic (medical) device is usually defined as “any medical device which is
a reagent, reagent product, calibrator, control material, kit, instrument, apparatus,
equipment, or system, whether used alone or in combination, intended by the
manufacturer to be used in vitro for the examination of specimens, including blood and
tissues derived from the human body” (NPAAC, Requirements for the development and
use of in-house in vitro diagnostic medical devices; third edition 2014). This IVD definition
thus includes all pathology testing equipment including POCT devices and associated
consumables. The term IVD is widely used as a generic description for medical laboratory
(pathology) testing equipment and associated consumables.
Global market segments for in vitro diagnostic and POCT devices
Point-of-care testing comprises approximately 25% of the global IVD market and is worth
approximately US$11.5 billion annually (based on 2007 data). Annual growth in POCT is
around 7% to 8%, nearly twice that of the total IVD market. The main POCT applications
to date have been in those areas where demands are not well met by laboratory testing,
with diabetes control and pregnancy determination using personal (home use) testing units
being very popular. The Roche company commands approximately 20% of the POCT
market.

Some estimates of the market segmentation for IVD’s can be found at a variety of internet
sites. Most require payment of many dollars to obtain a “market report”, so current
accurate information is not readily available.
The chart below obtained from Medscape provides an estimate of the global IVD market
( http://www.medscape.com/viewarticle/584399_5 ). The diagram represents the global
IVD market of US$45 billion in 2007, including POCT at $11.5 billion (or approximately
25% of the total).
The above chart (which does not include an exact percentage distribution), does give
some appreciation of the relative proportion of the various pathology disciplines in
comparison to POCT.
In the same manner as the distribution of IVD products can be proportioned across the
traditional pathology disciplines, POCT may also be categorised or segmented on a

discipline or test type basis. Again, current data regarding the segmentation of POCT is
difficult to obtain.
One estimate which describes the POCT market segmentation for Western Europe is
available in the journal European Hospital at ( http://www.european-
hospital.com/en/article/7895-POCT_brings_values.html ). A copy of this article is also
available in the topic library. The chart below has been redrawn from the data provides in
the European Hospital article.
European POCT market segments, 2009
7% 7%
7%
9%
38%
Diabetes
Infectious
10%
Fertility
Haematology
Cardiac
22% Coagulation
Blood gas & electrolytes
Segments of the Western European POC market by discipline in 2009
A useful feature provided by this type of information, is an understanding of the relative

growth and application of POCT within the various disciplines. The relative data provides
some insight into:
 the largest proportion is related to diabetes, largely due to personal testing and
home monitoring
 infectious disease screening is becoming important for public health assessment,
HIV, influenza (various types), bacterial infections, etc

 fertility; includes home testing for pregnancy (beta hCG) and ovulation (FSH)
 blood gas and electrolytes; very important but largely applicable to professional
use within critical care centres.
Supplementary reading
 M Shephard etal. Guidelines and recommendations for the quality assured

conduct of point-of-care testing for infectious diseases and drugs of abuse in
Australia. Aust J Med Sci 2012; 33: 143.
 JA DuBois. Point-of-care testing recognised as a distinct domain.

POC 2004; 3: 20.
 JF Dooley. Point-of-care diagnostic testing markets.

POC 2009; 8: 154.

Topic 9A
Analytical considerations for bench top devices
blood gas and acid-base measurement
 Learning objectives for topic 9A
 Overview of topic 9A
 Blood gas and metabolite analysis - definitions
 Electrode technology for blood gas and metabolite measurement
 Bench top and hand-held devices for blood gas and metabolites
 Special sampling requirements for blood gas and ionised (free) calcium
 Bench top devices for “general biochemical analysis”
 Comparison of POCT with central laboratory testing
 Co-oximetry and haemoglobin parameters
 Suggested reading for topic 9A
9A.1 Learning objectives for topic 9A

 Have a general understanding of the principles of electrochemistry.
 Be able to describe in general terms the operating principles of a reference
electrode, a pH (hydrogen ion) electrode, a potassium electrode, a sodium
electrode and a pCO2 electrode.

 Be able to describe in general terms to concept of polarography and the
operating principles of an oxygen electrode (pO2 electrode).
 Know the special sampling and handling requirements for acid-base and
blood gas analysis.
 Know the special sampling and handling requirements for the measurement
of ionised calcium.
 Be able to describe the possible differences between results obtained using
a POCT device and a traditional laboratory method (particularly the different
results which may be obtained for blood and plasma glucose, and for sodium
measured by direct and indirect potentiometric methods).
 Be able to describe the clinical application for acid-base, blood gas,
electrolyte, metabolite and co-oximetry measurements.
9A.2 Overview of topic 9A
Bench top POC analysers have generally been developed with the same goals as
hand-held devices. They are often used by personnel with limited technical training,
have a user-friendly interface and inbuilt quality management facilities.
Blood gas analysers represent one of the first forms of POCT device. They were
developed as a consequence of the Danish poliomyelitis epidemic by Poul Astrup
and his colleagues in association with the Radiometer company. The development
of the “Astrup Trolley” and its introduction by Radiometer in 1959 was a significant
event in the history of blood gas and acid-base analysis. The “Astrup Trolley” was a
portable blood gas analyser which used small samples of arterialised capillary blood
collected into two or three heparinised glass capillary tubes. The Astrup analyser
used a micro pH electrode for the measurement of blood pH, and the pH of two sub-
samples which had been equilibrated at two known pCO2 values. From these three
measurements the acid-base parameters can be calculated
Current blood gas and electrolyte analysers use electrochemical sensors

(electrodes) that rely on potentiometric or amperometric measurement of the
particular analyte. The analytical principles of these electrodes are essentially the

same as those used by Astrup, but technological advances have enabled electrodes
to be constructed in many different configurations. This has resulted in different
analyser types including single-use hand-held devices and larger multi-use bench
top units.
Metabolite components such as glucose or lactate are usually measured with

enzyme biosensors, where the product(s) of the particular measurement reaction
interact with an electrochemical sensor to establish a measurement signal.
The bench top analysers typically used in critical care facilities generally have the
largest test menu which may include: blood gas and acid-base parameters;
metabolites such as glucose, lactate, urea, creatinine, bilirubin; electrolytes such as
sodium, potassium, chloride, ionised calcium; and co-oximetry with the assessment
of haemoglobin derivatives.
POCT devices of all types, whether hand-held or bench top, may produce results
which differ from traditional laboratory procedures. It is therefore essential that
POCT managers and operators understand the basis for any difference and can
advise clinicians accordingly. These differences do not necessarily mean that a
particular testing system is incorrect, as in many circumstances both results are
actually correct as they are measuring a different property of the particular biological
sample or measurand (analyte). Understanding these differences and why they
occur is of utmost importance, as different reference intervals and clinical
interpretations may apply.
9A.3 Blood gas and metabolite analysis
Descriptions of blood gas, pH, acid-base and associated physiological metabolites

are well presented in clinical chemistry (clinical biochemistry) texts and journals.
Both the physiological and analytical aspects of these analytes are required for the
understanding of many disorders of renal, respiratory and cardiac function.
The terminology of the actual blood gas components and the actual analytes
considered as metabolites may vary from one situation to another, but the basic

concepts and physiology are important for an informed understanding of many
pathophysiological processes.
Definitions:
 blood gas; in physiology this usually refers to the partial pressure of oxygen
(pO2) and carbon dioxide (pCO2) in blood
 acid-base (balance); the relationship between acids and bases in the body
required to maintain a homeostatic balance. Acid-base balance is usually
determined by the measurement of pH and pCO2, with the calculation of
bicarbonate (HCO3- ) and/or one of the derived forms of bicarbonate
 pH; a measure of the degree of acidity or alkalinity. A chemical term
specifically defined as the negative logarithm of the hydrogen ion
concentration ( pH = - log [H+] )
 metabolite; a substance produced by metabolism. In the context of blood
gas and acid-base analysis, metabolite usually refers to substances which
are measured in association with the blood gas components on the same
instrument (for example, glucose, lactate, creatinine, urea). In some
situations or publications, the electrolytes sodium, potassium, chloride and
calcium may also be included (incorrectly) in this description.
 electrolyte; a chemical term which refers to a substance which ionises when
dissolved in a suitable solvent (water). In the current context, electrolyte
usually refers to substances such as sodium (Na+), potassium (K+), chloride
(Cl-), calcium (Ca++), magnesium (Mg++), bicarbonate (HCO3-).
 pCO2; partial pressure of carbon dioxide measured in defined units (either
mmHg or kilopascal (kPa))
 pO2; partial pressure of oxygen measured in defined units (either mmHg or
kilopascal (kPa))
 bicarbonate (HCO3-); the ionic component formed by “neutralising” on of the
two hydrogen ions in carbonic acid. The major “metabolic” component or
base in acid-base studies ( H2CO3 + H2O ↔ HCO3- + H3O+ )
 Co-oximetry; the relative determination of blood concentrations of
oxyhemoglobin, carboxyhemoglobin (carbon monoxide haemoglobin, CO),

methemoglobin and reduced hemoglobin by spectrophotometric
measurement.

 LA Kaplan and AJ Pesce. Clinical Chemistry, theory, analysis and correlation.
Acid-base control and acid-base disorders, Chapter 29, 5th edition 2010.
 Tietz, edited by CA Burtis, ER Ashwood and DE Burns. Fundamentals of
Clinical Chemistry. Physiology and disorders of water, electrolyte, and acid-
base metabolism, Chapter 35, 6th edition, 2008.
 National Academy of Clinical Biochemistry, Evidence based practice for Point-
of-Care Testing. Chapter 5, Critical care.
http://www.aacc.org/members/nacb/LMPG/OnlineGuide/PublishedGuidelines/p
oct/Pages/poctpdf.aspx
 Lab Tests Online (Australia). http://www.labtestsonline.org.au
 Radiometer, acute care testing. A number of excellent articles available on
this internet site, search index at: http://acutecaretesting.org
 RER Ashwood etal. Temperature correction of blood gas and pH
measurements. Clin Chem 1983; 29: 1877.
 TJ Morgan. The Stewart approach – One clinician’s perspective. Clin
Biochem Rev 2009; 30: 41. Article available in topic library.
9A.4 Electrode technology for blood gas and metabolite measurement
In a similar manner as described above for blood gas and acid-base physiology and
pathophysiology, detailed analytical procedures describing the measurement of
metabolites and electrolytes are well presented in clinical chemistry (clinical
biochemistry) texts and journals. For POCT devices, the testing technology is
based on ion selective electrode procedures.


 LA Kaplan and AJ Pesce. Clinical Chemistry, theory, analysis and
correlation. Electrochemistry: Principles and measurements , Chapter 12, 5th
edition 2010.
 Tietz, edited by CA Burtis, ER Ashwood and DE Burns. Fundamentals of
Clinical Chemistry. Electrochemistry and chemical sensors, Chapter 5, 6th
edition, 2008.
 U Oesch, D Ammann, W Simon. Ion-Selective Membrane Electrodes for
Clinical Use. Clin Chem 1986;32:1448.
 EJ Fogt. Continuous ex vivo and in vivo Monitoring with Chemical Sensors.
Clin Chem 1990;36:1573.
 CL Morgan, DJ Newman and CP Price CP. Immunosensors: Technology
and Opportunities in Laboratory Medicine. Clin Chem 1996;42:193.
 JE Pearson, A Gill and P Vadgama. Analytical aspects of biosensors
(review). Ann Clin Biochem 2000; 37: 119.
 MN Berry etal. Enzymatic determination of sodium in serum.
Clin Chem 1988; 34: 2295.
 NM Berry etal. Enzymatic determination of potassium in serum.
Clin Chem 1989; 35: 817.
 CR Lowe etal. Biosensors. Clin Biochem Rev 1992; 13: 22.
 FJ Vanstapel and WD Lissens. Free ionised calcium – a critical survey
(review). Ann Clin Biochem 1984; 21: 339.
 BM Buckley and LJ Russell. The measurement of ionised calcium in blood
plasma (review). Ann Clin Biochem 1988; 25: 447.
 GN Bowers etal. Measurement of ionized calcium in serum with ion-
selective electrodes: A mature technology that can meet the daily service
needs. Clin Chem 1986; 32: 1437.

PowerPoint presentations
The PowerPoint presentation describing electrochemistry and ion-selective

electrodes integrates many of the aspects of electrode chemistry. These concepts
apply equally to laboratory and POCT devices. An understanding of the
measurement principals and the relationship between whole blood and plasma
samples, and direct and indirect potentiometry are required for an understanding of
result interpretation and the correct utilisation of POCT technology.
A copy of the ion-selective electrode presentation is included as part
of topic 9.
9A.5 Bench top and hand-held devices for blood gas and metabolites
POC blood gas analysers which are primarily used in a defined location (for
example, critical care unit, emergency department, operating theatre, pathology
laboratory, etc) generally fall within the bench top classification. These are often
larger analysers, even if reagents, quality control and waste disposal all utilise a
replacement cartridge system. Hand held devices such as the i-STAT analyser, are
truly portable and may be transported for use next to the patient. Both types of
system are fully described in current literature.


 C Martin. Point-of-care testing for electrolytes and acid-base imbalances.
In: A Practical Guide to Global Point-of-Care Testing. Edited by Mark
Shephard. CSIRO Publishing, 2016.
 A St John and C Ritter. Benchtop instruments for POCT. Chapter 4, in: Point-
of-care testing. Needs, opportunity and innovation. Edited by CP Price, A St
John and LJ Kricka. AACC Press, 2010.
 JH Nichols. Point of Care Testing, 2003.
Chapter 9, Point-of-care blood gas and electrolyte testing: Its effect on patient
care and the clinical laboratory
 KA Erickson and P Wilding. Evaluation of a novel point-of-care system,
the i-STAT portable clinical analyser. Clin Chem 1993; 39: 283.
 United Kingdom Department of Health. Medical and healthcare products
regulatory agency (MHRA). Management and use of IVD point of care test
devices, 2010.
http://webarchive.nationalarchives.gov.uk/20141206133224/http://www.mhra.g
ov.uk/Publications/Safetyguidance/DeviceBulletins/CON071082
 Centre for Evidence based Purchasing (CEP), NHS (National Health Service,
UK). Search index at:
http://nhscep.useconnect.co.uk/CEPProducts/Catalogue.aspx?ReportType=M
arket+review
o A guide to blood gas analysers (on the UK market in May 2006)
o Survey of fourteen blood gas analysers. CEP report 05091, January 2006.
 Radiometer, acute care testing.
Search for articles at http://acutecaretesting.org

9A.6 Special sampling requirements for blood gas and ionised (free) calcium
Arterial blood is the specimen of choice for pH and blood gas analysis (PCO2 and
PO2). A capillary blood sample may also be used to determine acid-base status
(pH, PCO2 and bicarbonate), in situations in which arterial puncture may be too
traumatic or not always required (as in neonatal or paediatric patients).
Special collection procedures are required for pH, blood gas and ionised calcium
assays. The Clinical and Laboratory Standards Institute (CLSI, formerly NCCLS)
provides standards and guidelines that define terminology and discuss performance
characteristics from pre-analytical through to post-analytical procedures. These
documents represent the best practices in the field and are created through a
consensus process involving input from industry, government and the healthcare
professions. An overview of the standards for blood gas and acid-base analysis is
given by Melissa J D Archangelo (2007) at Radiometer’s “acute care testing”
internet site (Standards provide a quality approach to blood gas analysis
http://www.acutecaretesting.org (search for “Archangelo” in search field, then
mouse “click” on the article title)).
The collection of arterial specimens with glass syringes and immediate storage in
iced water was the accepted industry standard for many years. Practice has
changed over recent years to the use of commercially prepared pre-heparinised
plastic syringes specially designed for arterial blood gas collection. Cooling and
transport in iced water is still recommended when there is a delay in analysis.
If the specimen is also to be used for ionized calcium measurement additional
issues and collection procedures need to be considered:
 the binding of calcium by proteins and small anions is influenced by pH both
in vitro and in vivo
 albumin, with approximately 30 binding sites for calcium, accounts for 80%
of protein bound calcium
 increasing the pH of a specimen in vitro increases the negative charge on
albumin leading to an increase in protein bound calcium
 decreasing the pH in vitro causes a decrease in negative charge on albumin,
decreasing albumin bound calcium with an increase in free (ionised) calcium

 free calcium changes by about 5% for each 0.1 unit change in pH
 as ionised calcium is measured at the actual pH of the patient’s blood, it is
often converted (by calculation) back to a standardized pH of 7.40
 specimens for ionized calcium should be collected and handled
anaerobically to minimize changes in pH
 syringes and/or evacuated tubes should be completely filled to minimize CO2
loss and change in pH
 in the same manner as for blood gas collection, the production of lactate
should be minimized by cooling in ice water
 ionized calcium may be measured in whole blood or plasma
 heparin is the only suitable anticoagulant for ionized calcium, but specially
prepared heparin (calcium “titrated” heparin, electrolyte “balanced” heparin)
must be used.
 “normal” heparin binds calcium and decreases the results for ionized
calcium.

 SB Blonshine. Part 1; Arterial blood collection. Part 2; Arterial blood
collection: sampling and storage. Radiometer, acute care testing, 2005.
http://www.acutecaretesting.org (search for “Blonshine” in search field).
 C Higgins. Capillary blood gases – to arterialise or not. Radiometer, acute
care testing, July 2008. http://www.acutecaretesting.org
 American Association for Respiratory Care (AARC), Clinical practice
guidelines. Capillary blood gas sampling for neonatal and paediatric
patients. http://www.rcjournal.com/cpgs/cbgscpg.html
 BD Vacutainer. LabNotes 2009; number 1. Capillary blood collection: Best
practices. http://www.bd.com/vacutainer/labnotes/Volume20Number1
 V Shah and A Taddio. Neonatal capillary blood sampling. Radiometer,
acute care testing, July 2005. http://www.acutecaretesting.org
 American Association for Respiratory Care (AARC), Clinical practice
guidelines. Sampling for arterial blood gas analysis.
http://www.rcjournal.com/cpgs/sabgacpg.html

Supplementary references (continued)
 J Toffaletti. Use of novel preparations of heparin to eliminate interference in

ionized calcium measurements: Have all the problems been solved ?
 Clin Chem 1994; 40: 508.
 J Toffaletti etal. Dry electrolyte-balanced heparinised syringes evaluated for
determining ionized calcium and other electrolytes in whole blood.
 Clin Chem 1991; 37: 1730.
 BM Buckley and LJ Russell. The measurement of ionised calcium in blood
plasma (review). Ann Clin Biochem 1988; 25: 447.
 C Sachs etal. Anomalies in pH 7.40 correction in ionised calcium analysers.
 ABTJ Boink etal. IFCC recommendations on sampling, transport and storage
for the determination of the concentration of ionized calcium in whole blood,
plasma and serum. J Automatic Chemistry 1991; 13: 235.
Exercise
Review the information presented in topic 2:

 2.4 Analytical differences between whole blood, plasma and serum
 2.5 Differences for glucose (example) – blood and plasma, capillary
and venous
 2.6 Direct and Indirect potentiometry
 2.7 Specimen collection procedures
 2.8 Pre-analytical artefacts and errors
 2.9 Preservation of specimens and timing considerations
 2.10 Anticoagulants and preservatives

9A.7 Bench top devices for “general biochemical analysis”
Clinical personnel involved with POCT are usually orientated to obtaining results
rapidly for immediate use in patient care, and may not fully appreciate the need to
comply with all the procedural and regulatory requirements associated with testing.
To assist with compliance, a number of desirable features are often incorporated
into a POCT device. These features are similar to those of laboratory based
instruments, but significant differences include portability, ease of use and
automated documentation functions.
Instrument based POCT systems can be very sophisticated and highly automated,
requiring minimal routine and preventative maintenance, with automatic calibration
functions. This functionality has been made possible through advances in reagent
stability, development of miniaturised (flow-through) electrodes and immuno-
sensors, and by the developments in microcomputers and electronics.
9A.8 Comparison of POCT with central laboratory testing
As described in Topic 2 (2.4 Analytical differences between whole blood, plasma

and serum), there may be a significant difference in results depending on what
sample is used for analysis. Blood glucose is the “classical” example of this issue.
A significant difference in glucose concentration occurs for the equivalent whole
blood and plasma samples, and also between capillary and venous blood. It is
essential to take this into consideration when comparing whole blood POCT glucose
with a laboratory value measured on venous plasma (the usual venipuncture
specimen used in a laboratory).
Many analytes give results which are dependent on the type of specimen and/or the
particular analytical method involved. These differences are often confusing and
may be difficult to explain in all circumstances. It is not possible to review every
possible combination in a course of this type, but participants should be aware of
measurement procedures which may provide apparently “conflicting” results
(particularly between a POCT device and a traditional laboratory procedure).

POCT technology which provides results on whole blood usually incorporates some
form of “filter” which separates plasma water (and dissolved substances) from solid
elements such as erythrocytes. For example, many test strips which contain
“reaction pads” incorporate a semi-permeable membrane which prevents
erythrocytes from entering the reaction matrix. Water and the dissolved reactant(s)
diffuse through the membrane to interact with the colour producing chemicals. In
this type of reaction, the plasma water is also an important component as it acts as
solvent for the overall reaction. Colour producing blood glucose test strips are a
good example of this process.
Two other important specimen and measurement related issues are the differences
between direct and indirect potentiometry, and the requirement for specific
anticoagulants in some situations (for example, blood gas, electrolyte and ionised
calcium measurements).
9A.9 Co-oximetry and haemoglobin parameters

The non-invasive measurement of arterial oxygen saturation by pulse oximetry is
widely acknowledged to be one of the most important technical advances in patient
monitoring. Unlike blood gas analysers with facilities for co-oximetry, pulse
oximeters calculate blood oxygen saturation by measuring differences in the visible
and near infrared absorbances of oxygenated and deoxygenated blood. Blood
oxygen saturation can also be calculated from blood gas and co-oximetry
information, but requires a sample of arterial blood from the patient. Co-oximetry is
the relative determination of blood concentrations of oxyhemoglobin,
carboxyhemoglobin (carbon monoxide haemoglobin, CO), methemoglobin and
reduced hemoglobin by spectrophotometric measurement. There are many clinical
situations in which the use of the “co-oximetry parameters” can assist with patient
diagnosis and monitoring.


 R Chatburn. To Co-ox or not to Co-ox, oxygen saturation. Radiometer,
acute care testing. http://www.acutecaretesting.org (Check Co-oximetry box
on left hand margin … many good references).
 RF Moran. Application of hemoglobin derivatives in STAT analysis.
Radiometer, acute care testing. http://www.acutecaretesting.org
 Y Mendelson. Pulse oximetry: Theory and applications for non-invasive
monitoring. Clin Chem 1992; 38: 1601.
 KN Olson et.al. Accident or arson: Is co-oximetry reliable for
carboxyhaemoglobin measurement post-mortem? (Clinical case study).
Clin Chem 2010; 56: 515.
9A.10 Fitness for purpose
The fundamental role of medical testing laboratories and POCT providers is to

produce test results that are fit for their purpose. A test result must have appropriate
analytical accuracy and precision in order for it to be considered fit for purpose; that
is, suitable for the clinical purpose for which the test is being used. In order to
determine whether a method is routinely producing results which are fit for purpose,
there needs to be a relevant analytical goal against which the estimated uncertainty
of measurement can be compared. Some methods have internationally agreed
analytical goals (for example; glucose, cholesterol and haemoglobin A1c), but in their
absence various approaches have been used to set goals for both inaccuracy (bias)
and imprecision. A widely used and internationally recommended concept is to
define the upper acceptable limit for imprecision as a proportion of the intra-
individual biological variation of the measurand (analyte). With correct choice of the
proportionality factor, analytical imprecision should not contribute significant
additional variation to the test result when compared with the natural variation of the
substance being measured. A similar approach to goal-setting can be used for total

analytical error (inaccuracy plus imprecision). These concepts are clearly described
in the NPAAC document Requirements for the estimation of measurement
uncertainty and in the article by White and Farrance, Uncertainty of measurement in
The direct influence of analytical imprecision and inaccuracy (bias) on laboratory and
POC measurements are critical issues for the interpretation of patient results and the
fitness for purpose of testing devices. These concepts apply equally to all
measurement devices and methods, including all POCT procedures.
When reviewing the fitness for purpose of a POCT device or laboratory method, the
following steps should be considered in the order indicated:
1. Define the quality that is required for each test (based on clinical
requirement, intra-individual biological variation or total biological variability).
Quality management begins with knowledge of the quality required of the
test procedure.
2. Ascertain the performance of the POCT device (for example, blood gas
measuring unit) or method under consideration. The estimated precision
(uncertainty) and bias of the device should be evaluated in advance of
purchase (and certainly prior to the commencement of patient testing).
3. Calculate the relevant fitness for purpose. Ideally, uncertainty of
measurement should be less than 0.5 times the intra-individual biological
variation of the particular measurand as outlined in the article by GH White
and I Farrance, Uncertainty of measurement in quantitative medical testing.
Clin Biochem Rev 2004; 25, supplement (ii): S1.
4. Choose appropriate QC and EQA to establish and maintain the stability of
the testing procedure.

9A.11 Suggested reading for topic 9A
Suggested reading
 C Martin. Point-of-care testing for electrolytes and acid-base imbalances.

 G Herd and SMA Mussaad. Point-of-care testing in the hospital setting.

Shephard. CSIRO Publishing, 2016
 A St John and C Ritter. Benchtop instruments for point-of-care testing.

Chapter 4, in: Point-of-care testing. Needs, opportunity and innovation.
Edited by Price CP, St John A and Kricka LJ. AACC Press, 2010.
 P Holloway. POCT in the intensive care unit.



Topic 9B
Clinical interpretation of Blood Gas Measurements
• Learning objectives for topic 9B

• Overview of topic 9B
• Definitions
• pH as an indicator of acidity or alkalinity
• Blood gas terminology – pH, pKa, pCO2, pO2, bicarbonate (HCO3-), other forms of
bicarbonate, etc.
• Henderson Hasselbalch equation and its application to acid/base physiology
• Acidosis and alkalosis
• Metabolic and respiratory components of acid/base balance
• Haemoglobin and oxygen saturation
• Suggested reading for topic 9B
9B.1 Learning objectives for topic 9B

• Know the definitions and terminology of physiological blood gas and acid/base
reactions including; acid, base, pH, buffer, pKa, pCO2, pO2, acidosis and alkalosis.
• Know the Henderson Hasselbalch equation and its application to acid/base
physiology.
• Be able to describe in general terms the physiological formation, excretion and
buffering of hydrogen ions.
• Be able to describe in general terms the respiratory and metabolic components of
acid/base balance.
• Be able to describe in general terms the concepts of compensated and
uncompensated, acidosis and alkalosis.
• Have an understanding of the “normal” physiological values for blood pH, pCO2,
pO2 and bicarbonate.
1 of 11 Ian Farrance
9B.2 Overview of topic 9B
In the simplest chemical definition, an acid is a substance that provides hydrogen ion(s)
(H+), whereas a base is a substance which accepts a hydrogen ion(s). Both acids and
bases are further defined by their affinity for H+. Acid/base chemistry deals with the
interactions involving the transfer of hydrogen ions. Resistance to changes in hydrogen
ion concentration are provided by other chemical species called buffers. Many substances
present in blood and tissues act as buffers, the most important include substances such as
bicarbonate, phosphate, proteins (in general) and haemoglobin.
The bicarbonate – carbonic acid buffer system occupies a dominant position among the
many buffer systems present in living organisms because of its many links with
physiological regulatory mechanisms. Carbonic acid arises from the hydration of dissolved
CO2 and dissociates to yield bicarbonate and hydrogen ions. These interactions achieve
equilibrium extremely rapidly, permitting a description of the system by relatively simple
chemical expressions. One of the major end products of metabolism, gaseous CO2
produced by the tissues, is soluble in water. In this dissolved state, the concentration of
dissolved CO2 is proportional to the partial pressure of CO2 (or pCO2), with the interchange
of dissolved and gaseous forms being assisted by the enzyme carbonic anhydrase.
The interactions between all the major components which regulate the physiological
acid/base state are well described in Clinical Biochemistry texts as outlined in the
suggested reading for this topic.
Understanding the clinical utility of blood gas and acid/base metabolism is based on an
understanding of the bicarbonate – carbonic acid buffer system and the changes which
occur in disease. These clinical situations are generally described by changes in the
relative proportions of the components which comprise the (blood) bicarbonate – carbonic
acid buffer system and the corresponding level of hydrogen ion or pH
CO2(gas) ↔ CO2 (dissolved) ↔ H2CO3 (with H2O) ↔ H+ + HCO3−
In addition to components of the CO2/HCO3− buffer system, blood gas analysis (or
acid/base analysis) usually includes the measurement of pCO2 (partial pressure of CO2)
and pO2 (partial pressure of O2). The assessment of oxygen status and the relative
proportion of various haemoglobin derivatives (as part of CO-oximetry analysis) in
association with blood gas analysis, provide additional information with regard to
respiratory function and its influence on patient outcome.
9B.3 Definitions
Much of the terminology and biochemical relationships which underlie blood gas and acid-
base interactions are derived from basic chemistry. The concept of pH and its
mathematical definition based on hydrogen ion concentration, the relationship of pH to
acids and bases expressed by the Henderson Hasselbalch equation and the calculation of
the various “blood gas” parameters, are essentially exercises in basic chemistry.
Definitions:
• blood gas; in physiology this usually refers to the partial pressure of oxygen (pO2)
and carbon dioxide (pCO2) in blood
• acid-base (balance); the relationship between acids and bases in the body required
to maintain a homeostatic balance. Acid-base balance is usually determined by
the measurement of pH and pCO2, with the calculation of bicarbonate (HCO3- )
and/or one of the derived forms of bicarbonate
• pH; a measure of the degree of acidity or alkalinity. A chemical term specifically
defined as the negative logarithm of the hydrogen ion concentration ( pH = - log
[H+] )
• metabolite; a substance produced by metabolism. In the context of blood gas and
acid-base analysis, metabolite usually refers to substances which are measured in
association with the blood gas components on the same instrument (for example,
glucose, lactate, creatinine, urea). In some situations, the electrolytes sodium,
potassium, chloride and calcium may also be included (incorrectly) in this
description.
• electrolyte; a chemical term which refers to a substance which ionises when
dissolved in a suitable solvent (water). In the current context, electrolyte usually
refers to substances such as sodium (Na+), potassium (K+), chloride (Cl-), calcium
(Ca++), magnesium (Mg++), bicarbonate (HCO3-).
• pCO2; partial pressure of carbon dioxide measured in defined units (either mmHg
or kilopascal (kPa))
• pO2; partial pressure of oxygen measured in defined units (either mmHg or
kilopascal (kPa))
• bicarbonate (HCO3-); the ionic component formed by “neutralising” on of the two
hydrogen ions in carbonic acid. The major “metabolic” component or base in acid-
base studies ( H2CO3 + H2O ↔ HCO3- + H3O+ )
• CO-oximetry; the relative determination of blood concentrations of oxyhemoglobin,
carboxyhemoglobin (carbon monoxide haemoglobin, CO), methemoglobin and
reduced hemoglobin by spectrophotometric measurement.
• Henderson Hasselbalch equation; an equation that defines the relationship between
pH, an acid and its conjugate base. In medicine, the acid is usually carbonic acid (as
a function of CO2 gas) and bicarbonate ion (HCO3-).
( http://en.wikipedia.org/wiki/Henderson%E2%80%93Hasselbalch_equation )
9B.4 pH as an indicator of acidity or alkalinity
The acidity (or alkalinity) of a solution such as blood ( whole blood, plasma or serum) or a
chemical buffer, is often expressed as pH. It is a convenient notation for describing
relatively low levels of acidity. pH is a measure of hydrogen ion concentration ( [H+] ) and
is calculated using the formula pH = − log10 [H+] .
The pH scale covers the range 0 to14. A pH of 7.0 is “neutral”, a pH less than 7 is acidic,
and a pH greater than 7 is basic. As the pH scale is logarithmic (see definition and
equation above), each whole pH value below 7 is ten times more acidic than the next
higher value. For example, pH 4 is ten times more acidic than pH 5 and 100 times (10
times 10) more acidic than pH 6. The same holds true for pH values above 7, each of
which is ten times more alkaline (basic) than the next lower whole value. For example, pH
10 is ten times more alkaline than pH 9 and 100 times (10 times 10) more alkaline than pH
8. Pure water is “neutral”. When chemicals are mixed with water however, the mixture
can become either acidic or basic. Examples of acidic substances are vinegar and lemon
juice. Milk of magnesia (magnesium hydroxide) and ammonia solution are examples of
basic substances. Blood (serum, plasma) is normally very slightly alkaline at pH 7.4.
Acid Neutral Base
pH 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 pH
E
x
pure
coca tap
a water, deter- deter-
Gastric cola, water, soap, house-
m 1M 0.1M sea gents, gents, 0.1M 1M
juice lemon wine coffee saliva, baking hold
p HCl HCl water tooth- washin NaOH NaOH
juice, cows soda cleaner
l Blood paste g soda
vinegar milk
(7.4)
e
s
9B.5 Blood gas terminology – pH, pKa, pCO2, pO2, bicarbonate (HCO3-), other
forms of bicarbonate, etc.
Exercise
• Review the definitions as outlined in item 9B.3 (above).

• Review the information provided in Topic 2 – Pre-Analytical variables.
• Review the information provided in Topic 9A – Analytical considerations for
bench top devices, blood gas and acid-base measurement.
Normal cell metabolism depends on the maintenance of blood pH within very narrow limits
(reference interval approximately 7.35-7.45). Even relatively mild deviations outside this
normal pH range can have deleterious effects, including reduced oxygen delivery to
tissues, electrolyte disturbances and changes in heart muscle contractility. Survival is rare
if blood pH falls below 6.8 or rises above 7.8. The problem for the body is that normal
metabolism is associated with continuous production of hydrogen ions (H+) and carbon
dioxide (CO2), both of which tend to reduce pH. The mechanism which overcomes this
problem and serves to maintain normal blood pH (that is, to preserve acid-base
homeostasis) is a complex series of interactions which involve chemical buffers in blood,
the red cells (erythrocytes) which circulate in blood and the function of three organs (the
lungs, kidneys and brain).
The acid-base status in the body is described in terms of arterial pH, arterial pCO2 and
bicarbonate (or one of the related forms of bicarbonate). These are all related through the
Henderson Hasselbalch equation …
pH = pK + log ( [HCO3-] / [H2CO3] )
( where H2CO3 is directly proportional to pCO2 )
The most usual method for determining acid-base parameters is to measure blood pH and
pCO2 (in an anaerobic arterial blood sample), and calculate bicarbonate and/or one of the
related forms of bicarbonate. At the same time, blood pO2 and other constituents may
also be measured (refer to topic 9A). The use of various forms of bicarbonate to more
clearly define the “metabolic” component of acid-base metabolism may have added more
complexity instead of providing simplification as intended. The different forms of
“bicarbonate” include: bicarbonate concentration (or actual bicarbonate), total bicarbonate,
standard bicarbonate, base excess and buffer base.
The bicarbonate concentration or actual bicarbonate is the quantity of bicarbonate in unit

volume of blood plasma. This is the form in which most carbon dioxide (90 %) is
transported in blood. This parameter is obtained by calculation using the Henderson
Hasselbalch equation, using measurements of pH and pCO2.
Exercise
Using appropriate references, determine the definitions of the major alternate forms
of “bicarbonate” which provide an indication of the “metabolic” component of acid-
base status:
o standard bicarbonate
o total bicarbonate
o buffer base
o base excess

• Tietz, edited by CA Burtis, ER Ashwood and DE Burns. Fundamentals of
Clinical Chemistry. 6th edition, 2008.
Chapter 24, Electrolytes and blood gasses
Chapter 35, Physiology and disorders of water, electrolyte, and acid-base
metabolism.
• National Academy of Clinical Biochemistry, Evidence based practice for
Point-of-Care Testing. Chapter 5, Critical care.
http://www.aacc.org/members/nacb/LMPG/OnlineGuide/PublishedGuidelines/
poct/Pages/poctpdf.aspx
• Acid base homeostasis. http://en.wikipedia.org/wiki/Acid-base_homeostasis
• The Carbonic-Acid-Bicarbonate Buffer in the Blood
http://en.wikipedia.org/wiki/Bicarbonate_buffering_system … and …
http://www.chemistry.wustl.edu/~edudev/LabTutorials/Buffer/Buffer.html
http://www.chemistry.wustl.edu/~courses/genchem/Tutorials/Buffers/carbonic.htm
9B.6 Henderson Hasselbalch equation and its application to acid/base physiology
Three of many good articles available through Radiometer, acute care testing.
http://acutecaretesting.org (search by title and scroll down through articles).
1. C Higgins. An introduction to acid-base balance in health and disease, June

2004.
2. C Higgins. Parameters that reflect the carbon dioxide content of blood,
October 2008.
3. J Kofstad. All about base excess – to BE on not to BE, July 2003.
9B.7 Acidosis and alkalosis
9B.8 Metabolic and respiratory components of acid/base balance
The PowerPoint presentation describing blood gas and acid/base status form part of
the information provided for topic 9B. An understanding of the relationship between
the various acid/base and blood gas parameters are required for result interpretation
and the correct utilisation of POCT technology.
A copy of the acid-base balance presentation is included as part

of topic 9B.
9B.9 Haemoglobin and oxygen saturation
All cells require a continuous supply of oxygen and one of the principal life-preserving
functions of blood is the delivery of oxygen present in inspired air from lungs to tissues.
An adequate supply of oxygen to tissues thus depends on adequate oxygenation of blood.
Blood oxygen is carried in two forms. Approximately 98 % is carried bound to
haemoglobin, but a very small amount is transported as oxygen dissolved in blood plasma.
The two forms are in chemical equilibria and the haemoglobin bound oxygen acts as a
reservoir of “available” oxygen. It is the small amount of dissolved unbound oxygen that
determines pO2. Although pO2 reflects only a very small portion of the total oxygen in
arterial blood, it is the major determinant of the amount of oxygen carried by hemoglobin
and thus the total amount of oxygen carried by blood. The relationship between partial
pressure of oxygen in blood (pO2) and hemoglobin oxygen saturation (sO2) is described by
the oxygen dissociation curve as shown below. When arterial pO2 is within the normal
range of approximately 80 to 100 mmHg (approximately 10 to 12 kPa), arterial hemoglobin
is almost fully saturated (sO2 > 97 %). Providing that blood contains a normal amount of
hemoglobin, an arterial sO2 of approximately 97 % indicates adequate oxygenation.
However, if arterial pO2 is reduced, hemoglobin binds significantly less oxygen and blood
becomes less oxygenated. It is evident from examination of the linear portion of the
oxygen dissociation curve that only a slight reduction in arterial pO2, below approximately
60 mmHg (or 8 kPa) is associated with a significant fall in sO2 and a marked reduction in
the oxygen content of blood. Respiratory failure is defined by an arterial pO2 of less than
60 mmHg (or 8kPa).
The sigmoid shape of the oxygen dissociation curve is a result of the cooperative binding
of oxygen to the four polypeptide chains of haemoglobin. Cooperative binding is a
characteristic of hemoglobin, which provides a greater ability to bind additional oxygen
after a subunit has already bound an oxygen. As a consequence of this type of binding,
hemoglobin is most attracted to oxygen when three of the four polypeptide chains have
already bound to oxygen.
Factors which may influence oxygen binding to haemoglobin include:
• Temperature − increasing the temperature disrupts the bond between oxygen and
haemoglobin. This increases the amount of free oxygen and deoxygenated
haemoglobin, and decreases the concentration of oxyhemoglobin (or oxygen
bound to haemoglobin). The dissociation curve shifts to the right.
• pH − a decrease in pH (increase in [H+]) by the addition of carbon dioxide or other
acids, causes more oxygen to be released from haemoglobin as oxygen pressure
(pO2) increases. The dissociation curve shifts to the right.
• Organic phosphates − 2,3-diphosphoglycerate (DPG) is the primary organic
phosphate in erythrocytes. It binds with greater affinity to deoxygenated
hemoglobin (for example, when the erythrocyte is near respiring tissue) than it
does to oxygenated hemoglobin (for example, in the lungs). In bonding to partially
deoxygenated hemoglobin, it allosterically causes release of the remaining oxygen
bound to that hemoglobin. This binding, enhances the ability of erythrocytes to
release oxygen near tissues that need it most. DPG binds to hemoglobin causing
rearrangement and decreasing the affinity of hemoglobin to bind oxygen. The
curve shifts to the right.
• Carbon monoxide − haemoglobin binds with carbon monoxide (CO) 240 times
more readily than with oxygen. The presence of carbon monoxide on one of the
four haem sites causes the oxygen on the other haem sites to bind with greater
affinity. This makes it difficult for the haemoglobin to release the oxygen to the
tissues and has the effect of shifting the curve to the left. With an increased level
of carbon monoxide (with the formation of carboxyhaemoglobin), a person may
suffer from severe hypoxemia while maintaining a normal pO2.
• Left shift of the oxygen-haemoglobin dissociation curve is a sign of increased
affinity of haemoglobin for oxygen (as in the lungs). Similarly, right shift shows
decreased affinity, as would occur with an increase in body temperature, hydrogen
ion concentration, DPG or carbon dioxide concentration.

• Oxygen haemoglobin dissociation curve.
http://en.wikipedia.org/wiki/Oxygen-haemoglobin_dissociation_curve
• R Chatburn. To Co-ox or not to Co-ox, oxygen saturation. Radiometer,
acute care testing. http://acutecaretesting.org (Check Co-oximetry box on
left hand margin … many good references).
• RF Moran. Application of hemoglobin derivatives in STAT analysis.
Radiometer, acute care testing. http://acutecaretesting.org
• KN Olson etal. Accident or arson: Is co-oximetry reliable for
carboxyhaemoglobin measurement post-mortem ? (Clinical case study).
Clin Chem 2010; 56: 515.
• C Higgins. Blood oxygenation and spurious hypoxemia, January 2010.
Radiometer, acute care testing. http://acutecaretesting.org
9B.10 Suggested reading for topic 9B
Suggested reading
• C Martin. Point-of-care testing for electrolytes and acid-base imbalances.

• KM Hurth etal. POCT in the operating room.

• LA Kaplan and AJ Pesce. Clinical Chemistry, theory, analysis and

correlation. Acid-base control and acid-base disorders, Chapter 29, 5th
edition 2010
Topic 10 – Diabetes, Glucose Metabolism

and Haemoglobin A1c
 Overview of topic 10 − diabetes and glucose metabolism
 Glucose metabolism
 Pathophysiology and diabetes
 Types of diabetes and importance of glucose control
 Home monitoring of “blood” glucose
 The role of haemoglobin A1c (HbA1c)
 Analytical requirements for accuracy and precision
 DCCT and related control and monitoring requirements
 Suggested reading for topic 10

 Have a general understanding of the intermediary metabolism of glucose and be
able to describe the major mechanisms which control blood glucose (including the
action of insulin, C-peptide, glucagon, cortisol, somatotropin (growth hormone)).
 Be able to describe in general terms the biochemistry of diabetes mellitus.
 Understand the criteria for the diagnosis and classification of diabetes mellitus,
acute and chronic complications, and appropriate laboratory investigations for the
diagnosis and monitoring of diabetes.
 Be able to describe in general terms the biochemistry of hypoglycaemia, how
hypoglycaemia can be identified taking into consideration the limitations of POCT
devices for the assessment of hypoglycaemia.
 Understand the importance of accurate and precise monitoring of blood glucose,
HbA1c and albumin / creatinine ratio as indicators of diabetes control.
 Understand the advantages, limitations and fitness for purpose of POCT devices
for monitoring diabetes mellitus.
10.2 Overview of topic 10 − diabetes and glucose metabolism
The main focus of this topic is directed towards the important clinical aspects of glucose
metabolism and diabetes mellitus (usually just referred to as “diabetes”).
This includes:
 a basic understanding of the intermediary metabolism of glucose and related
carbohydrates
 the mechanisms which control the biochemical and physiological interactions of
glucose
 the role of key hormones such as insulin, C-peptide, cortisol, growth hormone, etc
 hyperglycaemia and hypoglycaemia
 the metabolic interactions between glucose, fatty acids, electrolytes, pH and acid base
metabolism
 protein glycation and the interaction of glucose with tissues and tissue proteins
 glycated haemoglobin(s) and HbA1c
 the pathophysiology of diabetes mellitus and the types of diabetes (including
gestational diabetes)
 symptoms of diabetes
 procedures and criteria for the diagnosis of diabetes
 treatment and monitoring of diabetes
 measurement and monitoring of blood (plasma) glucose and HbA1c: self
monitoring, ward monitoring, clinic monitoring, laboratory monitoring.
Even though this topic provides a clinical focus, the analytical aspects discussed
previously (particularly in module 1) form an integral component of all POCT as they
provide the important link between analytical measurement and clinical interpretation.
The best descriptions of glucose metabolism and the metabolic aspects of diabetes can be
found in standard Clinical Biochemistry textbooks or speciality texts on diabetes. Journal
articles extend what is available in standard texts and discuss new or controversial
concepts such as the biochemical diagnosis of diabetes, the role of HbA1c,
standardisation procedures and estimated average glucose.
The WHO report which describes the “definition and diagnosis of diabetes mellitus and
intermediate hyperglycaemia” also provides a good general overview of diabetes. It is the
WHO definitions and diagnostic criteria which are followed in Australia. The American
Diabetes Association (ADA) provides similar information, but differs in some respects to
WHO. There is often confusion as to which criteria apply as there are many cross
reference to both WHO and ADA in the clinical literature. An appendix within the 2006
WHO report provides a table of the differences between the two groups. By
consensus however, Australia follows the WHO criteria.
Also of current interest, is the general agreement that HbA1c can be used for the
diagnosis of diabetes mellitus provided that the assay used is of suitable analytical
performance.
10.3 Glucose metabolism
Concise descriptions of glucose metabolism and its hormonal control are provided in
textbooks of Clinical Biochemistry. The metabolic interactions and regulatory processes
which occur as part of normal glucose metabolism form the basis for an informed
understanding of the clinical aspects of diabetes mellitus.
Suggested reading
 RF Dodds. Diabetes mellitus, chapter 38. LA Kaplan and AJ Pesce. Clinical

Chemistry, theory, analysis and correlation, 5th edition 2010.
10.4 Pathophysiology and diabetes
10.5 Types of diabetes and importance of glucose control
The PowerPoint presentation describing glucose and diabetes provides an overview

of glucose metabolism, the pathophysiology of diabetes and the procedures for
monitoring blood glucose and diabetes control.
A copy of the glucose and diabetes presentation is included as part

of topic 10.
The diagnosis of diabetes often relies on the results obtained from an oral glucose
tolerance test (OGTT or GTT) or from the accurate measurement of HbA1c. The
performance and interpretation of a GTT are defined by WHO. The performance criteria
include preliminary patient preparation (important factors which are often neglected),
glucose dose considerations, specimen collection, analytical considerations and
interpretation. The appendix to the 1999 WHO report contains a good description of these
performance criteria (see supplementary references below).
The WHO recommended the adult dose of glucose to be used in a GTT as 75 gram of
anhydrous glucose (or equivalent). However, the usual form of commercial crystalline
glucose is glucose monohydrate (that is, glucose with one molecule of water included in
the crystal structure). This is still a “pure” form of glucose, but the increased formula
weight needs to be considered. To obtain the required 75g anhydrous glucose as defined
for the GTT, 82.5 of glucose monohydrate is required. This may already have been
considered if commercial “glucose drinks” are used, but one should always check the
bottle contents and product description. Some brands of commercial “glucose drink” may
actually state the contents as “75 gram of glucose”, but actually only contain 75 gram of
glucose monohydrate (not the required 75 gram of anhydrous glucose or 82.5 gram of
glucose monohydrate). Always check the glucose dose carefully !
Exercise
Review the appendices contained within the 1999 and 2006 WHO reports (which
outline GTT performance criteria and result interpretation). Important information
includes …
 recommendations not to use some types of POCT device for GTT glucose
measurement
 patient preparation prior to the performance of a GTT
 specimen collection and crucial performance criteria
 GTT result interpretation
(provided as part of the topic library)
WHO 1999. Definition, diagnosis and classification of diabetes mellitus and its
complications.
WHO 2002. Laboratory diagnosis and monitoring of diabetes mellitus.
WHO 2006. Definition and diagnosis of diabetes mellitus and intermediate
hyperglycaemia.

 DB Sacks. Carbohydrates, chapter 22. Tietz, Fundamentals of Clinical
Chemistry, edited by CA Burtis, ER Ashwood and DE Burns, 6th edition,
2008.
 AK Aarsand etal. Diagnosis and management of diabetes mellitus. Chapter
6, in National Academy of Clinical Biochemistry, Evidence based practice for
Point-of-Care Testing. https://www.aacc.org/science-and-research/practice-
 Diabetes Control and Complications Trial (DCCT), links to DCCT information.

http://www.bsc.gwu.edu/bsc/oneproj.php?pkey=5 … and …
http://diabetes.niddk.nih.gov/dm/pubs/control/index.htm
 Diabetes in the United Kingdom, 2004.

http://www.diabetes.org.uk/Documents/Reports/in_the_UK_2004.doc
 National evidence based guideline for blood glucose control in type 2

diabetes, NHMRC 2009.
http://www.nhmrc.gov.au/_files_nhmrc/file/publications/synopses/di19-
diabetes-blood-glucose-control.pdf
 WHO 1999. Definition, diagnosis and classification of diabetes mellitus and

its complications.
http://whqlibdoc.who.int/hq/1999/WHO_NCD_NCS_99.2.pdf
 WHO 2006. Definition and diagnosis of diabetes mellitus and intermediate

hyperglycaemia.
http://www.who.int/diabetes/publications/Definition%20and%20diagnosis%20
of%20diabetes_new.pdf
 Schneider H, Shaw J, Zimmet P. Guidelines for the Detection of Diabetes

Mellitus – Diagnostic Criteria and Rationale for Screening. Clin Biochem
Revs 2003; 24: 77-80.
 Jones GRD et.al.. RCPA and AACB working party report. Position statement
on impaired fasting glucose. Clin Biochem Rev 2008; 29: 113.
 National (Australian) evidence based guidelines relating to diabetes are

available through the National Health and Medical Research Council internet
site.
o National Evidence Based Guideline for Patient Education in type 2
Diabetes.
o National Evidence Based Guideline for Case Detection and Diagnosis of
Type 2 Diabetes.
o National Evidence Based Guideline for Diagnosis, Prevention and
Management of Chronic Kidney Disease in Type 2 Diabetes.
o National Evidence Based Guideline for Blood Glucose Control in Type 2
Diabetes.
o National Evidence Based Guideline for the Primary Prevention of Type
2 Diabetes.
To search - copy and paste title into your web browser
10.6 Home monitoring of “blood” glucose
Exercise
 Review the information provided in Topic 2 – Pre-Analytical variables.

 Review the information provided in Topic 3 – Analytical considerations for
hand held devices.

 M Shephard, J Shaw and P Zimmet. Point-of-care testing for diabetes:
Haemoglobin A1c. In: A Practical Guide to Global Point-of-Care Testing.
 S Coster etal. Monitoring blood glucose control in diabetes mellitus: A

systematic review. Health Technology Review 2000; 4 (12):1 to 93.
 ES Kilpatric etal. The effect of haemolysis on blood glucose measurement.

Diabet Med 1995; 12: 341.
 RF Louie etal. Point of care glucose testing, effect of critical care variables,
influence of reference instruments and a modular glucose meter design.
Arch Pathol Lab Med 2000; 124: 257.
 LMC Welschen etal. Self-monitoring of blood glucose in patients with type 2

diabetes who are not using insulin. Diabetes Care 2005; 28: 1510.
10.7 The role of haemoglobin A1c
Haemoglobin A1c (HbA1c) plays an important role in the diagnosis of diabetes and for
monitoring glucose stabilisation in its treatment. HbA1c is now approved in some
countries for the diagnosis of diabetes (Australia, USA and United Kingdom). The
important role in monitoring glucose stabilisation was clearly demonstrated in the Diabetes
Control and Complications Trial (DCCT). The glycation mechanism(s) which occur at
increased blood glucose levels are described in Clinical Biochemistry textbooks and the
attached PowerPoint lecture set.
An important consideration with the use of HbA1c in both the diagnosis and monitoring of
diabetes, is the analytical precision and degree of systematic error (bias) with which it can
be measured. Due to the narrow reference range for HbA1c and the small deviations
which are considered clinically important, a high analytical precision with a low bias is
required for its measurement. The effect of analytical precision on clinical interpretation
has been widely discussed. To provide a clear analytical distinction between the blood
HbA1c decision levels of 53 mmol/mol (“old” NGSP units of 7.0%) and 64 mmol/mol (“old”
NGSP units of 8.0%), an analytical uncertainty (method uncertainty) expressed statistically
as a coefficient of variation (CV) of less than 3% is required. The relationship of analytical
precision, biological variation and reference intervals to clinical interpretation has also
been discussed in Module 2.
Units for HbA1c. A recent consensus statement on the worldwide standardisation and
reporting of HbA1c measurement has recommended that: (1) method calibration should be
traceable to the International Federation of Clinical Chemistry and Laboratory Medicine
(IFCC) reference method; and (2), that Systeme International (SI) units (mmol/mol) be
used for reporting results. A concise summary of these recommendations is provided in
the article by GRD Jones et al, Change of HbA1c reporting to the new SI units, MJA 2011;
195: 45 (article available in topic library).
Suggested reading

 L Al-Ansary etal. Point-of-care testing for HbA1c in the management of

diabetes: A systematic review and metaanalysis. Clin Chem 2011; 57(4):568.
 WHO. The use of Glycated Haemoglobin (HbA1c) in the Diagnosis of

Diabetes. (See separate item for this topic).
 GRD Jones etal. Change of HbA1c reporting to the new SI system. MJA
2011; 195: 45.
 RR Little. Performance of haemoglobin A1c assay methods: Good enough ?

Clin Chem 2014; 60: 1031.

 DB Sacks. Carbohydrates, chapter 22. Tietz, Fundamentals of Clinical
Chemistry, edited by CA Burtis, ER Ashwood and DE Burns, 6th edition, 2008.
 CG Fraser and PH Petersen. Analytical performance characteristics should be
judged against objective quality specifications. Clin Chem 1999; 45: 321.
 RR Little etal. Status of Haemoglobin A1c measurement and goals for
improvement: From chaos to order for improving diabetes care. Clin Chem
2011; 57(2): 205.
 WG John etal. HbA1c standardisation: History, science and politics. Clin
Biochem Rev 2007; 28: 163. Article available in topic library.
 AK Aarsand etal. Diagnosis and management of diabetes mellitus. Chapter 6,
in National Academy of Clinical Biochemistry, Evidence based practice for
Point-of-Care Testing (see references) or https://www.aacc.org/science-and-
 Consensus committee. Consensus statement on the world wide
standardisation of HbA1c measurement. Diabetes Care 2007; 30: 2399. Or
Diabetologia 2007; 50: 2042.
 DE Burns and JC Boyd. Few point-of-care hemoglobin A1c assays meet
clinical needs. Clin Chem 2010; 56: 4.
 E Lenters-Westra and RJ Slingerland. Six of eight HbA1c POCT instruments
do not meet the generally accepted analytical performance criteria. Clin Chem
2010; 56: 44.
 E Lenters-Westra and RJ Slingerland. Three of 7 haemoglobin A1c point-of-
care instruments do not meet generally accepted analytical performance
criteria. Clin Chem 2014; 60: 1062.
 A Woodworth etal. Utilisation of assay performance characteristics to estimate
haemoglobin A1c result reliability. Clin Chem 2014; 60: 1073
Exercise
Review the information provided in Topic 3.6 – HbA1c: instrument technology and
general considerations.
10.8 Analytical requirements for accuracy and precision
Suggested reading

Glucose. In: A Practical Guide to Global Point-of-Care Testing. Edited by

Edited by Mark Shephard. CSIRO Publishing, 2016
 CG Fraser and MG Scott. Analytical performance requirements for point-of-

care testing: Glucose as an example. Chapter 12, in: Point-of-care testing.
Needs, opportunity and innovation. Edited by CP Price, A St John and LJ
Kricka. AACC Press, 2010.
10.9 DCCT and related control and monitoring requirements

 Diabetes Control and Complications Trial (DCCT), links to DCCT information.
o http://en.wikipedia.org/wiki/Diabetes_Control_and_Complications_Trial
o http://www.bsc.gwu.edu/bsc/oneproj.php?pkey=5
o http://diabetes.niddk.nih.gov/dm/pubs/control/index.htm
 TME Davis and S Colagiuri. The continuing legacy of the United Kingdom
Prospective Diabetes Study. MJA 2004; 180: 104.
http://www.mja.com.au/public/issues/180_03_020204/dav10798_fm-2.html
 United Kingdom Prospective Diabetes Study (UKPDS).

http://en.wikipedia.org/wiki/United_Kingdom_Prospective_Diabetes_Study
accuracy and precision in order for it to be considered fit for purpose; that is, suitable for
The practical problems associated with selecting appropriate quality goals (quality
specifications) are exemplified by the most widely used POCT device, the glucose meter.
The American Diabetes Association (ADA) and the National Academy of Clinical
Biochemistry (NACB) have documented this dilemma in considerable detail (National
Academy of Clinical Biochemistry (NACB). Guidelines and recommendations for
laboratory analysis in the diagnosis and management of diabetes mellitus, 2011.
https://www.aacc.org/science-and-research/practice-guidelines/diabetes-mellitus and
Sacks DB, etal. Guidelines and recommendations for laboratory analysis in the diagnosis
and management of diabetes mellitus. (Clin Chem 2002; 48: 436). They reinforce the
WHO recommendation that a laboratory measurement of plasma glucose be used for the
diagnosis of diabetes mellitus, and not whole blood glucose measured on a glucose meter
with the statement “Portable meters are used by healthcare workers in acute and chronic
care facilities, in physicians’ offices, and by patients. Because of the imprecision and
variability among meters, they should not be used to diagnose diabetes and have limited
value in screening.” One of the issues which should also be considered as part of this
argument, particularly for hospital in-patients whose insulin infusion (and consequent
insulin dosage) is adjusted in association with POC glucose testing, is that relatively small
changes in measured glucose are associated with relatively large changes in insulin
infusion rate. Blood glucose monitoring is used to guide therapy in two distinct situations,
one to adjust insulin dosage in diabetic patients, and one for adjusting insulin requirement
in acutely ill patients on “tight glucose control”. The quality specifications for these two
situations have been reviewed by Price (Price CP. Glucose monitoring and patient safety.
Point of Care 2008; 7: 253). In a recent evaluation study by Freckmann etal (Freckmann
G, etal. System accuracy evaluation of 27 blood glucose monitoring systems according to
DIN EN ISO 15197. Diabetes Technology and Therapeutics 2010; 12: 221.), eleven of
twenty seven home-use blood glucose monitoring systems (40%) did not fulfill the rather
generous minimum accuracy requirements of ISO 15197 (that is; +/- 0.83 mmol/L at blood
glucose less than 4.2 mmol/L, or +/- 20% at blood glucose levels 4.2 or greater). From
their study, Freckmann etal concluded that inaccurate results which lead to false treatment
decisions by diabetic patients may cause severe health injury.
In addition to portable POC glucose meters, the fitness for purpose of portable devices
which measure HbA1c has also been questioned recently (Burns DE and Boyd JC. Few
point-of-care haemoglobin A1c assay methods meet clinical needs (editorial). Clin Chem
2010; 56: 4. and Lenters-Westra E and Slingerland RJ. Six of eight haemoglobin A1c
point-of-care instruments do not meet the general accepted analytical performance criteria.
Clin Chem 2010; 56: 44.). The main reason for this concern is the questionable ability of
some HbA1c devices to provide a clear analytical distinction between the recommended
decision (treatment) levels of 53 mmol/mol and 64 mmol/mol. As discussed above, to
clearly distinguish between HbA1c values of 53 mmol/mol and 64 mmol/mol, an analytical
(method) uncertainty (expressed as coefficient of variation, CV%) of less than 3.0% is
required. Methods which produce analytical CV’s of 4.0% or greater are not considered
appropriate, as this degree of analytical imprecision cannot distinguish changes in patient
HbA1c levels of less than 11 mmol/mol. A more detailed discussion of these concepts and
of the relationship between HbA1c measurement and analytical performance is given by
White and Farrance (Uncertainty of measurement in quantitative medical testing. Clin
Biochem Rev 2004; 25, supplement (ii): S1).
The direct influence of analytical imprecision and inaccuracy (bias) on laboratory and POC
measurements are critical issues for the interpretation of patient results and the fitness for
purpose of testing devices. These concepts apply equally to all measurement devices and
methods, including all POCT procedures.
1. Define the quality that is required for each test (based on clinical requirement,
intra-individual biological variation or total biological variability). Quality
management begins with knowledge of the quality required of the test procedure.
2. Ascertain the performance of the POCT device or method under consideration.
The estimated precision (uncertainty) and bias of the device should be evaluated
in advance of purchase or commencement of patient testing.
3. Calculate the relevant fitness for purpose. Ideally, uncertainty of measurement
should be less than 0.5 times the intra-individual biological variation of the
particular measurand as outlined in the article by GH White and I Farrance,
Uncertainty of measurement in quantitative medical testing. Clin Biochem Rev
2004; 25, supplement (ii): S1.
4. Choose appropriate QC and EQA to establish and maintain the stability of the
testing procedure.
Suggested reading

Glucose. In: A Practical Guide to Global Point-of-Care Testing. Edited by

Edited by Mark Shephard. CSIRO Publishing, 2016
 RF Dodds. Diabetes mellitus, chapter 38. LA Kaplan and AJ Pesce. Clinical

Chemistry, theory, analysis and correlation, 5th edition 2010.
 L Al-Ansary etal. Point-of-care testing for HbA1c in the management of

diabetes: A systematic review and metaanalysis. Clin Chem 2011; 57(4):568.
 CG Fraser and MG Scott. Analytical performance requirements for point-of-

care testing: Glucose as an example. Chapter 12, in: Point-of-care testing.
Needs, opportunity and innovation. Edited by CP Price, A St John and LJ
Kricka. AACC Press, 2010.
 GH White and I Farrance. Uncertainty of measurement in quantitative medical

testing. Clin Biochem Rev 2004; 25, supplement (ii): S1. Article available in
topic library.
 WHO. The use of Glycated Haemoglobin (HbA1c) in the Diagnosis of

Diabetes. (See separate item for this topic).
 GRD Jones etal. Change of HbA1c reporting to the new SI system. MJA
2011; 195: 45.
 RR Little. Performance of haemoglobin A1c assay methods: Good enough ?

Clin Chem 2014; 60: 1031.
Topic 11
POCT for Haematology and Coagulation

• Overview of topic 11
• Definitions
• Haemoglobin
• Haemostasis
• Coagulation and the coagulation cascade
• Anticoagulation therapy (including warfarin)
• International Normalised Ratio (INR)
• POCT devices for monitoring warfarin therapy
• Activated clotting time
• Global tests for haemostasis
• Platelet function tests
• Fitness for purpose
• Summary
• Supplementary references for topic 11
1 of 20 Ronda Greaves and Ian Farrance


• Know the terminology associated with laboratory testing in haematology and
coagulation.
• Know the definitions of common haematological measurands.
• Be able to describe the importance of pre-analytical sample handling in
haematology.
• Understand the various professional guidelines which describe POCT in
haematology and coagulation.
• Have a general knowledge of the coagulation cascade and its relevance to tests for
haemostasis.
• Be able to discuss the technologies available for POCT in haemostasis.
• Be able to discuss the concept of International Normalised Ratio (INR), its role in
monitoring anticoagulation therapy, POCT devices which are available for INR
testing and their fitness for purpose.
Haematology (also spelt hematology) is the study of the morphology and physiology of
blood, blood forming organs and blood diseases. This also includes diseases which affect
the production of blood cells and their components, the oxygen transport protein
haemoglobin, and the mechanism of coagulation. The medical and scientific staff who
work in a haematology laboratory perform an array of tests that investigate the cellular
elements of blood, cellular proteins such as haemoglobin and plasma proteins involved
with clotting. Haematological tests are performed to diagnose diseases such as
leukaemia, anaemia, genetic variations in the structure of haemoglobin, and abnormalities
of blood coagulation. Blood transfusion procedures are often included as part of a
haematology laboratory.
The most commonly requested test in a haematology laboratory is a Full Blood Count (or
Full Blood Examination or Complete Blood Count; FBC or FBE or CBC). The FBC test
provides information about the number, size and variability of the red blood cells (RBC),
platelets, and the commonly encountered white blood cells (WBC). Current (large)
laboratory-based analysers can generate many other parameters that may be of use in the
diagnosis of various conditions. However, even with modern laboratory-based analysers,
haematology scientists must be aware of the limitations associated with such
instrumentation. As POC analysers are even more limited in their technical performance
and with the parameters which they can provide, great care must be taken with result
interpretation.
The main utility of POCT is to improve the time at which the results of a test become
available. In haematology testing, the use of POC instruments for haemoglobin estimation
is well-established as is the use of coagulation monitoring in operating theatres, critical
care units, primary practice and home care. However, modern technology is continually
providing improved instrumentation and new devices which monitor different parameters.
Various professional and international guidelines outline the management and use of
POCT in primary care and in hospital networks. Professional guidelines play an important
role in the development and introduction of POCT as they provide recommendations which
include: appropriate venues for POCT utilisation, the range of tests which may be
appropriate, operator training, quality control and quality assurance, result interpretation
and fitness for purpose, device maintenance, patient and operator safety. For POCT in
haematology and coagulation, specific guidelines have been provided by the British
Society of Haematology (BSH), The International Council for Standardisation in
Haematology (ICSH) and the National Association of Clinical Biochemistry in the USA.
These guidelines are considered as compulsory reading for this topic. In addition, there
are various national agencies with an interest in POCT - the Advisory Committee on
Medical Devices (ACMD) provides independent medical and scientific advice to the
Minister of Health and to the Therapeutic Goods Administration (TGA) on safety, quality
and performance of medical devices supplied in Australia.
In common with many laboratory and POCT procedures, the type of specimen used in a
POCT is of critical importance for both the analytical procedure and for result
interpretation. The blood specimen used for FBC diagnostic testing in a clinical laboratory
is usually venous or arterial blood anticoagulated with EDTA. While venous and arterial
blood give similar results for haematology parameters, capillary samples which are usually
required for POCT are often more concentrated due to the capillary sampling method.
Even when the collection is standardised and expertly performed, this slight increase in
concentration results in a small but significant increase in measured haemoglobin and
other RBC parameters. In addition, if a sample of anticoagulated venous blood can be
used as a replacement for capillary blood in a given POCT procedure, it is critical that the

sample is well-mixed to overcome gross errors in estimation which may otherwise occur.
Pre-analytical errors of this type require careful consideration for all POCT assays as
discussed in Topic 2. Both Topic 2 and Topic 3 should be reviewed in association with this
current topic.
Coagulation (or thrombogenesis) is the process by which blood forms clots. It is the
principal mechanism which inhibits blood loss from a damaged blood vessel. The clotting
process involves several steps but essentially involves the damaged blood vessel wall
being covered by a platelet and fibrin clot. Clotting starts almost immediately following
injury, it stops bleeding and begins repair of the damaged vessel. Tissue thromboplastin
released from endothelial cells and leucocytes initiate changes to platelets and the plasma
protein fibrinogen. Platelets quickly form a plug at the site of injury, while the plasma
protein “clotting factors” respond in a complex cascade to form insoluble fibrin which
strengthens the platelet plug. Disorders of coagulation can lead to an increased risk of
bleeding (haemorrhage) or obstructive clotting
(thrombosis).http://en.wikipedia.org/wiki/Coagulation - cite_note-isbn1-4051-8460-4-1#cite_note-
isbn1-4051-8460-4-1
Suggested reading
• Briggs C, etal. Guidelines for point-of-care testing: haematology.

Brit J Haem. 2008:142: 904–915.
• Briggs C, etal. ICSH Guideline for worldwide point-of-care testing in

haematology with special reference to the complete blood count.
Int J Lab Hem 2008;30:105–1160.
• NACB. Evidence based practice for POCT. Chapter 4, Coagulation.
• Harper P and Sheppard B. Point-of-care testing for the management of oral

• Williamson M. Point-of-care testing for markers of haematology disorders. In:

A Practical Guide to Global Point-of-Care Testing. Edited by Mark Shephard.

11.3 Definitions
• Activated partial thromboplastin time (aPTT, APTT): A laboratory test which

measures the efficacy of both the "intrinsic" and the “common” coagulation
pathways. In order to activate the intrinsic pathway, phospholipid, an activator
(such as silica, celite or kaolin) and calcium are mixed into a blood plasma sample.
The time taken for a clot (thrombus) to form is the APTT.
• Anaemia (anemia): A decrease in number of red blood cells (RBC’s) or less than
the normal quantity of haemoglobin in the blood
• Erythrocyte: Red blood cell (RBC).
• Haematology (Hematology): The study of the morphology and physiology of blood,
blood forming organs and blood diseases. This includes the cellular components
of blood, haemoglobin, blood cell proteins and the mechanism of coagulation.
• Haemoglobin (hemoglobin): The iron-containing oxygen transport metalloprotein in
red blood cells
• Haemorrhage: Bleeding; profuse bleeding from ruptured blood vessels
• Haemostasis: The name given to the process which prevents blood loss once the
vascular system has been breached. The cessation of bleeding.
• INR: International Normalised Ratio.
• Leucocyte: White blood cell (WBC). White blood cells are cells of the immune
system, they are involved with defending the body against both infectious diseases
and foreign materials.
• Platelet (or thrombocyte): A small, disk shaped clear cell fragment, derived from
fragmentation of precursor megakaryocytes. The average lifespan of a platelet is
normally just 5 to 9 days. Platelets are a natural source of growth factors. They
circulate in the blood of mammals and are involved in haemostasis.
• Polycythaemia: A disease in which the proportion of the blood volume that is
occupied by red blood cells increases. Blood volume proportions (the ratio or
proportion of blood cells to blood volume) can be measured as haematocrit.
Polycythaemia can be due to an increase in the number of red cells or to a relative
decrease in the volume of plasma.
• Thrombocyte: See platelet.
• Thrombogenesis: The formation of a (blood) clot or thrombus.
• Thrombosis: A blood clot.

11.4 Haemoglobin
Haemoglobin is the oxygen carrying protein found in red blood cells (RBC’s). It is a
tetramer of two pairs of different globin chains with a molecular mass of 64,000 daltons.
The physiological reference intervals for blood haemoglobin vary considerably: male and
female, newborns, infants, children up to puberty, and adults. There is also considerable
variability associated with physical exercise, posture and altitude. High altitude causes
increased haemoglobin, as lower oxygen (at higher altitudes) stimulates haemoglobin
production.
Anaemia, the term used to describe a decreased blood haemoglobin or a decreased

number of red blood cells, occurs in three main situations: impaired production (as in
aplastic anaemia), increased destruction (as in haemolytic anaemia), and excessive blood
loss (as in haemorrhage). Excessive blood loss and/or iron deficiency result in iron
deficiency anaemia, where the lack of iron produces a decrease in erythropoiesis (the
process by which RBC’s are produced).
An increased blood haemoglobin concentration (polycythaemia) occurs in polycythemia

vera, erythrocytosis, dehydration, chronic lung disease, altitude, renal disease, some types
of malignancy, iatrogenic (as in blood doping).
POCT for haemoglobin has been available for many years and is used in many different
medical settings and includes: blood donor collection centres, emergency departments,
obstetric units, renal dialysis units, outpatient clinics (especially haemato-oncology),
operating theatres, and now even home use applications. POCT for haemoglobin is
thought to be useful in any situation where a patient may be bleeding, but great care is
required when interpreting a result in acute haemorrhage, as fluid shifts occur to correct
the blood volume deficit and the expected change in haemoglobin concentration may be
masked or delayed.
Haemoglobin testing is usually performed by a photometric procedure, with various

corrections being applied which depend on the specific instrumentation. Dual wavelength
instruments usually correct for lipaemia, leucocytosis and other sources of turbidity. In
addition to hand-held devices, haemoglobin is often measured on blood gas analysers
which have co-oximetry capability. With co-oximetry, the blood concentrations of
oxyhemoglobin, carboxyhemoglobin (carbon monoxide haemoglobin, CO),

methemoglobin and reduced hemoglobin are determined by spectrophotometric
measurement. As part of the blood gas procedure with co-oximetry measurements,
various derived parameters may be calculated such as oxygen saturation and oxygen
carrying capacity.
A recent development which describes a “Disposable platform provides visual and color-
based point-of-care anemia self-testing” and measures haemoglobin by a modification of
the reaction between haemoglobin, hydrogen peroxide, and 3,3′,5,5′-tetramethylbenzidine
(TMB). Haemoglobin catalyzes a redox reaction between TMB and hydrogen peroxide,
leading to oxidized TMB products that can be detected colorimetrically or assessed
visually. This reaction has been well characterized and has been used commercially for
colorimetric (spectrophotometric) detection of haemoglobin in plasma, qualitative detection
of blood in faecal samples, and crime scene investigation. A recent article describes its
use for self-testing for anaemia (EA Tyburski, etal. Disposable platform provides visual
and color-based point-of-care anemia self-testing. J Clin Investigation, DOI:
10.1172/JCI76666. (Article available in topic library)).
11.5 Haemostasis
Blood flow within the body is critical to keep all tissues oxygenated. Complex systems
keep the blood remaining fluid, while other systems prevent blood loss from the
cardiovascular system. Haemostasis is the name given to the process which prevents
blood loss once the vascular system has been breached. Primary haemostasis involves
platelets and von Willebrand factor binding to the subendothelial tissues (adhesion),
followed then by platelet−platelet interaction which results in the formation of a platelet
plug. The secondary haemostatic system involves the coagulation proteins (such as
fibrinogen) which stabilise the platelet plug.
YouTube has several animated versions of this process – some of which contain major
errors. However, the following examples are satisfactory:
http://www.youtube.com/watch?v=--bZUeb83uU
http://www.youtube.com/watch?v=xNZEERMSeyM

POC haemostasis testing may be performed in several clinical settings:
• The bleeding patient
• In association with heparin anticoagulation therapy. For example; in
cardiopulmonary bypass, heparin is used as anticoagulant to prevent blood
clotting, the effectiveness of anticoagulation (and indirectly the dose of heparin) is
monitored in order to maintain correct fluid (blood) flow.
• To ensure that there is adequate anticoagulation (but not over anticoagulation) in
specific situations such as warfarin therapy (that is; to monitor “blood thinning”)
• To monitor clot breakdown in patients with an intravascular clot or thrombotic
stroke. In this situation, the patient may be treated with a clot lysing agent, which
dissolve the clot and minimise any ischaemic damage. As this process may
become systemic rather than localised to the vessel required, it must be monitored.
The type of testing required and the analytes (measurands) actually measured depend
largely on the type of anticoagulation treatment and the specific clinical situation. For
example; monitoring heparin anticoagulation requires different test procedures to that of
warfarin anticoagulation. The requirements for cardiopulmonary bypass monitoring or for
an intensive care unit are different to those which may be appropriate for primary care
(general practitioner or home use). Platelet related test procedures may be appropriate for
critical care hospital units and operating theatres, but are unlikely to be used in primary
care. Prothrombin time (or more correctly, the international normalised ratio test) could
well be important in both situations.
International normalised ratio (INR, a type of prothrombin time testing) is discussed in

more detail below.

Suggested reading


11.6 Coagulation and the coagulation cascade
Coagulation (or thrombogenesis) is the process by which blood forms clots. It is an

important part of haemostasis. To stop bleeding and to begin repair, the damaged blood
vessel wall is covered by a platelet and fibrin containing clot. Disorders of coagulation can
lead to an increased risk of bleeding (hemorrhage) or obstructive clotting (thrombosis).
http://en.wikipedia.org/wiki/Coagulation - cite_note-isbn1-4051-8460-4-1#cite_note-isbn1-4051-
8460-4-1
When blood vessel endothelium is damaged, the underlying collagen is exposed to

circulating platelets. These platelets bind directly to collagen with collagen-specific
glycoprotein surface receptors. This adhesion is strengthened by von Willebrand factor, a
glycoprotein present in (blood) plasma. Platelet adhesion also initiates “platelet
activation”, which in turn activates more platelets and secondary haemostasis. The
coagulation cascade of secondary haemostasis has two pathways which lead to fibrin
formation. These are the intrinsic pathway (also known as the contact activation pathway)
and the extrinsic pathway (also known as the tissue factor pathway). These two pathways
merge to a common pathway for clot formation as outlined in the diagram below.

Diagram illustrating the various coagulation proteins and their interaction in the
test tube to generate a fibrin clot. This is not the in vivo schema, but is what
happens in our tests and may be used to determine the problems in vivo.
The coagulation system may be assessed by the following tests:

i. Prothrombin time (PT), is used to assess both the extrinsic and common
pathways. PT results may be reported as time in seconds, the ratio of the patient
result to a “normal” patient result (a prothrombin ratio), or as the International
Normalised Ratio (INR) as discussed further below.
ii. Activated partial thromboplastin time (APTT or aPTT), which is used to assess
the intrinsic and common pathways.
iii. Activated clotting time (ACT) does not correlate with other coagulation tests and
is used to monitor the effectiveness of high-dose heparin therapy. It is commonly
used in operating theatres to monitor patients on cardiopulmonary bypass.

Exercise
Coagulation factors are usually described by Roman numerals instead of their

“proper” biochemical name. The use of Roman numerals rather than eponyms or
systematic names was agreed by a committee of haemostasis experts starting about
1955. This expert committee eventually evolved into the present-day International
Committee on Thrombosis and Hemostasis (ICTH).
List the “proper” names for the coagulation factors shown in the coagulation cascade
(above diagram). For example:
• Factor I Fibrinogen (forms fibrin clot)
• Factor II Prothrombin (converted into thrombin, a serine protease enzyme)
• Factor IIa “Activated” prothrombin or thrombin
• Factor III … etc …(make your own list for the remaining factors).
11.7 Anticoagulation therapy
Anticoagulants are the class of drug which prevents blood clotting. There are both
injectable and oral forms. Previously, most patients who required injectable
anticoagulation received heparin, while those patients requiring oral anticoagulation
received warfarin (see additional notes below). Due to the narrow therapeutic range
associated with warfarin and need for frequent laboratory or POCT monitoring, alternative
anticoagulants have been actively sought. As a consequence, many “new” anticoagulants
have been developed in recent years. However, both warfarin and heparin are widely
used and are still probably the two most popular anticoagulants.
Patients receive anticoagulation therapy to reduce clot formation, retard fibrin deposition
on established thrombi and prevent the formation of new thrombi. The type and use of a
particular anticoagulant is based on the individual risks and benefits for a given patient.
The biggest risk of anticoagulation therapy is the increased risk of bleeding. In an
otherwise healthy person, the increased risk of bleeding is minimal when using the correct
drug dose. For patients post-surgery however, there may be both an increased risk of clot

formation without anticoagulant therapy plus increased risk of bleeding with anticoagulant
therapy. Generally, the benefit of anticoagulation is the prevention of thromboembolic
disease and/or the reduction in the progression of a disease. Some indications for
anticoagulant therapy that are known to have benefit from therapy include:
• Atrial fibrillation
• Coronary artery disease and arthrosclerosis
• Deep vein thrombosis (DVT), which may lead to pulmonary embolism
• Pulmonary embolism
• Stroke
• Hypercoagulable states and haematological disorders
• Clot formation following hip (replacement) or knee (reconstruction) surgery
• Narrowing of a blood vessel following vascular surgery, cardiac surgery or
following insertion of a stent.
In the above examples, anticoagulation therapy can prevent dangerous clots from forming
or prevent further clot development. However, the type of anticoagulant will depend on
the situation applicable to the particular patient and the clinical requirement.
A number of anticoagulants are currently available. Both warfarin and heparin are still in
widespread use, but a number of new agents have been introduced that are collectively
referred to as the novel oral anticoagulants or directly acting oral anticoagulants.
These agents include inhibitors of factor IIa (dabigatran) and factor Xa (rivaroxaban,
apixaban and edoxaban) and do not require regular laboratory monitoring. All of these
new oral anticoagulants have all been shown to be as good or even slightly better than the
coumarins (warfarin), with possibly less serious side effects. However, they are generally
more expensive than the traditional warfarin or heparin and should be used with care in
patients with renal impairment. These drugs are generally not used as a “general”
anticoagulant, but in specific clinical situations. In addition, as there is no antidote for the
factor Xa inhibitors it is difficult to stop their effects in an emergency situation (accidents,
urgent surgery, etc). In contrast, the effects of dabigatran (factor IIa inhibitor) may be
reversed with a drug called idarucizumab.

11.7.1 Warfarin
Warfarin (a dicoumarol derivative) is an oral anticoagulant that is often used long-term in

patients that require a reduction in clot formation. Warfarin prevents inappropriate blood
clotting, principally through the antagonism of endogenous vitamin K dependent clotting
factors. Even though warfarin can prevent thrombosis, the reduction in haemostatic ability
must be closely monitored to prevent over anticoagulation whilst maintaining the required
level of anticoagulation.
Vitamin K is required in its reduced form to modify several clotting factors and allow them
to become fully active. Warfarin prevents the recycling of Vitamin K into its active form
and thus inhibits the activity of coagulation Factors II (prothrombin), VII, IX and X and the
natural anticoagulants Protein C and Protein S. As indicated in the above schema,
Factors II and X are in the common coagulation pathway; Factor IX is in the intrinsic
pathway and Factor VII is in the extrinsic pathway.
Warfarin anticoagulation activity is usually monitored by measuring prothrombin time, a

prothrombin time ratio or preferably the INR. As warfarin has a very narrow therapeutic
window with a considerable patient to patient variation (inter-individual biological variation),
careful monitoring is required. Warfarin is also susceptible to interaction with other drugs
and to changes in the dietary intake of vitamin K. POC monitoring of INR is widely used in
medical clinics and for self-testing.
11.8 International Normalised Ratio
The introduction of the International Normalised Ratio (INR) by the World Health
Organisation (WHO) in 1982, provided a major advance in the monitoring of oral
anticoagulation therapy. The universally utilised formula which defines the INR is based
on a model derived from empirical data but has no theoretical foundation. By definition,
INR is described mathematically as:
ISI
 PT 
INR =  
 MNPT 
Where:
• PT is the measured prothrombin time for a given patient,

• MNPT is the geometric mean of the prothrombin times for a healthy adult
population,
• ISI is the International Sensitivity Index.
The ISI, the power term in the INR function, provides an adjustment to align a given
thromboplastin reagent and analytical device combination with a WHO international
thromboplastin standard. If the given thromboplastin reagent is identical to the WHO
standard, the ISI equals 1.0. However, depending on the particular batch of reagent
thromboplastin and the measurement system (equipment) employed, ISI values can range
from approximately 0.9 to approximately 2.0 (or even greater). The theory behind the use
of INR, is to minimise inter-laboratory and inter-device variation when comparing patient
results and therapeutic ranges based on prothrombin time derived test procedures.
Suggested reading
• Poller L. International Normalised Ratios – The first 20 years.

J Thromb Haemost 2004; 2: 849-860.


11.9 POCT devices for monitoring warfarin therapy
The POC monitoring of PT, INR and APTT is now widely used in medical clinics,
emergency departments and the home (home use for self-testing). Two of the most
popular POC coagulation monitoring analysers currently available in Australia are the
Abbott iSTAT and the Roche CoaguChek. However, other devices are available from a
variety of suppliers.
• Examples of POC coagulation analysers,

http://www.captodayonline.com/productguides/instruments/coagulation-
analyzers-may2013.html
• iSTAT from Abbott diagnostics,

http://www.abbottpointofcare.com/
• CoaguChek from Roche Diagnostics,

http://www.coaguchek.com/coaguchek_hcp/en/home/coagulation_monitoring/m
edical_and_scientific_information.html/
11.10 Activated clotting time
Activated clotting times are rarely performed in diagnostic laboratories. They are most
often performed as POC procedures in operating theatres to assess the anticoagulation
efficacy of high-dose heparin or direct thrombin inhibitors such as argatroban
( http://en.wikipedia.org/wiki/Argatroban ), lepirudin ( http://en.wikipedia.org/wiki/Lepirudin )
or bivalirudin ( http://en.wikipedia.org/wiki/Bivalirudin ). The pathology laboratory rarely
receives samples for comparison of ACT results, as the interpretation and assessment of
treatment options occurs in the operating theatre.
11.11 Global tests of haemostasis
In addition to PT, INR and ACT, whole blood tests of haemostasis also include tests to
evaluate the effects of platelets, RBC and WBC on the haemostatic mechanism. There
are also instruments which measure the visco-elastometric properties of clot formation
(thromboelastography, TEG), a procedure which depends on the interaction of fibrinogen,
platelets, and fibrinolytic proteins. In addition to the “traditional” TEG method, a new

technology called rotational thromboelastometry (ROTEM) has recently become available.
Both TEG and ROTEM instruments are considered useful for the intra-operative
assessment of blood therapies which are required for the bleeding patient. These
techniques may also be useful in the monitoring and reversal of thrombolytic therapy.
Useful internet sites:
• http://www.practical-
haemostasis.com/Miscellaneous/Miscellaneous%20Tests/teg.html
• https://www.rotem.de/en
• http://www.jaypeejournals.com/eJournals/ShowText.aspx?ID=2821&Type=FRE
E&TYP=TOP&IN=_eJournals/images/JPLOGO.gif&IID=221&isPDF=YES
11.12 Platelet Function Testing
Anti-platelet therapy (aspirin, clopidogrel) may be used to minimise platelet thrombi

following various procedures such as coronary angioplasty.
• POC instruments such as the PFA-100 (Siemens Medical Solutions USA, Inc.,
Malvern, Pennsylvania, USA), may be used to measure the time taken for
blood, drawn through a fine capillary to clot in membranes coated with either
collagen and epinephrine or collagen and adenosine diphosphate (ADP). It is
thus a measure of both platelet adhesion and aggregation.
• The Plateletworks (Helena Laboratories, Beaumont, Texas, USA) provides a

complete haematology profile, including a standard FBC, platelet count plus
platelet aggregation. Aggregation testing is based on performing a platelet
count before and after intentional platelet activation using either collagen, ADP
or arachidonic acid as the agonist.
• The VerifyNow (Accumetrics, San Diego, California,USA) device measures

platelet responsiveness to antiplatelet medications using a turbidimetrically
based optical detection system that evaluates the agglutination of whole blood
samples that have been exposed to specific agonists.

been used to set goals for both bias (systematic error) and imprecision. A widely used and
be used for total analytical error (bias plus imprecision). These concepts are clearly
The direct influence of analytical imprecision and bias bias on laboratory and POC

testing procedure.
11.14 Summary
The pressure to improved POCT analysers is likely to increase in the future, with many
more instruments with different technologies becoming available. As a consequence, the
introduction of quality management procedures and a full understanding of the relevant
technology is of vital importance. Medical laboratory personnel with expertise in
haematology measurement procedures should also be encouraged to assist with POCT
applications. Professional guidelines which outline the technical and quality management
requirements for POCT, are essential documents for the successful implementation of
POC procedures.

11.15 Suggested reading for Topic 11
Suggested reading for Topic 11
• Briggs C, etal. Guidelines for point-of-care testing: haematology.

Brit J Haem. 2008:142: 904–915.
• Briggs C, etal. ICSH Guideline for worldwide point-of-care testing in

haematology with special reference to the complete blood count.
Int. J Lab Hem 2008;30:105–1160.


• Poller L. International Normalised Ratios – The first 20 years.

J Thromb Haemost 2004; 2: 849-860.
• Heneghan C, Perera R and Garcia-Alamino JM. Self-monitoring of oral

anticoagulation therapy.
• NACB. Evidence based practice for POCT. Chapter 4, Coagulation

11.16 Supplementary references for Topic 11 (See below).
Supplementary references (also refer to topic library)

• United Kingdom department of healthy. Centre for evidence-based purchasing
(CEP) Type into the search box - POCT or INR.
http://nhscep.useconnect.co.uk/CEPProducts/Catalogue.aspx
• Moran RF. Application of hemoglobin derivatives in STAT analysis. Radiometer,

acute care testing. http://acutecaretesting.org/ (search by name “Moran” and
scroll down, search be “haemoglobin” and scroll down, etc …).
• Hirsh J etal. Oral Anticoagulants - Mechanism of action, clinical effectiveness,

and optimal therapeutic range. Chest 1998; 114: 445(S).
• Hirsh J etal. Heparin – Mechanism of action, pharmacokinetics, dosing

considerations, monitoring, efficacy and safety. Chest 1995; 108: 259(S).
• Rao LV, Ekberg BE, Connor D. Evaluation of a new point of care automated
complete blood count (CBC) analyzer in various clinical settings.
Clin Chim Acta 2008: 389: 120.
• van Berkel C, Gwyer JD, Deane S et al. Integrated systems for rapid point of
care (POC) blood cell analysis. Lab Chip 2011; 11: 1249.
• Rhee AJ and Kahn RA. Laboratory point-of-care monitoring in the operating

room. Curr Opin in Anesth. 2010; 23: 741.
• Solvik UO, Petersen PH, Monsen G, Stavelin AV, Sandberg S. Discrepancies in

International Normalized Ratio Results between Instruments: A Model to Split
the Variation into Subcomponents. Clin Chem 2010; 56: 1618.
• Laposata M. Point-of-care coagulation testing: Stepping gently forward.

Clin Chem 2001; 47: 801.
• Prisco D and Paniccia R. Point-of-Care Testing of Hemostasis in Cardiac

Surgery. Thrombosis Journal 2003; 1: 1.
• EA Tyburski, etal. Disposable platform provides visual and color-based point-of-

care anemia self-testing. J Clin Investigation, DOI: 10.1172/JCI76666.

Topic 12 – POCT for Drugs of Abuse

• Overview: Measurement of drugs in body fluids
• Therapeutic drug monitoring
• Toxicology and the poisoned patient
• Non-clinical applications and POCT drug testing
• Sampling requirements
• Chain of custody
• Special NATA accreditation for specimen collection and specimen referral
procedures
• Ethanol
• Workplace testing
• Technology for POCT ethanol and drug testing
• Assessment of smoking status
Page 1 of 25 Ian Farrance and Ronda Greaves


• define the terminology and describe the differences between therapeutic drug
monitoring, drugs used in excess of therapeutic requirements, drugs of abuse and
toxicology
• outline the common toxicological reasons for coma and how these may be
distinguished
• describe laboratory tests which may be useful in the evaluation of an unconscious
patient
• describe what types of POC test may be available to assist in the evaluation of an
unconscious patient
• describe the major drug classes for which workplace POCT may be undertaken
• describe the use of POCT for detecting drugs of abuse, the precautions required,
the special requirements for sample collection and the validity (legal acceptability)
of the results obtained
• describe “chain of custody”
• outline the type(s) of analytical procedures used for drug POCT devices
• outline the procedures used for the assessment of smoking status.
12.2 Overview: Measurement of drugs in body fluids
Therapeutic drug monitoring (TDM) may be considered as the use of drug measurements
in body fluids as an aid to the management of patients receiving drug therapy for the cure,
alleviation or prevention of disease. It has long been customary to adjust the dosage of a
drug in accordance with its characteristics and its therapeutic effect on the particular
individual concerned. Dosage individualisation is much more difficult for drugs which
cannot be easily assessed clinically or when toxic effects cannot be detected until severe
toxicity is present. Measurement of drug levels in body fluids (blood, serum, urine, saliva
etc) is now an established clinical practice which assists with optimising and individualising
therapy.
Drug(s) of abuse (or substance abuse) has a wide range of definitions, all relate to the
taking of a psychoactive drug or performance enhancing drug for a non-therapeutic or non-
medical effect. Some of the drugs most often associated with this term include alcohol
(ethanol), amphetamines, barbiturates, benzodiazepines, cocaine and opioids. Use of
these drugs may lead to criminal penalty (depending on the local regulatory situation), in
addition to possible physical, social, and psychological harm. Testing for the presence of
such drugs is common practice in many clinical, social or legal environments.
Toxicology is the study of toxins, their physiology, biochemistry, clinical effect(s), and the
molecular basis of their activity. In the current (clinical) context, toxicology relates to the
investigation of a poisoned patient and identification of the relevant toxin (poison).
Poisoning may occur for many reasons: deliberate self poisoning, suicide, as a
consequence of substance abuse, homicide, euthanasia, occupational poisoning
(unintentional exposure), iatrogenic (therapeutic error, prescription error, failure to
recognise impairment of the principal organ responsible for elimination of a particular
drug), etc.
Point of care applications are available for many of the drugs which may be considered in
the above situations. POCT may be useful in a hospital emergency department to assess
a critically ill (poisoned) patient on admission, as part of a workplace screening program or
as part of a police driver screening initiative. Detecting substance abuse using POCT is a
multimillion dollar business. Overdose and drug of abuse testing is generally performed
on urine even though saliva is being used in defined situations and breath is the preferred
sample for ethanol screening. The purported flexibility and relative ease of use make
POCT devices an attractive analytical tool in a variety of settings and by an equally varied
range of operators. This also is one of their disadvantages; untrained operators with the
potential for inappropriate use and result interpretation may make results unreliable.
Certainly from a legal perspective, POCT drug (and ethanol) results are usually only
considered as preliminary or screening results. Positive screening results usually require
confirmation by an accredited laboratory procedure. In Australia and many countries, drug
testing which is required for legal purposes (for example; workplace testing, with a positive
test involving a penalty or dismissal), requires confirmatory testing to be performed to
Australian Standard AS 4308.
The use of POCT for drugs of abuse in both medical and non-medical situations raises
many issues, including; operator training, quality control, quality assurance, accuracy of
results, limitations of the technology and economics. In particular, POC drug testing is
generally designed to screen for the presence of designated drugs or groups of drugs.
None are intended to serve as a definitive or confirmation test.
12.3 Therapeutic Drug Monitoring

(This section is provided for background information)
12.3.1 Why monitor plasma drug concentrations ?

TDM is used to avert toxicity, to optimize dose and therapeutic response, to detect
changes in pharmacokinetics and to monitor compliance. In order for it to be useful and
cost effective, it is important that there should be a relationship between the plasma drug
concentration and the therapeutic response and/or toxicity. To justify laboratory
measurement there should be a relatively poor relationship between plasma concentration
and the dosage administered. There should be good clinical indications for the test such
as lack of response to treatment, suspected non-compliance or signs of toxicity. Other
requirements are the collection of an appropriately timed and dated specimen with proper
patient information including adequate clinical information to allow the interpretation of
results.
The Therapeutic Index: is the ratio of the toxic dose to the therapeutic dose and
determines whether monitoring is beneficial or not. Examples of drugs with a low
therapeutic index include Lithium, Phenytoin, Phenobarbital, Vancomycin, Digoxin and
some of the immunosuppressives.
12.3.2 Pharmacokinetics
The ‘ADME’ parameters (Absorption, Distribution, Metabolism and Excretion) should be
taken into account and it should be realised that these vary with gender, age, formulation
and method and route of administration.
12.3.3 Drug absorption

The rate of drug absorption is determined by its distribution into the various body
compartments (one or two compartment models) and whether it undergoes first-pass
metabolism by the liver. Once the drug is absorbed into the blood, it begins to distribute to
tissues depending on lipophilicity and protein binding. This partitioning of drug between
blood and tissues is expressed quantitatively as the Volume of Distribution (Vd). Drugs

with low Vd (gentamycin, phenytoin, theophylline, ethanol) are contained mostly in the
plasma, because they are highly water soluble (plasma water content is higher than
tissues), or they are highly protein bound (which prevents them from freely diffusing into
tissues. Drugs with high Vd (Digoxin, carbamazepine, propranolol) are distributed mostly
in tissues, and plasma levels may not reflect body burden.
12.3.4 Enzyme Induction

Various substances can induce increased cytochrome P450 enzyme levels in the
hepatocytes. These enzymes are responsible for the metabolism and breakdown of most
drugs. An increased level of these enzymes can mean that the drug is metabolised more
rapidly than would otherwise be the case. Diet, smoking, alcohol and other therapeutic
drugs can cause this induction to take place.
12.3.5 Steady State

Since plasma drug levels reflect the outcome of absorption, tissue uptake and metabolism
and excretion, plasma drug levels take about 5 half lives to achieve steady state. Thus it is
usually appropriate to delay for this time before first attempting to assess drug
concentration.

12.3.6 Elimination
Drug elimination can be renal, biliary, secretory or respiratory and often depends on the
metabolism of the drug. It is the most variable pharmacokinetic parameter. Thus the
factors that affect the rate of drug elimination are hepatic and renal function, urine pH,
genetic factors and the presence of other drugs being metabolised.
12.3.7 Common reasons for sub-therapeutic or toxic levels
Sub-therapeutic Toxic Levels
Non compliance Overdose
Dose too low Dose too high
Malabsorption Dose too frequent
Rapid metabolism Impaired renal function
Reduced hepatic metabolism

12.3.8 Factors affecting plasma drug concentration
Drug in dosage form
Patient compliance
Drug in solution
Absorption
Plasma Compartment
Protein-bound drug
Free Drug Kidney
Liver Excretion
Metabolites Reabsorption
Metabolites
Non Target Tissue Site of Action

Pharmacologic response
The diagram above indicates the various compartments of the body that can affect the
drug level found in plasma. The effect of disease, particularly of the liver or kidney can
influence the drug and/or its metabolites and therefore the pharmacological response.

12.3.9 Some definitions of key words
Drug Half-life. The time taken for the plasma drug concentration to decrease to one half
of its initial level.
Therapeutic Index. The relationship between the minimum effective concentration (MEC)
and the minimum toxic concentration (MTC).
First-pass metabolism. Changes that occur to the drug between administration and
arrival at the site of action in the body.
Pharmacokinetics. The activity of the drug in the body including absorption, distribution,
localisation, biotransformation and excretion.
Volume of distribution. Quantifies the body “compartment” where the drug is available.
High values indicate the drug is in the tissues, low values indicate the drug is mainly in the
plasma.
Peak and trough drug concentration. Peak concentration is usually immediately after
the dose is administered and trough concentration is usually immediately before.
12.3.10 Clinically important therapeutic drugs
Digoxin
Used in managing cardiac failure with atrial
fibrillation since it improves cardiac output.
The low therapeutic index of digoxin means
dose adjustment must be performed with
caution. Plasma digoxin measurements are
used to assess dose and to diagnose
digoxin toxicity. Non-compliance is also
common. To ensure that distribution of
digoxin is complete, samples are taken at
least 6 h post-dose. Digoxin elimination is
strongly influenced by renal function. The therapeutic range for serum digoxin is 1.3 – 2.6
nmol/L (1.0 - 2.0 ng/mL; ug/L) and toxic levels > 2.6 nmol/L (>2.0 ng/mL; ug/L) produce
arrhythmias, GI, CNS symptoms with digoxin toxicity increasing exponentially.

Carbamazepine
This anticonvulsant is monitored because there is a variable relationship between dose
and plasma concentration. It induces its own metabolism so that following the initiation of
therapy, it takes 2 - 4 weeks to obtain a steady state. Treatment therefore, commences
with low doses, increasing at weekly intervals for the first month.
Therapeutic range for serum carbamazepine: 20 – 50 umol/L (4.5 - 12 µg/mL; mg/L).
Toxicity may still occur in some patients, even with a serum result within this range.
Phenytoin
Most frequently prescribed anticonvulsant. Monitoring is imperative as a guide to dosage
adjustment, since phenytoin has a low therapeutic index and exhibits saturation kinetics
(i.e. small increases in dose result in large increases in concentration). Therapeutic range
for serum phenytoin: 40 – 80 µmol/L (10 - 20 µg/mL; mg/L) with toxicity (blurred vision,
loss of coordination, drowsiness/coma) at > 80 µmol/L (> 20 µg/mL; mg/L).
Valproic acid
Broad spectrum anticonvulsant, but mostly used for petit mal seizures. The anticonvulsant
activity and toxicity of valproic acid show no simple relationships to its plasma
concentration. Hence there is generally little clinical value in the measurement of valproic
acid. The short plasma half-life results in large variations in plasma concentrations
between doses. Therapeutic range for serum valproate is usually quoted to be
approximately 400 – 800 umol/L (50 – 115 ug/mL; mg/L)
Theophylline
This vasodilator and CNS stimulant is used to treat broncho-constriction. It can produce
nausea and headache even at concentrations within the therapeutic range. Serious
toxicity leading to cardiac arrhythmia occurs at levels above 110 mmol/L and TDM is
valuable to optimise dose and/or confirm toxicity.
Therapeutic range for serum theophylline: 55 – 110 umol/L (10 – 20 µg/mL; mg/L)

Case history 1 – Theophylline
A chronic asthmatic well controlled on theophylline, developed a severe chest infection.

She was prescribed erythromycin and later presented to her GP complaining of
tachycardia and dizziness. Her serum theophylline concentration was found to be
140 µmol/L (therapeutic range 55-110 µmol/L).
Case History Explanation

Erythromycin inhibits the metabolism of theophylline. Since the erythromycin treatment
was still required, her theophylline was stopped for 2 days and restarted at a lower dose.
Once the infection was clear she was restarted on her original dose of theophylline.
12.4 Toxicology: Investigation of the poisoned patient
Identification of a toxic substance in a comatose patient can be very difficult. This problem
is compounded if the patient has suffered trauma such as a motor vehicle accident and/or
drug abuse by more than one substance. Detoxification methods are extremely limited but
identification and quantitation of the toxic agent can aid in prognosis and identify likely
permanent damage.
12.4.1 Common causes of coma

• Intracerebral pathology
• Trauma, infection, haemorrhage
• Intoxication
• Alcohols, hypnotics, antidepressants, sedatives, barbiturates, heavy metals, gases,
salicylate
• Metabolic
• Endocrine, glucose, renal, hepatic, oxygen
• Miscellaneous
• Epilepsy, hysteria, hypo and hyperthermia
12.4.2 Laboratory evaluation of the unconscious patient
• Acid base and blood gases, osmolality, electrolytes, creatinine, glucose, calcium
• Specific tests determined by clinical symptoms
• Liver function tests (LFTs), endocrine tests, CSF analysis, etc
• Toxicological analysis
• Drug screen or quantitative analysis
12.4.3 POCT analytes routinely available to evaluate of the unconscious patient

• Acid base and blood gases, osmolality, electrolytes, creatinine, glucose, calcium
• Specific tests determined by clinical symptoms; for example, cardiac markers, urine
dipsticks
• Toxicological screen, drugs of abuse POCT screening tests, blood alcohol (BAC)
12.4.4 Toxins for which biochemical tests are useful
Paracetamol
In normal circumstances,
paracetamol is metabolised to
harmless glucuronide and
sulphate conjugates which
are excreted in urine. Small
quantities of hepatotoxic and
nephrotoxic N-acetyl p-
benzoquinoneimine (NAPQI)
are formed through the P450
enzyme system and
detoxified by conjugation with
glutathione. In overdose, the glucuronidation and sulphonation pathways are saturated
and more NAPQI is formed. The glutathione required for detoxification is limited and soon
becomes depleted. In paracetamol overdose, plasma paracetamol is used to predict liver
damage and determine the therapeutic use of N-acetylcysteine which promotes
glutathione synthesis. Enzyme induction (with alcohol or phenytoin for example) also
conveys an increased risk of liver damage.

Case History 2 - paracetamol
A 24 year old man presented to his accident and emergency department within four hours
of ingesting 64 paracetamol tablets (32 g). He had a history of alcohol misuse (about 100
“drinks” per week).
A POCT breath alcohol (ethanol) assessment was performed on the patient by the
attending nurse, which returned a result of 0.09%. Four hours after the overdose he had a
serum paracetamol concentration of 178 mg/L which was measured by the supporting
laboratory.
He was given 50 g of activated charcoal and discharged. Over the next two days he had
worsening abdominal pain, was vomiting, and became jaundiced. He returned to the
emergency department where POC investigations were performed on arterial blood for:
glucose of 2.1 mmol/L, creatinine of 218 µmol/L, international normalised ratio (INR) of 8,
and pH of 7.3.
He was started on acetylcysteine within the Emergency Department before admission to a

hospital ward. Despite developing grade 3 encephalopathy, his liver recovered to the
extent that his INR fell to 2.2. His progress was complicated by oliguric renal failure, lobar
pneumonia, and ventricular arrhythmias. He died from septicaemia and persistent
multifocal seizures on the sixth day after the overdose.
The history of alcohol misuse suggests that his serum paracetamol should have been
compared with the high risk line (100 mg/L at four hours) rather than the standard
treatment nomogram.

Salicylate poisoning
Aspirin (acetyl salicylic acid, salicylate) overdose causes metabolic acidosis and
respiratory alkalosis. Serum salicylate is a valuable test for determining the need for
therapy and to monitor efficacy. The excretory conjugation mechanism becomes
saturated, so urinary excretion is major route and alkalinised urine leads to decreased
tubular reabsorption and increased excretion (thus bicarbonate is infused). Activated
charcoal can reduce gut absorption.
Case History 3 – salicylate
A 45 year old woman had attempted suicide by ingestion of a bottle of aspirin tablets. She
was found by her son who called an ambulance. On arrival at the Emergency Department,
arterial blood was collected for general chemistry and drug analysis.
The samples for drug analysis were sent to the central laboratory whilst the general
chemistry analysis was performed immediately by the nurse on the department’s (POC)
blood gas analyzer.
Arterial blood Sodium 142 mmol/L (137-145)

Potassium 5.5 mmol/L (3.1-4.2)
Chloride 98 mmol/L (98-106)
Urea 12.0 mmol/L (3.0-8.0)
Creatinine 0.15 mmol/L (0.05-0.12)
Anion Gap 30 mmol/L (7-17)
pH 7.38 (7.35-7.45)
H+ 42 nmol/L (35-45)
PCO2 26 mm Hg (35-45)
PO2 60 mm Hg (80.110)
HCO3– 15 mmol/L (23-33)
Note the mixed acid base disorder and the high anion gap

Suggested reading
• AHB Wu etal. Recommendations for the use of laboratory tests to support

poisoned patients who present to the emergency department. NACB medical
practice guidelines. Clin Chem 2003; 49: 357 (copy of article in topic library).
• ID Watson. Laboratory analyses for poisoned patients: joint position paper.

Ann Clin Biochem 2002; 39: 328 (copy of article in topic library).
• S Vazquez and B Spaeth. Point-of-care testing for drugs of abuse. In: A

12.5 Non-clinical applications and POC testing
Drug testing for non-clinical purposes is very common. For example, workplace drug of
abuse testing is a common requirement for some occupations, but positive POCT
screening tests must be confirmed by a legally defensible test performed in an accredited
laboratory. None of the POCT devices currently available (with the possible exception of
breath alcohol analyzers) are sufficiently specific to be considered as a definitive or
confirmatory test. In Australia, confirmatory testing should be performed to Australian
Standard AS 4308:2008.
12.6 Sampling requirements
Australian standards AS/NZS 4308:2008 (Procedures for specimen collection and the
detection and quantitation of drugs of abuse in urine) and AS 4760:2006 (Procedures for
specimen collection and the detection and quantitation of drugs in oral fluid) outline the
procedures required for a laboratory to demonstrate its technical competency. They are
prescriptive in the procedures required to ensure appropriate specimen collection,
handling, storage, transport, on-site screening, initial testing and confirmatory testing. The
procedures are intended for (but not limited to) medico-legal, workplace, correctional
services or court directed testing, of any or all of the following classes of drugs:
• Amphetamine type substances.
• Benzodiazepines.
• Cannabis and metabolites.
• Cocaine and metabolites.
• Opiates.
The Standards covers all aspects of specimen collection and testing, and defines
procedures which must be followed to allow proper legal scrutiny.
These procedures include:
• Specimen collection, storage, handling and dispatch
• Integrity and identity of the specimen
• General laboratory requirements
• Laboratory screening procedures and test cut-off levels (cut-off drug
concentrations), recording of results and reporting
• Laboratory security, specimen reception, specimen integrity and storage
• Quality control
• Laboratory confirmatory procedures
• Laboratory personnel and training
POCT for drugs of abuse has evolved over many years, with urine as the most tested
sample. Cassette devices for urine drug testing are available with sensitivities and
specificities similar to enzyme immunoassays used in the central laboratory urine
screening procedures. The analytical cut-off concentrations used in POCT devices are
similar (or the same) as those used in laboratory procedures. POCT drug assays are
generally immunoassay procedures which use the same antibodies as used in laboratory
drug screening tests. Urine POC drug testing devices target the same drug, and/or class
of drug, and/or drug metabolites as those detected with laboratory drug screening
procedures. The labeling of these devices with respect to what is measured or what is
detected is particularly important. Many users may not fully understand that most of these
tests are designed to detect classes or groups of drugs and are not necessarily “specific”
for the stated drug. POC drug testing devices are sometimes inappropriately labeled as
detecting a specific drug, when actually detecting a class of drugs. This mislabeling may

lead to interpretational false positives or negatives when testing personnel do not
understand the specificity.
Saliva or oral fluid is sometimes considered as an alternative to urine and has reported
advantages and disadvantages. Saliva collection is regarded as easy and noninvasive,
and the specimen is less likely to be adulterated. Justification for the use of saliva
because of ease of collection may be disputed by the fact that oral fluid is potentially more
infectious than urine. Immunoassays developed for urine are not optimally formulated for
the parent drug or metabolite in oral fluid. Also, as drug concentrations differ in oral fluid
and urine, alternative cut-offs have been advised. In contrast to urine, drugs which may be
of interest in oral fluid are usually present as the parent drug where it is typically present at
a higher concentration level relative to its metabolite’s. This means that most devices
designed for detecting the urinary metabolites will not be useful for oral fluid testing. For
roadside testing in law enforcement, an advantage of oral fluid is that drug detection
relates more directly to current subject impairment than does drug presence in urine, with
its longer detection periods for metabolites. Collection procedures and devices for
collection are not standardized, and drug concentration can differ depending on collection
method.
Breath analysis for drugs of abuse is confined to testing for alcohol (ethanol), see item
below.
• S George and RA Braithwaite. Use of on-site testing for drugs of abuse.
Clin Chem 2002; 48: 1639.
• OH Drummer. Drug testing in oral fluid. Clin Biochem Revs 2006; 27: 147.
• WM Bosker and MA Huestis. Oral fluid testing for drugs of abuse.
Clin Chem 2009; 55: 1910.
• M Groschl. Current status of salivary hormone analysis. Clin Chem 2008; 54:
1759.
• P von Lode. Point of care immunotesting: approaching the analytical performance

of central laboratory methods. Clin Biochem 2005; 38: 591.

• A Richardson. POCT devices and drugs of abuse testing. The Clinical Biochemist
Newsletter June 2004.
12.7 Chain of custody
Chain of custody is a legal term which refers to the chronological documentation or “paper
trail”, showing the seizure, custody, control, transfer, analysis, and disposition of evidence,
either physical or electronic. In the current context, chain of custody applies to all aspects
of the collection and testing procedure including; specimen collection (for urine, this
includes observing actual urination into the collection container), specimen labeling,
handling, transport, analysis, storage, etc. Because analytical results may be used as
evidence in court all aspects of the collection and analytical process must be performed in
a scrupulous manner to avoid later allegations of tampering or misconduct. The idea
behind recording the chain of custody is to establish that the alleged evidence (drug
testing result) is in fact related to the correct person. The chain of custody documentation
should include the conditions under which the specimen was collected, who observed
urination or actually collected the specimen (for blood or saliva, etc), the identity of all
persons who handled the specimen, security conditions while handling or storing the
specimen (locked room, locked safe, etc), and the manner in which the specimen was
transferred to the laboratory or subsequent custodians.
A good overview of chain of custody is given at Wikipedia :

http://en.wikipedia.org/wiki/Chain_of_custody
12.8 NATA accreditation for specimen collection and specimen referral procedures
Refer to NATA News, December 2009, page 5. Copy of NATA documents available under
separate heading.
12.9 Ethanol
Both the clinical and non-clinical use of breath ethanol testing is quite common. Hospital
emergency departments often use breath ethanol to assess the degree of alcohol
intoxication of a patient being investigated. The presence of alcohol may be a contributing
factor to a patient’s “clinical condition” or may modify other drug or medical treatment.
The non-clinical use of breath ethanol is widely used by police as a screening procedure
for road safety, while many industrial situations require staff to be alcohol free as part of
their occupational duty (airline pilots, machinery operators, bus and truck drivers, etc)
An overview of police drug testing in Victoria and Queensland is given at
http://www.police.qld.gov.au/programs/roadsafety/drugdriving.htm and drink and drug
driving (road safety Victoria, key issues) http://www.roadsafety.vic.gov.au/key-
issues/drug-driving.html
12.10 Workplace drug testing
In workplace drug testing, two Australian standards are often used as the basis for
workplace testing:
• AS/NZS 4308:2008 Procedures for specimen collection and the detection and
quantitation of drugs of abuse in urine
• AS 4760:2006 Procedures for specimen collection and the detection and
quantitation of drugs of abuse in oral fluid
A good overview of workplace drug testing in Australia is provided by NATA in several of

their information documents. In particular refer to: Workplace drug testing – A guide to
industry (NATA information paper number 10, November 2013) at
https://www.nata.com.au/nata/component/phocadownload/category/50-nata-tech-notes-
info-papers (Scroll down to Information paper 10).
Refer also to NATA News, December 2009, page 5 - Copy available under separate
heading.
Also check out Civil Liberties Australia at

http://www.cla.asn.au/0805/index.php/articles/2009/workplace-drug-testing-caution-urged
(type workplace drug testing into the search box).
Symbion Pathology. Understanding drug and alcohol testing in the workplace (document
available in topic library).

12.11 Technology for POC ethanol and drug testing
An ethanol breath testing analyser (breathalyzer) usually employs a semi conductor

sensor or a fuel cell sensor. Semi conductor sensors are commonly used in alcohol
screening devices, personal breathalysers and for low volume professional testing. This
type of breathalyser has been shown to perform well and give accurate readings when well
maintained and calibrated regularly. A breathalyser that utilises a fuel cell sensor is using
one of the most advanced methods of determining breath ethanol. A fuel cell measures
ethanol concentration by creating a chemical reaction that oxidises the alcohol in the
sample and produces an electrical current. Fuel cell breathalysers are more specific for
ethyl alcohol and reduce false readings from substances such as ketones that are similar
in chemical structure. They are generally more accurate and stable than semi conductor
based devices. Fuel cell sensors are more expensive but commonly used in industrial or
evidential grade breathalysers.
A good general overview of breathalysers is given by Andatech at:

http://www.andatech.com.au/breathalysers . Andatech also have information on urine and
saliva drug testing.
As stated above, POCT drug assays are generally immunoassay procedures which use
the same antibodies as used in laboratory drug screening tests. Urine POC drug testing
devices target the same drug, and/or class of drug, and/or drug metabolites as those
detected with laboratory drug screening procedures. The labeling of these devices with
respect to what is measured or what is detected is particularly important. Many users may
not fully understand that most of these tests are designed to detect classes or groups of
drugs and are not necessarily “specific” for the stated drug. POC drug testing devices are
sometimes inappropriately labeled as detecting a specific drug, when actually detecting a
class of drugs. This mislabeling may lead to interpretational false positives or negatives
when testing personnel do not understand the specificity.

Suggested reading
S Vazquez and B Spaeth. Point-of-care testing for drugs of abuse. In: A

12.12 Assessment of smoking status
12.12.1 Overview
The long term consequences of smoking have been recognized for the last three decades.
The Cancer Council in Victoria and others interstate and worldwide have led the campaign
to promote the adverse health outcomes associated with smoking and to encourage
quitting. This has resulted in the incidence of smoking dropping by almost 20% in Victoria
in the last ten years; from a smoking rate of 21.3% in 1998 to 17.3% in 2007. Despite this
fall in smoking rate, tobacco use is still a significant contributor to major health problems.
Approximately one in five Victorians are now regular smokers with no statistical difference
between the number of men (18.6%) compared to women (16.1%) smoking. The
incidence of smoking varies with age, with 18.6% of people smoking who are between 18
and 29 years, 21.8% of people between the ages of 30 to 49 years and 11.9% aged over
50 years of age. However, the most alarming figures relate to the incidence of smoking
during pregnancy, with the latest figures indicating that 43.2% of teenage mothers smoke
during pregnancy (compared to 17.9% of all mothers).
To accurately categorise a patient or volunteer as a smoker or a non-smoker is important

to assess risk of complications with pathologies such as diabetes, stratification of
participants for clinical trials, confirm if a participant of a quit smoking project has complied
and in some situations for insurance purposes.

Studies looking at smoking habits in different groups have identified “fib” rates varying from
5 to 25% for self reporting. The variation in this false report rate is influenced by the
perceived guilt of the patient / study volunteer. As these self assessment reports of
smoking behaviour are flawed due to the variation in the false reporting rate, biochemical
analysis can play a part in positively identifying smokers and environmental tobacco
smoke (ETS) exposure.
12.12.2 Central Laboratory Testing of Smoking Status:

Cotinine is the accepted standard for the chemical validation of smoking status. Cotinine
is the predominant metabolite of the drug nicotine and where nicotine has a short half life
of approximately two hours, cotinine has a half life of 24 hours in non-smokers and a
shortened half life of 18 hours in smokers due to the up regulation of the cytochrome P450
enzyme CYP2A6.
Structural relationship of nicotine and its major metabolite cotinine
As a measure of smoking behaviour and ETS exposure cotinine has been measured in a
variety of body fluids including serum, urine, saliva and hair. Traditional analysis was
chromatographic by either GC with FID or MS detection or HPLC with spectophotometric
detection. Immunoassay techniques such as RIA and ELISA are also in common use, the
antibodies in these techniques often measure cotinine and additional nicotine metabolites
due to cross reactivity. An automated immunoassay for cotinine and other nicotine

metabolites has been developed by Siemens Medical Solutions for the analysis on their
routine Immulite platforms, making the analysis simple, convenient and cost effective.
12.12.3 Point of care assessment of Smoking Status

Point of care testing for smoking status is primarily focused on providing evidence to
support the patient’s or volunteer’s claim of smoking status. Currently POCT is not used to
assess exposure to ETS. Assessment of smoking status at the point of care is
advantageous for the clinician, researcher, nurse or allied heath professional to enable
them to provide immediate feedback to the patient or study volunteer. Such feedback
could include on the spot clinical advice to quit smoking or stratification / exclusion from
research projects.
Urine dip stick technology has been developed to assess urine cotinine levels at the point
of care. The major limitation for these dip sticks is the lack of sufficient sensitivity,
potentially causing false negative results.
Increased levels of carbon monoxide exhaled in the breath have been proposed as a
indirect assessment of smoking status. Collection of breath samples is considered to be a
non-invasive form of sample collection. Hand held POCT breath carbon monoxide
instruments have been developed to provide a non-invasive assessment of smoking status
at the point of care. Such tests are qualitative and indirect and other causes of increased
carbon monoxide levels in exhaled breath need to be excluded to ensure accurate
interpretation of the results. If there is doubt and confirmation is required, measurement of
cotinine levels by a central testing laboratory should be considered. In practice these
central laboratory tests are infrequently ordered.
(Note: carbon monoxide breath technology can also used by paramedics to assess carbon
monoxide poisoning. Usually these are separate instruments to the analysers used to
assess smoking status due to the analytical range required).

Hand held Bedfont “smokerlyzer” to measure carbon monoxide levels at the POC.
• Cancer Council Quit campaign website. http://www.quit.org.au

• Holl R, Grabert M, Heinze E, Debatin K. Objective assessment of smoking habits
by urinary cotinine measurement in adolescents and young adults with type 1
diabetes. Reliability of reported cigarette consumption and relationship to urinary
albumin excretion. Diabetes Care 1998; 21(5): 787.
• Greaves R, Trotter L, Brennecke S, Janus E. A simple high-pressure liquid
chromatography cotinine assay: validation of smoking status in pregnant women.
• Bedfont Scientific Limited. http://www.bedfont.com/smokerlyzer


(analyte), but this is generally not appropriate for therapeutic drug monitoring, drug of
abuse testing or toxicology testing. Uncertainty of measurement for therapeutic drugs may
be evaluated using drug half life as discussed by White and Farrance, Uncertainty of
measurement in quantitative medical testing – A laboratory implementation guide
(available in topic library).
therapeutic drug half-life, or legislative requirement (for drugs of abuse)). Quality
in advance of purchase (and certainly prior to the commencement of patient
testing).
3. Calculate the relevant fitness for purpose. As outlined in the article by GH White
and I Farrance, Uncertainty of measurement in quantitative medical testing. Clin
Biochem Rev 2004; 25, supplement (ii): S1.
testing procedure.

Suggested reading
• ID Watson etal. Drugs and ethanol. Chapter 7, NACB Laboratory medicine

practice guidelines. Evidence based practice for point of care testing
(document available in topic library).
• AHB Wu etal. Recommendations for the use of laboratory tests to support
poisoned patients who present to the emergency department. NACB medical
practice guidelines. Clin Chem 2003; 49: 357 (document available in topic
library).
• S Vazquez and B Spaeth. Point-of-care testing for drugs of abuse. In: A
• RA Mirghani. Current Status of Point-of-Care Testing in clinical toxicology
(focus on drugs of abuse). POC 2008; 7: 266.
• K Lewandrowski etal. Implementation of POC rapid urine testing for DOA in
the Emergency Department of an academic medical center.
Am J Clin Pathol 2008 129: 796.
• S George and RA Braithwaite. Use of On-Site Testing for Drugs of Abuse.
Clin Chem 2002; 48: 1639.
• SW Cotton, etal. Unexpected interference of baby wash products with a
cannabinoid (THC) immunoassay.
Clin Biochem 2012; 45: 606. Article available in topic library.
• GRD Jones etal. Mass or molar ? Recommendations for reporting
concentrations of therapeutic drug. MJA 2013; 198: 1. Article available in topic
library.

Topic 13A
Troponin and markers of cardiac function
• Learning objectives for topic 13A
• Overview of topic 13A
• Troponin and markers of cardiac function
• Issues to consider
• Heart failure and BNP
• Supplementary references
• Suggested reading for topic 13A
13A.1 Learning objectives for topic 13A
• Describe how a myocardial infarct is identified.
• Explain the temporal relationship between a myocardial infarct and the various
biochemical markers.
• Describe the troponin complex and the relevance of troponin I and troponin T as
markers of infarction.
• Explain why assay results for troponin I and troponin T can not be interchanged.
• Explain the relevance of low levels of troponin and why the limit of detection is an
important consideration for the measurement of troponin.
• Describe how BNP contributes to the diagnosis of heart failure.
13A.2 Overview of topic 13A
The definition of myocardial infarction (MI) now relies on the measurement of cardiac
troponin, either cardiac troponin I or cardiac troponin T. Although there has been
extensive evaluation of other markers, there has been no conclusive evidence that these
are superior to troponin. As analytical methods have evolved and become more sensitive,
it has also been shown that the measurement of troponin allows the diagnosis of a majority
of acute MI patients at first presentation. Measurement of troponin at POC may be
performed by semi-quantitative or quantitative test procedures. Early testing for a MI
marker such as troponin with associated early diagnosis, allows prompt intervention with
thrombolytic therapy.
Heart failure, the clinical condition which occurs when the heart is unable to provide
adequate blood supply to support metabolic needs, may present as ischaemic heart
disease or as ventricular dysfunction. Measurement of N-terminal pro B-type natiuretic
peptide or B-type natiuretic peptide assists in the diagnosis of both acute and chronic
forms of heart failure. POC assays are available to assist with diagnosis.
This topic is also covered as part of Module 1 Topic 3 Analytical Consideration for Hand
Held Devices (refer to items 3.5 and 3.9). These sections should be reviewed as part of
this topic.
13A.3 Troponin and markers of cardiac function
Coronary artery disease continues to be the most common cause of morbidity and
mortality among adults throughout the western world. The renewed interest in risk
stratification of patients with acute coronary syndromes (ACS) is largely due to an
improved knowledge of cardiac pathology, improved assay technology for existing
biochemical markers (and in particular the troponins), and new methods of treatment. An
important feature of this progress is a greater understanding of what is now called acute
coronary syndrome(s). This syndrome encompasses a variety of forms of unstable
ischaemic heart disease and can be considered as a dynamic progression from unstable
angina through to myocardial infarction. This spectrum of coronary artery disease is
determined by total plaque burden, plaque stability and plaque fissuring, with presentation
dependant on the degree of plaque disruption, plaque location, the degree of stenosis and
the presence of thrombosis.
In acute ischaemic heart disease, the most important biochemical tests are the cardiac
troponins (either troponin I or troponin T). These are proteins found exclusively in heart
muscle cells and released during the ischaemic episode and cell death. Increased blood
(plasma / serum) levels indicate heart muscle damage (myocardial infarction). There is
little evidence to indicate which of the troponins (troponin I or troponin T) is a better marker
of cardiac damage, but the ability of the assay to detect a “low” troponin level is an
important assay feature. In addition, as troponin T assays are patented, they are only
available through Roche Diagnostics.
Myoglobin, the oxygen binding protein of cardiac and skeletal muscle has also been
proposed a marker of cardiac muscle damage. It can certainly be detected at an early
stage following muscle damage, but as it is present in both cardiac and skeletal muscle, it
is not a specific marker for cardiac muscle damage. For this reason, it has not been fully
accepted as a reliable marker for myocardial infaction.
Suggested reading
P Venge, B Lindahl and H Halls. Point-of-care testing for cardiac markers.

13A.4 Issues to consider
• The use of troponin T or troponin I. These are two different proteins with different
concentration levels, assay results must not be interchanged
• Different troponin I assays (assays from different manufacturers and possibly

different assays from the same manufacturer), may not be interchangeable
• For prompt diagnosis, turn-around-time (TAT) from blood collection to result is

important. Most guidelines suggest a TAT of 60 minutes or less, with 30 minutes
or less preferred
• Time from onset of chest pain is an important diagnostic factor … results on blood
collected less than 3 hours following chest pain may give a negative result
• Increase in cardiac troponin (either T or I) in conditions other than ACS
• Possible increase in troponin T in renal failure
• The detection level and “low end” sensitivity of troponin assays
• POC troponin assays used as a screening (or diagnostic) procedure for AMI and
“comparison” with laboratory troponin values (which may give different troponin
values due to different methodology or different troponin (I or T, refer to comments
above).
13A.5 Heart failure and BNP
Heart failure or congestive cardiac failure is the ineffective pumping by the heart. This
leads to fluid accumulation (congestion) in the lungs. Incorporated in the formal definition
of heart failure is a wide spectrum of clinical conditions which relate to impairment of pump
function. Brain (or B-type) natriuretic peptide (BNP), is a hormone originally found in
porcine brain tissue. However, its main source of release is from ventricular myocardium
in response to elevations of end-diastolic pressure and volume. The main circulating
forms are proBNP, the N-terminal fragment of proBNP (NT-proBNP), and the
physiologically active BNP. Measurement of BNP or NT-proBNP give a biochemical
assessment of ventricular function. Blood (plasma / serum) levels are increased in chronic
heart failures and correlate closely with its severity. A normal BNP virtually excludes left
ventricular systolic dysfunction (LSVD), whereas an elevated BNP (or NT-proBNP)
indicates the presence of cardiac disease and the need for an echocardiogram. While
increased BNP is indicative of LSVD, it is also elevated in other cardiovascular conditions
such as pulmonary hypertension and right ventricular overload.
There have been many recent articles describing the merits of various troponin and BNP
assays and their role in diagnosis and clinical management. POC devices are widely
available for troponin and / or BNP measurement. All systems utilise an immunochemical
assay procedure, the specific details of which are very much method or device related.
Exercise
1) Why is prompt diagnosis and initiation of treatment so critical for patients suspected
of MI ?
2) When should specimens for cardiac biomarkers (troponin) be collected from

patients with suspected MI ?
3) How important are changing levels of cardiac biomarker (troponin) in the diagnosis
of MI ?
13A.5 Fitness for purpose
glucose, cholesterol and haemoglobin A1c and troponin), but in their absence various
approaches have been used to set goals for both inaccuracy (bias) and imprecision. A
widely used and internationally recommended concept is to define the upper acceptable
limit for imprecision as a proportion of the intra-individual biological variation of the
measurand (analyte). With correct choice of the proportionality factor, analytical
imprecision should not contribute significant additional variation to the test result when
compared with the natural variation of the substance being measured. A similar approach
to goal-setting can be used for total analytical error (inaccuracy plus imprecision). These
concepts are clearly described in the NPAAC document Requirements for the estimation
of measurement uncertainty and the article by White and Farrance, Uncertainty of
measurement in quantitative medical testing – A laboratory implementation guide.
When reviewing the fitness for purpose of a POCT device or laboratory method (including
POCT devices for troponin and BNP measurement), the following steps should be
considered in the order indicated:
testing procedure.
13A.6 Supplementary references
• European Society of Cardiology and American College of Cardiology. Myocardial

infarction redefined – A consensus document of the joint European Society of
Cardiology / American College of Cardiology committee for the redefinition of
myocardial infarction. J Am Coll Cardiol 2000; 36: 959.
http://content.onlinejacc.org/article.aspx?articleID=1126658
• PO Collinson et al. Impact of European Society of Cardiology / American College

of Cardiology guidelines on diagnostic classification of patients with suspected
acute coronary symdromes. Ann Clin Biochem 2003; 40: 156.
• AHB Wu et al. National Academy of Clinical Biochemistry standards of laboratory

practice: Recommendations for the use of cardiac markers in coronary artery
disease. Clin Chem 1999; 45: 1104.
• Association of Clinical Biochemists in Ireland. Guidelines for the use of
biochemical cardiac markers and risk factors, ACBI 2002. (If this link does not
work correctly, copy and paste directly into your internet browser).
• Various articles on cardiovascular disease plus the “Recommendation on use of

biochemical markers in acute coronary syndrome: IFCC proposals”, ejifcc volume
14, number 2. http://www.ifcc.org/ifcc-communications-publications-division-
(cpd)/ifcc-publications/ejifcc-(journal)/e-journal-volumes/ejifcc-2003-vol-14/vol-14-
n%C2%B0-2/
• National Heart Foundation, Cardiac Societies of Australia and New Zealand.

Guidelines for the management of acute coronary syndromes 2006 with updates
and summaries. Scroll down to view and access the various documents.
http://www.heartfoundation.org.au/information-for-professionals/Clinical-
Information/Pages/acute-coronary-syndrome.aspx
• M Gorenjak. Natriuretic peptides in assessment of ventricular dysfunction. ejifcc

volume 14, number 2. http://www.ifcc.org/ifcc-communications-publications-
division-(cpd)/ifcc-publications/ejifcc-(journal)/e-journal-volumes/ejifcc-2003-vol-
14/vol-14-n%C2%B0-2
• A Clerico and M Emdin. Diagnostic accuracy and prognostic relevance of the

measurement of cardiac natriuretic peptides: A review. Clin Chem 2004; 50: 33.
• AB Storrow et al. Use of cardiac biomarkers for acute coronary syndromes.

Chapter 3, in National Academy of Clinical Biochemistry, Evidence based practice
for Point-of-Care Testing (see references) or … https://www.aacc.org/science-and-
• PO Collinson et al. Measurement of cardiac troponins (review). Ann Clin Biochem

2001; 38: 423.
• EJ Lamb et al. The significance of serum troponin T in patients with kidney

disease: A review of the literature. Ann Clin Biochem 2004; 41: 1.
• WE Kelley et al. Increases of cardiac troponin in conditions other than acute

coronary syndrome and heart disease (review). Clin Chem 2009; 55: 2098.
• AAA Ismail et al. Interference in immunoassay is an underestimated problem. Ann

Clin Biochem 2002; 39: 366.
• DC Gaze and PO Collinson. Multiple molecular forms of circulating cardiac
troponin: Analytical and clinical significance (review). Ann Clin Biochem 2008; 45:
349.
• JH Nichole et al. Clinical outcomes of Point-of-Care testing in the interventional

radiology and intensive cardiology setting. Clin Chem 2000; 46: 543.
• TJ Clark et al. The Triage cardiac panel. Point of care J 2002; 1: 42.
http://journals.lww.com/poctjournal/Fulltext/2002/03000/The_Triage_Cardiac_Pane
l__Cardiac_Markers_for_the.11.aspx
• PO Collinson et al. A randomised controlled trial of point-of-care testing on the

coronary care unit. Ann Clin Biochem 2004; 41:397.
• FS Apple et al. Future biomarkers for detection of ischemia and risk stratification in
acute coronary syndrome. Clin Chem 2005; 51:810.
• G Wilcox et al. Measurement of cardiac troponin I levels in the Emergency

Department: Predictive value for cardiac and all-cause mortality. MJA 2001; 174:
170.
• AHB Wu. Risk stratification of cardiac troponin in ischaemic and non-ischaemic

cardiac diseases and procedures. Clin Biochem Revs 2000; 21: 79.
• P Tiderman. Risk stratification of cardiac patients: Point of care testing, progress

and practicalities. Clinical Biochemist Newsletter June 2004, page 18.
• M Panteghini et al, IFCC committee. Quality specifications for cardiac troponin

assays. Clin Chem Lab Med 2001; 39: 174.
• Centre for Evidence based Purchasing (CEP), NHS (National Health Service, UK),
A number on interesting evaluation reports and other information at:
o Nine troponin assays (CEP report 05085, December 2005).
o Three point of care devices for troponin measurement (CEP report 06020, May
2006).
13A.7 Suggested reading for topic 13A
Suggested reading
• P Venge, B Lindahl and H Halls. Point-of-care testing for cardiac markers.

• RH Christenson and PO Collinson. Point-of-care testing in acute coronary

syndromes, heart failure, stroke and venous thromboembolism: Is there
evidence of “a need for speed” ? Chapter 37, in: Point-of-care testing.
LJ. AACC Press, 2010.

disease. In: A Practical Guide to Global Point-of-Care Testing. Edited by
Mark Shephard. CSIRO Publishing, 2016
• JR Burnett. Coronary artery disease: Lipid metabolism. Chapter 37 in,

Clinical Chemistry, theory, analysis and correlation. LA Kaplan and AJ
Pesce, 5th edition, 2010
• DA Morrow etal. NACB practice guidelines: Clinical characteristics and

utilisation of biochemical markers in acute coronary syndromes.
Circulation 2007; 115: e356. Article available in topic library
• V Blaton. The role of lipids in the development of atherosclerosis and

coronary heart disease: Guidelines for diagnosis and treatment,.
eJIFCC 2003 14(2). Article available in topic library.
• M Panteghini. Recommendations on use of biochemical markers in acute

coronary syndrome: IFCC proposals. eJIFCC 2003 14(2). Article available in
topic library.
• I Aganovic. Diabetes mellitus and cardiovascular disease. eJIFCC 2003

14(2). Article available in topic library.
• AHB Wu. Recent advances in point-of-care diagnostics for cardiac markers.

eJIFCC 2014; 25 (2): 40. Article available in topic library.
Topic 13B
Lipids and cardiac risk factors
• Learning objectives for topic 13B

• Overview of topic 13B
• Lipids and lipoproteins
• Transport of lipids
• Factors affecting plasma lipid levels
• Measurement of lipids
• Calculation of LDL using the Friedewald formula
• Disorders of lipid metabolism
• Cholesterol, triglycerides and HDL: instrument technology and general
considerations
• Cardiovascular risk factors
• Supplementary references
• Suggested reading for topic 13B
13B.1 Learning objectives for topic 13B

• The metabolism of cholesterol, triglycerides, HDL and LDL cholesterol
• In general terms the various forms of hyperlipidaemia and the pathophysiology of
lipid abnormalities, including the relationship between hyperlipidaemia,
atherosclerosis, coronary heart disease and stroke
• Risk factors for cardiovascular disease
• The types of method used for the measurement of lipids and for POC lipid testing
• The analytical performance required for lipid measurement and the relationship
between analytical performance, intra-individual biological variation and the
interpretation of lipid results.
13B.2 Overview of topic 13B
Cardiovascular disease is a general description which covers all diseases and conditions
of the heart and blood vessels. There are many different causes for cardiovascular
disease. However, in developed countries such as Australia, the main underlying problem
is atherosclerosis. In this condition, there is an abnormal deposition of fat, cholesterol
and other substances (plaque) within the inner lining of arteries. Atherosclerosis is most
serious when it affects the blood supply to the heart (causing angina or heart attack) or to
the brain (which can lead to a stroke). Atherosclerosis is a slow and complex process,
often starting in childhood and progressing with age.
The major preventable risk factors for cardiovascular disease are tobacco smoking, high
blood pressure, high blood cholesterol (particularly low density lipoprotein cholesterol, LDL
or LDLC), insufficient physical activity, overweight and obesity, poor nutrition and diabetes.
Risk strongly increases with age and is higher for men, Aboriginal and Torres Strait
Islander peoples, and people from lower socioeconomic groups. New research suggests
that other factors, including depression and social factors may also play a role. Risk
factors for cardiovascular diseases are also shared with other chronic diseases such as
diabetes and chronic obstructive pulmonary diseases, so coordination of prevention, early
detection and other strategies for these conditions will lead to better patient outcomes.
POCT for increased blood (plasma, serum) lipids can assist with screening, diagnosis and
treatment monitoring. In rural and remote centres, POCT may provide the primary means
for lipid testing.
This topic is also covered as part of Module 1 Topic 3 Analytical Consideration for Hand
Held Devices (refer to items 3.5 and 3.9). These sections should be reviewed as part of
this topic.
13B.3 Lipids and Lipoproteins
More than 60,000 people die each year in Australia from coronary heart diseases (CHD) or
associated diseases (angina, myocardial infarction). This is the leading cause of death at
any age. It is estimated that 7% to 9% of total health care budget is spent on treating
CHD.
Cholesterol or rather LDL-cholesterol is the single most important risk factor associated
with CHD death.
• Cholesterol. Approximately 90% of body’s cholesterol is synthesized in the liver
and gut, from simple molecules especially acetate. Dietary cholesterol comes
mainly from animal products such as meat, eggs, etc. About 30-60% of intestinal
cholesterol is absorbed each day. This occurs through the formation of mixed
micelles that facilitate uptake across lumen (bile acids are necessary for uptake).
70% of plasma cholesterol is esterified to enhance the lipid carrying capacity of the
lipoproteins. This esterification is catalysed by the enzyme LCAT (lecithin
cholesterol acyl transferase) in the plasma or ACAT in the hepatocyte. Free
cholesterol rapidly exchanges with cholesterol in cell membranes but cholesterol
ester does not form part of this exchange.
• Fatty acids and triglycerides. Fatty acids are stored in the adipose tissue as
triglycerides. These are fatty acid esters of glycerol usually containing a mixture of
fatty acids; palmitic, oleic or linoleic acids. Triglyceride synthesis occurs in the
liver, muscle, small intestine. The fatty acid composition of triglycerides is strongly
influenced by dietary fatty acids. Plasma triglycerides increase steadily after a
meal to a maximum at 4 to 6 hours and then fall at 8 hours. Fatty acids are
released from the adipose tissue into plasma as free fatty acids by lipolysis. This is
governed by hormone sensitive lipase, the activity of which is increased by stress
hormones, fasting and diabetes mellitus. Free fatty acids (FFA) are transported to
liver or muscle which are the main sites of utilization. They are usually bound in
plasma to albumin or prealbumin and utilized rapidly.
• Lipoproteins. All lipids, except FFA’s, are transported in plasma in the form of
lipoproteins. These have a thin hydrophilic layer of protein, phospolipid and free
cholesterol which surrounds the hydrophobic lipid core of triglyceride and
cholesterol ester. Lipoproteins differ in size and composition continuously due to
their metabolic activity. Lipoprotein particles can be classified on the basis of their
size, density and protein content into populations (very low density lipoprotein,
intermediate density lipoprotein, low density lipoprotein, high density lipoproteins).
These densities are determined by ultracentrifugation; with the “large” low density
Chylomicrons > VLDL > IDL > LDL > HDL. It is also possible to separate the
various lipoprotein classes by electrophoresis.
• Apoprotein main functions. Assist in solubilising cholesterol esters and
triglycerides. Regulate reactions of lipids with enzymes, LCAT, lipoprotein lipase
and hepatic lipase. Bind to cell surface receptors, determining the sites of uptake
and rates of degradation of other lipoprotein constituents.
13B.4 Transport of lipids
13B.4.1 Exogenous
Dietary fat is absorbed in the small

intestine and is packaged into
chylomicrons. These enter the
circulation via the thoracic duct and are
the major transport form of endogenous
lipid, consisting of 90% triglyceride.
In the capillaries of adipose tissue and
skeletal muscle, chylomicrons are
degraded by lipoprotein lipase (LPL) in
the presence of Apo C-II removing
triglyceride (delivering FFA to these tissues) and the smaller remnants are then taken up
by hepatocyte receptors that recognise apo E on the remnant particle surface.
13B4.2 Endogenous
VLDL are synthesized by the liver and

transport triglyceride to the tissues.
VLDL also undergo degradation to
remnants by LPL releasing triglyceride
and the particle becomes the smaller
IDL. This is removed by uptake into the
liver by a receptor that recognizes the
apo B-100 on the remnant, or
undergoes further degradation by
hepatic lipase to LDL (apoB-100 but no
apoE). As the proportion of triglyceride
in the particle is being reduced, the predominant lipid in LDL is cholesterol ester.
LDL particles are the principal carriers of cholesterol (as cholesterol ester). The LDL
receptor (liver and tissues) recognizes apo B-100 and internalizes the particle allowing
release of the cholesterol. This inhibits HMG-CoA reductase (rate limiting for endogenous
cholesterol synthesis), downregulates LDL receptor synthesis and stimulates cholesterol
esterification by ACAT.
13B.4.3 Reverse cholesterol transport
HDL are synthesized in the liver and gut.

They acquire cholesterol from tissues and
other lipoproteins. LCAT is activated by
apo A-I and esterifes the cholesterol in
HDL. The cholesterol ester is transferred
to remnant particles (CM and IDL) in
exchange for triglyceride (mediated by
CETP) and is thus returned to the liver for
excretion in bile.
HDL is involved in reverse cholesterol
transport, removing potentially
atherogenic lipids from peripheral tissues
and transporting them back to the liver for processing.
13B.5 Factors affecting plasma lipid levels
• Age − lipids generally rise through life, mainly due to increased LDL cholesterol
• Gender − females tend to have higher HDL cholesterol and lower triglycerides than
males from puberty onwards. Probably due to influence of sex hormones.
Oestrogens lower LDL-C
• Body Weight − triglycerides increase with body weight. Cholesterol increases only
slightly with body weight. HDL-C tends to decrease
• Physical Activity − reduces fasting triglycerides, total cholesterol, LDL-C and
increases HDL-C
• Dietary – diets high in fat particularly saturated fats increase cholesterol and CHD
incidence
• Smoking lowers HDL
• Alcohol in moderation increases HDL and decreases LDL. Excess alcohol
increases VLDL and triglycerides.
13B.6 Measurement of lipids
Exercise
Review the types of method available for the measurement of plasma lipids.
Good method reviews are to be found at the internet site of the Clinical Chemistry
textbook, Clinical Chemistry, theory analysis and correlation., by LA Kaplan and AJ
Pesce, 5th edition, 2010.
Evolve method reviews. Need to register, but this is free.
https://evolve.elsevier.com/cs/Satellite/StudentHome?Audience=Student
Method documents for cholesterol, triglycerides and HDL cholesterol are available in
the topic library.
13B.7 LDL cholesterol using the Friedewald formula
Both indirect and direct methods can be used to measure LDL cholesterol. Direct
measurement is not usually available using POC technology. The most widely used
indirect method for LDL cholesterol uses the Friedewald calculation based on total
cholesterol, triglycerides and HDL cholesterol. This calculation may form part of an
automatic (in-built) procedure for a POCT device for lipid measurement.
The Friedewald calculation is based on actual measurements for:
Total cholesterol = HDL + LDL + VLDL
VLDL = triglycerides / 2.2
Thus, total cholesterol = HDL + LDL + (triglycerides / 2.2)
So, LDL cholesterol = total cholesterol - HDL - (Triglycerides / 2.2)
Provided triglycerides < 4.5 With all values in mmol/L
13B.8 Disorders of lipid metabolism
Prolonged hyperlipidaemia results in the accumulation of tissue lipids with adverse clinical
manifestations. Lipids may accumulate in arterial walls (intima) and cholesterol
accumulation, leads to cell proliferation. This fibrous tissue leads to atheromatous plaque
formation, occlusion of blood flow and potential ischaemia. Cholesterol can also
accumulate in subcutaneous tissues causing xanthomatas of the tendons, knees hands or
around and in the eye (corneal arcus).
• Familial hypercholesterolaemia (FH). This is due to a genetic mutation which

leads to deficiency of LDL receptors causing accumulation of LDL and increased
cholesterol. Xanthomata and corneal arcus are common. It is usually
heterozygous and affects 0.2% of the population (1 in 500) who have cholesterol
2x normal. Rarely (1 in 106) there is a double homozygous mutation with
cholesterol levels up to 20 mmol/L and with early onset of CAD. If untreated these
individuals die young, There are 4 classes of mutation with over 500 different
mutations (receptor negative or receptor defective) now identified.
• Common (polygenic) hypercholesterolaemia. Raised cholesterol is less high

than in FH and appears influenced by a number of genes. Environmental factors
such as diet are important.
• Familial Combined Hyperlipidaemia. Overproduction of hepatic apo B-100 leads

to overproduction of VLDL and LDL. Both cholesterol and triglycerides are raised.
Relatively common (0.5% of population with 10-15% of CAD’s having the problem.
Hard to diagnose due to lack of distinctive clinical features, requiring family studies.
Treatment includes use of gemfibrozil to reduce triglycerides.
• Remnant Removal Disease. Also called dysbetalipoproteinaemia or type III
hyperlipoproteinaemia. Is caused by the accumulation of remnant particles due a
decreased affinity of ApoE for the hepatocyte receptor and results in increased
cholesterol and triglyceride. Patients present with palmar xanthomas and have
increased risk of coronary and cerebral vascular disease. Half the patients have
premature atherosclerosis. Apo E polymorphisms are responsible with the E2/E2
phenotype associated with RRD but disease is not present in every case (1:100
have the phenotype but only 1:10,000 get the disease, so other factors such as
obesity or alcohol have a role). E3/E3 is the commonest (normal) phenotype and
E4 is linked to Alzheimer’s disease.
• Familial hypertriglyceridaemia. Affects 1% of the population with increased

levels of larger than normal VLDL particles. This results in moderately increased
triglycerides (3.5 – 12.0 mmol/L). HDL is often decreased and there may be a
increased risk of CHD. Patients are often obese, diabetic and in severe cases
have a risk of pancreatitis.
13B.9 Cholesterol, triglyceride and HDL: instrument technology and general

considerations
Cholesterol, triglycerides and high density lipoprotein cholesterol (HDL or HDLC) are all
available on POC devices. The clinical utility and clinical application of these tests are well
described in standard Clinical Chemistry textbooks such as those suggested in the general
references.
The main issue with POC devices for lipid testing is their ability to provide an appropriate
level of analytical performance. Diagnosis and monitoring of lipid disorders require test
methods that provide both good accuracy and precision. The usual criteria used to assess
analytical performance for cholesterol and HDL measurements are those developed by the
Centres for Disease Control (CDC) and the National Heart, Lung and Blood Pressure
Institute (NHLBI). The performance expected for cholesterol from laboratories who do
not participate in the cholesterol reference method laboratory network is a bias of 3% or
less and an overall imprecision of 3% or less. This compares to the Australian Quality
Assurance Programme (QAP) total allowable error no greater than +/- 0.50 for cholesterol
values less than 10.0 mmol/L. Equivalent CDC values for HDL are a bias of 5% or less
and an overall imprecision of 4% or less, with QAP total allowable error no greater then +/-
0.2 for HDLC values less than 2.0 mmol/L. These performance specifications should be
compared to the desirable performance specifications based on intra-individual biological
variation as discussed in Module 2. The concept of analytical performance which is based
on biological variation is discussed in detail by CG Fraser in his book Biological Variation
and by Westgard et al (see references below). The intra-individual biological variation for
plasma cholesterol in a healthy individual is approximately 5.4%.
POC lipid testing systems usually measure cholesterol, triglycerides and HDL
simultaneously, with low density lipoprotein cholesterol (LDL or LDLC) being derived by
calculation (refer to Friedewald calculation above). Enzymatic reaction(s) form the basis of
the measurement process as described in manufacturers’ literature and in the article by
MDS Shephard et al (mandatory reading, below). The reaction sequences are essentially
the same as those used in standard laboratory testing and similar interfering factors should
thus be expected (and in particular ascorbic acid, a common interferent in hydrogen
peroxide coupled reactions).
Suggested reading
• Fraser CG. Optimal analytical performance for point of care testing.

Clin Chim Acta 2001; 307: 37 (Article available in topic library).
• MDS Shephard et al. Comparative performance of two point-of-care analysers

for lipid testing. Clin Lab 2007; 53: 561 (reference available in topic library).

disease. In: A Practical Guide to Global Point-of-Care Testing. Edited by Mark
13B.10 Cardiovascular risk factors
According to the Australian Heart Foundation and the American Heart Association, major
risk factors are those that research has shown to significantly increase the risk of heart
and blood vessel (cardiovascular) disease. Coronary artery disease is preventable, but
the more risk factors one has, the greater ones chance of developing coronary heart
disease.
Exercise
Review the information describing risk factors for cardiovascular disease at

Heart Foundation (of Australia):
https://www.heartfoundation.org.au/for-professionals
http://www.heartonline.org.au
13B.11 Fitness for purpose
testing procedure.
13B.12 Supplementary references
• CDC. Cholesterol reference method laboratory network for the notional reference
system for cholesterol, certification protocols for clinical laboratories, 2004.
http://www.cdc.gov/labstandards/pdf/crmln/CertProtocolClinLabsMay04.pdf
• CDC. National reference system for cholesterol, HDL cholesterol certification for
manufacturers, 2002.
http://www.cdc.gov/labstandards/pdf/crmln/MFRHDLNov2002final.pdf
• National Heart Foundation of Australia and the Cardiac Society of Australia and
New Zealand. Position statement on lipid management 2005.
https://www.heartfoundation.org.au/for-professionals/clinical-information/lipid-
management
• CG Fraser. Biological Variation, from principles to practice. AACC press 2001.
• Westgard et al. Guest essays and biological variation database compiled by C

Ricos et al. Westgard QC at http://www.westgard.com/biodatabase1.htm
• RH Ng et al. Direct measurement of high-density lipoprotein cholesterol by the

Reflotron assay with no manual precipitation step. Clin Chem 1991; 37: 435.
• DA Wiebe and JO Westgard. Cholesterol – A model system to relate medical

needs with analytical performance. Clin Chem 1993; 39: 1504.
• WT Law et al. Whole blood test for total cholesterol by a self-metering, self-timing
disposable device with built-in quality control. Clin Chem 1997; 43: 384.
• C Luley et al. Point-of-care testing for triglycerides: Evaluation of the Accutrend

triglycerides system. Clin Chem 2000; 46: 287.
• M du Plessis et al. Analytical quality of near patient blood cholesterol and glucose
detarminations. Clin Chem 2000; 46: 1085.
• Centre for Evidence based Purchasing (CEP), NHS (National Health Service, UK),
A number on interesting evaluation reports and other information at:
(Search for the measurand you wish to investigate. Type “cholesterol” for
example, to obtain cholesterol measuring devices).
13B.13 Suggested reading for topic 13B
Suggested reading
• Fraser CG. Optimal analytical performance for point of care testing.

Clin Chim Acta 2001; 307: 37 (Article available in topic library).

disease. In: A Practical Guide to Global Point-of-Care Testing. Edited by Mark
• P Venge, B Lindahl and H Halls. Point-of-care testing for cardiac markers.

• Gialamas A; St John A; Laurence CO; Bubner T, and POCT, Management

Committee. Point of Care Testing for patients with diabetes, hyperlipidaemia
or coagulation disorders in the general practice setting: a systematic review.
Family Practice 2010;27:17 (Article available in topic library).
http://fampra.oxfordjournals.org/content/27/1.toc
• JR Burnett. Coronary artery disease: Lipid metabolism. Chapter 37 in, Clinical

Chemistry, theory, analysis and correlation. LA Kaplan and AJ Pesce, 5th
edition, 2010.
• MDS Shephard et al. Comparative performance of two point-of-care analysers

for lipid testing. Clin Lab 2007; 53: 561 (Article available in topic library).
Topic 14 – POCT for Infectious Disease

 Define the terms listed in bold throughout this unit
 Describe the main infections where POCTs are used
 Evaluate the role of POCT in the diagnosis of an infectious disease
 Interpret the results of a POCTs in infectious disease
14.2 Overview:
POCT in the area of infectious disease is limited in scope and number of tests performed
due to the need for high sensitivity (a measure of how many sick people are correctly
identified as positive). Many pathogenic bacteria can exist harmlessly on the body as part
of the natural human microbiota and thus some interpretation of the results needs to be
performed by highly trained laboratory staff. However POCT has made it possible to
provide faster turn-around-times for a number of infections and those POCTs currently
widely used will be covered in this unit. The POCTs outlined in this unit will be based on
either detecting the antibodies that patients produce in response to infection (blood is
usually the specimen) or detection of the actual infecting agent called an antigen detection
test (body fluids required related to the primary site of infection).
Pre-reading: Topics for pre-reading are listed in the section as a search term for Wikipedia
(Wikipedia (www.en.wikipedia.org). Wikipedia is useful for reading and learning, however it
should not be used as a reference. Other easily available internet references will be
supplied throughout the unit and in-text references will be given at the end of the unit.
For those of you with the recommended text: Point-Of-Care Testing (Editors: Price, St
John, Kricka), AACCPress, the relevant chapter is 33 (Listed as Marks, 2010 in the
reference list at the end of this unit). However you will be guided to selected readings that
are available on the internet and thus it is not mandatory to have the recommended text.
Page 1 of 17 Dr Danilla Grando 2011

The answers to ‘Thinking Point (TP)’ questions can be found at the end of this document.
Answers to the Reading questions must be obtained from the reading material. Review
question answers can also be found at the end of the document but please note, many of
the choices are correct – you must choose the best answer! The types of Exam questions
will be multiple choice and will be based on the objectives, the ‘Thinking point’ and
‘Reading’ and ‘Review’ questions.
14.3 Urinary Tract infections

Pre-reading: ‘Urinary tract infection’ on Wikipedia
Urinary tract infections are infections involving any part of the urinary tract and are
particularly common in women who are sexually active. Usually the patient will have lower
abdominal discomfort, pain on passing urine (dysuria) and difficulty passing urine so that
only small quantities of urine are passed (oligouria). The presence of bacteria in the
urinary tract will cause inflammation and hence the presence of leucocytes and
erythrocytes in the urine (Figure 1)
Leucocytes
Erythrocytes
s
Bacteria
Squamous epithelial cells
Figure 1. Urine microscopy (phase contrast) to count leukocytes, erythrocytes, epithelial

cells and microorganisms
14.3.1 Diagnosis of Urinary tract infections

Normally the bladder should be sterile. The urine must pass through the urethra in order to
be collected and tested. The lowest part of the urethra is colonized with skin microbiota
(skin flora) so there will always be a degree of contamination when urine is collected. To
minimize contamination a mid-stream urine (MSU) is collected.

Thinking point:
Look at the urine microscopy in Figure 1. There is evidence of inflammation due to the
presence of polymorphs and erythrocytes. However the presence of large flat epithelial
cells (squamous epithelia) indicates external skin contamination (urethral and bladder
cells are smaller and have larger nuclei). The bacteria that can be seen may have been in
the bladder or they may have entered the urine from the end of the urethra.
TP 1. If external skin bacteria can contaminate a urine sample, what will happen to the
numbers of bacteria present if the urine is not stored in the refrigerator before testing?
14.3.2 What role has POCT in UTI diagnosis?

POCT for the diagnosis of UTI are possible through dipstick testing. Dipsticks are
configured with a number of tests, the most common configurations include leucocyte
esterase, nitrate, protein, glucose and blood.
For this section we will concentrate on two tests; leucocyte esterase and nitrate as they
are both used as indicators of UTI.
Reading activity:
Download the following article on Urine Dipstick Testing and then answer the following
review questions.
http://archive.student.bmj.com/search/pdf/09/02/sbmj68.pdf
(Student British Medical Journal, February, 2009).
RQ 1. Before performing the urine dipstick test, what should be checked? (Hint: See
paragraph two). What needs to be done to perform the test correctly? (see same
paragraph)
RQ 2. What is the main purpose of using POCT in the diagnosis of UTI? (Hint: see last
paragraph in the section marked ‘Nitrites and leucocyte esterase’.
Issues arising from using dipsticks to identify patients requiring further testing
From the article you will have learnt that many bacteria need to be present in order to give
a positive nitrate test. However numbers of bacteria quickly increase when samples are
left at room temperature. In nursing homes, urine dipstick testing is very important as
patients with dementia or seriously ill patients may not be able to communicate that they
have the symptoms associated with a UTI. If samples are left at room temperature then

the dipstick will give a false positive result. Unfortunately it is often not possible to obtain a
mid-stream urine from these patients. If too much skin contamination is present, this will
result in a false positive leucocyte esterase test.
Quality control
Urine dipsticks must be stored according to manufacturer instructions and used within the
expiry period. No quality control is required if correct storage of test strips in used.
14.3.3 Review
1. What do the letters MSU stand for?
a) microscopy and sensitivity of urine
b) monthly screening of urine
c) mid-stream specimen of urine
d) minor screen of urine
2. Which of the following is most likely to lead to a false positive nitrite test?
a) patient has taken Vitamin C before having sample tested
b) a MSU was unable to be collected so there was excessive contamination
c) the pH of the urine is too high
d) the sample was not refrigerated
3. Which of the following is most likely to lead to a false positive leucocyte esterase
test?
a) patient has taken Vitamin C before having sample tested
b) a MSU was unable to be collected so there was excessive contamination
c) the pH of the urine is too high
d) the sample was not refrigerated
4. What is the main purpose of using POCTs for screening samples for UTI?
a) it means that more expensive laboratory tests can be avoided

b) samples negative for nitrite and leucocyte esterase need not be tested further
c) patients with positive samples for nitrite and leucocyte esterase can commence
antibiotics
d) patients can be quickly identified as infected and thus it avoids the delay of sending
a sample to an analytical laboratory
14.4 Upper respiratory tract infections

Pre-reading: First read about the bacteria that is the most common cause of bacterial sore
throat at http://www.textbookofbacteriology.net/streptococcus.html
Then read about Pharyngitis at http://www.aafp.org/afp/2004/0315/p1465.html

(American Family Physician, March 15, 2004)
You will notice in Table 1 of the American Family Physician article that the sensitivity and
specificity of each test is mentioned. If you do not understand these terms you should
look each one up in Wikipedia.
14.4.1 The diagnosis of pharyngitis

Traditionally throat swabs have been sent to the laboratory to culture for Group A
streptococcus (primarily but other bacteria can be responsible). The problem is that a
minimum of 18 hours is required for the sample to be cultured. Then further testing may be
required that requires a further 18 hours.
14.4.2 POCT for streptococcal pharyngitis

Recent improvements is the type of POCTs available for screening patients for Group A
strep in the clinic mean that properly performed negative POCTs no longer need to be sent
to a laboratory for confirmation (Mayes and Pichichero, 2001).
TP 1. What do you think it means to properly perform these POCTs?
14.4.3 Review
1. What causes the majority of cases of pharyngitis?
a) Streptococcus pyogenes otherwise known as Group A strep
b) Influenza virus
c) a large variety of viruses
d) golden staph

2. What is the main reason that POCT for pharyngitis is desirable in clinics?
a) pharyngitis is one of the most common reasons that people visit their doctor
b) if pharyngitis is not treated it could lead to serious complications
c) it can be difficult for the doctor to tell viral infections from bacterial infections
d) POCT can replace the laboratory diagnosis of pharyngitis
3. What is the main danger of performing POCTs for streptococcal pharyngitis?

a) some positive samples might be missed
b) the sensitivity of the test is insufficient
c) the specificity of the test is too low
d) a positive coloured line may be interpreted as a positive test
14.5 Lower respiratory tract infections

Patients with infections of the lower respiratory tract have their ability to exchange
respiratory gases compromised and thus these infections represent possible life-
threatening situations. Additionally patients can act as a source of highly infectious viruses
which pose an infection risk to other patients. Two of the most important viruses that need
to be identified are RSV (children) and Influenza (all age groups).
Pre-reading: Influenza on Wikipedia
This is a very good Wikipedia site. After reading you should know the difference between
Influenza A and Influenza B and be able to explain why influenza A is most likely to
cause pandemics.
After you feel you are an expert on influenza, you need to read about RSV (look up
‘Human Respiratory Syncytial Virus’ on Wikipedia. In the case of RSV you just need to
recognise that RSV is a serious respiratory infection in infants.
Specific antiviral agents are available for both influenza and RSV and thus early diagnosis
can lead to earlier implementation of appropriate therapy and a lowering of the infectious
dose carried by the patient.
14.5.1 The diagnosis of lower respiratory tract infections

Both viruses and bacteria cause serious infections of the lower respiratory tract. No
adequate POCTs exist to help direct therapy for bacterial lung infections. This is because

many of the bacteria that cause lung infections can be present in the upper respiratory
tract without causing the disease. Since the sample used in testing (sputum) is coughed
up and it makes contact with the upper respiratory tract, the presence of bacteria in the
sample may not correlate with the presence of infection in the lung.
On the other hand, viruses when present in the respiratory tract will indicate infection.
Luckily these viruses will be widespread through the respiratory tract when causing
infection so it is not necessary to get samples from the lower respiratory tract. In practice
we find that a nasopharyngeal aspirate or swab is the most important sample that gives
the best sensitivity of detection of the important respiratory viruses. In large hospitals,
there are a number of rapid tests that can be used that can identify a wider range of
viruses, however these tests may not be available out of hours. If the patient is not in the
hospital or the patient arrives at the Emergency Department out of hours, then there can
be delays in testing and thus POCT have been developed for Emergency Departments
and clinics.
14.5.2 POCT for lower respiratory tract infections

POCTs for Influenza viruses and RSV are amongst some of the most widely used POCTs,
particularly in Emergency Departments without 24 hour laboratory support.
Currently the POCTs available for Influenza and RSV have lower sensitivity than
laboratory tests, so POCT cannot be used to exclude these infections. However the
prevalence of these two groups of viruses makes these POCTs some of the most widely
used.
Interestingly there are some urine tests that can be used to diagnose some important
bacterial infections of the lower lungs. The most common of bacteria to cause community
acquired pneumonia (CAP) is Streptococcus pneumoniae. When a person has a serious
lung infection with bacteria, some of these bacteria may enter the circulation and be
attacked by white blood cells that can engulf them (phagocytic white blood cells). These
phagocytes then digest the bacteria, breaking them down. These bits of broken up
bacteria are released and contain antigens that can then be filtered by the kidney and thus
be detected in the urine using an antibody test. The problem with this test is that it can be
negative in early infection (Gray, 2010). In practice CAP is often a clinical diagnosis and
empirical antibiotic therapy is used until sputum culture results are received by the
laboratory. In the case of Legionellosis (see Wikipedia) the rapid urine antigen test can

result in rapid diagnosis. The prognosis of Legionnaires’ disease is linked to how early a
diagnosis is made making this POCT valuable. However because Legionnaires’ Disease is
not common, most Emergency Departments do not bother to perform this POCT and
rather request a laboratory to perform this test instead.
14.5.3 Review
1. Which of the following statements is correct?
a) Influenza B is a less serious condition than influenza A
b) Influenza A is a less serious condition that influenza B
c) A POCT should test for both influenza A and influenza B
d) There are no antiviral treatments available for influenza A
2. RSV is
a) a type of influenza virus
b) a serious respiratory infection in all age groups
c) a life threatening bacterial respiratory infection of infants
d) a life threatening viral respiratory infection of infants
3. POCT for Influenza and RSV

a) can be used to exclude infection with these viruses
b) can help direct antiviral therapy
c) if positive can rule out bacterial infections
d) do not need to be confirmed by laboratory tests
14.6 Genital infections

Genital infections can be classified as sexually transmitted infections or endogenous
infections (an organism part of the normal microbiota of the body at levels of growth
leading to symptoms). Sexually transmitted infections such as gonorrhoea, chlamydia
and syphilis have important public health consequences which include uncontrolled
spread if surveillance strategies are not used and possible transmission to the newborn if
mothers are infected while pregnant. Infected individuals may fail to return visit their health
provider, so POCT can ensure same day diagnosis and the prescription of appropriate
antimicrobial therapy. Another sexually transmitted infection that is not of great public
health importance but is easily diagnosed with a POCT is trichomoniasis. Infections that

are endogenous include yeast infections (thrush), and bacterial vaginosis which is a loss of
balance in the local flora that in women causes excessive malodorous discharge.
In order to understand the diagnosis of each of these infections, it is important first to
understand the characteristics of each of these responsible organisms. HIV will be dealt
with in section 14.7 Blood infections.
Neisseria gonorrhoeae
The disease gonorrhoea causes a purulent urethral discharge in most men, but women
may be asymptomatic. If women are not treated, the infection can ascend the upper gental
structures leading to inflammation of the peritoneum, scarring of the fallopian tubes,
possible ectopic pregnancy and infertility.
This is a Gram-negative diplococcus that lives on mucous membranes and does not
survive in the environment. It requires enriched agar that selects for this organism and an
atmosphere with 5% carbon dioxide in order for laboratory isolation. Isolates should be
grown in a diagnostic laboratory to check the susceptibility of isolates to currently used
antimicrobials.
Chlamydia trachomatis
Like gonorrhoea, chlamydia infection leads to a discharge that may also be asymptomatic
in women and cause the same complications.
Unlike gonorrhoea, this bacterium will not grow on enriched agar plates. It requires cell
culture for growth, however cell-culture has a low sensitivity. With the advent of molecular
biology techniques, the preferred method of diagnosis is PCR (polymerase chain reaction).
PCR is able to amplify nucleic acids such as DNA and thus these tests are often
abbreviated to NAATs (nucleic acid amplification test) Since culture of chlamydia is
difficult, these organisms are not routinely tested to report their susceptibility to antibiotics.
Syphilis
Syphilis starts off as a skin infection of the genitals. When one partner has a shallow wet
ulcer due to the syphilis bacterium and has sex with another person, the rubbing of the
sexual act can cause minor skin abrasion and thus the entry of the organism into the skin
of the partner. The organism multiplies in this shallow ulcer called a chancre and after a
period of up to 6 weeks from the time first infected the ulcer usually disappears.

Unfortunately the ulcer is painless so that if a woman has this ulcer in their vagina they
may not even realize they have been infected. When the ulcer disappears, it does not
mean that the patient is cured. From the skin the organisms travel around the body in the
blood stream. As soon as three weeks following the disappearance of the ulcer in
untreated patients a rash will appear and this is called secondary syphilis. This is a very
dangerous stage as the organism is blood borne and can cross the placenta. This is why
syphilis testing is one of the mandatory tests at the end of the first trimester of pregnancy.
After secondary syphilis, many infections will become latent. In some individuals this latent
infection can cause damage to special tissues and a patient may present up to 30 years
later with the symptoms of tertiary syphilis. These symptoms are mainly neurological and
can be as vague as dementia.
Trichomoniasis
Trichomonas vaginalis is a single cell protozoan that causes inflammation and malodorous
green discharge in women only.
Reading activity:
Go to the Centers for Disease Control (CDC) website to find factsheets on each of the
above sexually transmitted diseases. Then answer the following reading activity questions.
http://www.cdc.gov/std/default.htm
RQ 1. Which parts of the body need to be investigated for gonorrhoea infection?
RQ 2. How often should sexually active women (multiple partners) be screened for
Chlamydia?
RQ 3. What is the link between syphilis and HIV?
RQ 4. Can you catch trichomoniasis more than once?
14.6.1 Diagnosis of genital infections

Gonorrhoea
Treatment of gonorrhoea has changed over the years in response to the ability of this
organism to develop antibiotic resistance. Laboratory testing thus still relies on the

traditional methods of organism isolation on agar plates and then antibiotic susceptibility
testing.
Chlamydia
The most sensitive method is NAAT as mentioned above.
Syphilis
A blood test is required for screening of patients. The screening blood test has high
sensitivity but poor specificity. Many tests that are positive in the screening test will be
negative on confirmatory testing due to the phenomenon of biological false positives
(BFPs). The high rate of BFS arises from the fact that the screening test does not test for
antibodies to syphilis, rather it tests for cardiolipin antibodies which are formed when
syphilis organisms damage tissue. These antibodies can also be formed in other
conditions such as Rheumatoid Arthritis and during pregnancy. As syphilis is an important
diagnosis to be made during pregnancy it is important that all reactions in the screening
test be forwarded for further confirmatory testing.
Trichomoniasis
Traditionally the diagnosis of trichomoniasis has been too take the swab and make a
microscopic preparation so that the swimming cells of Trichomonas can be seen under the
microscope. You can see a movie of trichomonads under the microscope at
http://www.youtube.com/watch?v=9B1PFJXuJUs
The problem with viewing for live organisms is that they need to be alive! If long delays in
the transport of swabs occur then this can mean that the organisms are not easily seen as
they resemble white blood cells when they roll up and die.
14.6.2 POCT testing for genital infections

Neisseria gonorrhoea and Chlamydia trachomatis
POCTs for gonorrhoea and chlamydia are based on immunoassay and can give positive
results quickly, however these tests have much lower sensitivity than NAATs, so these are
only recommended where NAATs are not available.

Syphilis
Since the RPR is a simple card based format, it can be used as a POCT. The only draw
back is that it is performed on serum. All positive RPRs must be confirmed by a laboratory
due to the high BFPs.
Trichomoniasis
As mentioned above, the sensitivity of microscopy for trichomoniasis is less that 70%.
Simple POCT available for trichomoniasis have sensitivities of around 80% (Gray, 2010)
and are thus particularly useful for clinics specialising in STIs (sexually transmitted
infections).
Bacterial Vaginosis
The most simple POCT for bacterial vaginosis is microscopy to see the shift it the type of
flora present. In clinics where there is high turnover of genital specimens, it may be
possible to train health care staff in the interpretation of the microscopy. Unfortunately the
typical POCT devices available to detect this condition do not offer advantages over
simple microscopic procedures. One POCT that can be used simultaneously for Candida,
trichomoniasis and bacterial vaginosis is based on DNA probe technology and has good
sensitivity and specificity (Gray, 2010).
14.6.3 Review
1. What is a biological false positive in regards to syphilis testing?

a) this is a false positive test due to poor transport conditions
b) this is a false positive test following successful treatment of syphilis infection
c) this is a positive test that occurs during latent infection
d) it is a positive screening test that is negative in the confirmatory test
2. When is a POCT particularly useful for Trichomonas testing?

a) In resource poor settings
b) In STI clinics
c) When patients are pregnant
d) When deciding to give treatment

3. Chlamydia POCTs are more useful than Neisseria gonorrhoeae POCTs because
a) Chlamydia cause more infections than Neisseria
b) Chlamydia needs to be treated more urgently than Neisseria
c) Chlamydia does not need susceptibility tests to determine which antibiotics to use
d) Chlamydia cannot be grown in the laboratory
4. Syphilis POCT is useful because

a) syphilis is very common
b) syphilis is difficult to treat
c) syphilis may become resistant to antibiotics
d) syphilis is a common screening test in pregnancy
5. What do candida, trichomonas and Bacterial Vaginosis all have in common?

a) They are a common cause of vaginal discharge
b) They are all endogenous infections
c) They are always sexually transmitted
d) They are a common cause of discharge in men and women
14.7 Blood infections
HIV
HIV has become one of the top three causes of death (the other two being malaria and
tuberculosis). Because diagnosis is key to the implemention of prevention strategies, there
has been a big push to get POCT adopted in resource poor settings. It is important to
realise that all positive primary tests for HIV that are antibody tests, must be confirmed by
a laboratory test. These antibody primary tests are called screening tests and are
characterized by high sensitivity (very few positives are missed), but low specificity (some
of these positives may be false positives!).
14.8 Gastrointestinal infections

Very few POCT have been developed for common gastrointestinal infections, particularly
those leading to gastroenteritis symptoms. Most cases of gastroenteritis are causes by
viruses and there are a large variety of these viruses. However notable outbreaks have
occurred in hospital settings amongst infants and it is important that these not spread due
to the vulnerable population which could quickly succumb to death due to dehydration.

Of particular note are two groups of viruses that commonly cause outbreaks, namely
rotavirus and adenovirus.
Infection of the gastrointestinal tract can also occur in the stomach due to colonisation of
the stomach by Helicobacter pylori. Some strains of this bacterium carry toxin genes that
can destroy stomach lining cells and can lead to gastric ulcers. Some of these gastric
ulcers may progress to cancer. Because treatment involves a cocktail of antibiotics,
diagnosis of this infection must precede the prescribing of antibiotics. You can read more
about the discovery of this microorganism which lead to the awarding of a Nobel Prize at
http://www.helico.com/.
Reading activity:
Go to the Victorian Government’s Health Information Website’s Immunisation Fact Sheet
for Rotavirus http://www.health.vic.gov.au/immunisation/fact-sheets/factsheets/rotavirus
Then read the Centers for Disease Control website’s information on adenovirus at
http://www.cdc.gov/ncidod/dvrd/revb/respiratory/eadfeat.htm
RQ 1. Why is immunization recommended for Rotavirus?
RQ 2. How is rotavirus infection confirmed?
RQ 3. What are some of the prevention strategies recommended for adenovirus and how
are these related to safe handling of specimens in the laboratory?
14.8.1 Diagnosis of gastroenteritis infections

As most cases of gastroenteritis are self-limiting and not requiring treatment other than
rehydration therapies, laboratory diagnosis will only be warranted in severe cases or to
determine which patients may be infectious to others in healthcare institutions. Laboratory
diagnosis usually involves submitting a faecal sample to the laboratory and the common
pathogens investigated are Salmonella and Campylobacter.
RQ 4. Hopefully you are familiar with Salmonella as a cause of gastroenteritis. Are you
familiar with campylobacter infection? Go to the Victorian State Government’s Department
of Health IDEAS website (Infectious Disease Epidemiology and Surveillance)
http://www.health.vic.gov.au/ideas/diseases/gas-dcamp
What food is the main culprit of causing campylobacter infection when undercooked?

14.8.2 POCT for gastroenteritis infections
POCT test have been developed that detect Rotavirus and Adenovirus directly from stool.
TP. 1 From the online folder where you downloaded this unit description, you will find a
package insert for a POCT for “Adeno_Rota_method”. From this method, why are there
results shown for a variety of invalid test (explain each one).
Helicobacter pylori
This organism has now been shown to be the main cause of gastric and duodenal ulcer.
Stool testing can be used to detect Helicobacter antigen. Active infection of H Pylori can
be determined by a urea breath test.
Answers to Thinking Point Questions

Section 14.3. Urine
TP 1: Even if low numbers of bacteria contaminate the sample, their numbers will increase
dramatically at room temperature. This means that after a period of time, the original
insignificant count of bacteria will become large enough to now be considered as
significant and the report may read that the results are consistent with Urinary tract
infection. This means that some patients will receive unnecessary antibiotic treatment.
Section 14.4 Upper Resp

TP1: Many POCT have an internal control. If for example it is a test that produces coloured
lines, one coloured line will be for an internal control (to check all the reagents have been
added) and the second line will be for the patients sample and will either be positive (so
two lines appear) or negative (only one line will appear, the control line). If untrained
persons are used to perform the test (for instance agency staff unfamiliar or untrained with
the kit), then they mistake the appearance of one line as a positive for the sample!
The second form of quality control that is used is an external control. Depending on how
often the kit is used, it is usually recommended that once a week or if the kit has not been
used in a week that an external control be performed before the kit is used. The external
control is usually supplied with the kit and instead of using the patient’s swab, you test the
external control. This is particularly important to satisfy regulatory authorities and the date

and results of external testing must be recorded and kept for a period of time as specified
by inspecting authorities.
Section 14.8 Gastro
TP1: In order from First Left invalid to last
First left: Something in the specimen is interfering with the test, so no bands are appearing
and thus you don’t know whether a positive has been missed. Report as specimen found
to be inhibitory ino POCT. Suggest repeat specimen collection of referral of specimen to
analytical testing laboratory.
Second third and last picture of invalid tests: these are missing the band for the control.
This indicates that the test is inhibitory to the control and thus any other results may be
unreliable. Suggested reporting: Specimen results were not interpretable in the POCT due
to the sample being inhibitory to the test control. Suggest repeat specimen collection of
referral of specimen to analytical testing laboratory.
Answers to Review Questions:
Section 14.3 Urine
1. c 2. d 3. b 4. b
Section 14.4 Upper Resp
1. c 2. b 3. a
Section 14.5 Lower Resp
1. a 2. d 3. b
Section 14.6 Genital
1. d 2. b 3. c 4. d 5. a
References
Gray, JW. 2010. Point-of-Care Testing for Infectious Diseases. In Point of Care Testing, eds, Price, St John,
Kricka, AACCPress, Washington.

Mayes T, Pichichero ME. 2001. Are follow-up throat cultures necessary when rapid antigen detection tests
are negative for group A streptococci? Clin Pediatr [Phila]., 40:191–5.

2018 POCT Course Notes

Uploaded by

Copyright:

Available Formats

2018 POCT Course Notes

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2018 POCT Course Notes

Uploaded by

Copyright:

Available Formats

Course Notes

Point of Care Patient Testing

ONPS 2425 and ONPS 2426

Topics I to XIII written by Mr Ian Farrance, FAACB

School of Health & Biomedical Sciences (HE)

With support from

School of Vocational Health and Sciences (TAFE)

R. Greaves POCT Page 2

Course Code: ONPS2425 / ONPS2426

Topic 1 – Introduction to POCT

 Course learning objectives

1.1 Learning objectives

On completion of the course, participants will be able to:

Page 1 of 11 Ian Farrance

Page 2 of 11 Ian Farrance

1.3 Situations in which POCT may be advantageous

In Australia, POCT testing may be provided in a central laboratory environment by skilled

Page 3 of 11 Ian Farrance

1.5 Visually read indicator strips (“dip-stick”)

Page 4 of 11 Ian Farrance

1.6 Technology and overview with respect to types of POCT devices

In general, POCT may be categorised as hand-held or bench-top procedures. Hand-held devices as

Prior to the early 1980’s, the main POCT applications were …

Page 5 of 11 Ian Farrance

Astrup trolley acid-base Siggaard Andersen acid-base

Page 6 of 11 Ian Farrance

1960 • PO2 electrodes

• Semi-automated blood • Whole blood biosensors

1.7 Functionality of hand-held devices and bench top devices

Page 7 of 11 Ian Farrance

1.8 Policies, procedures and guidelines for POCT

1.9 Where to obtain information

Page 8 of 11 Ian Farrance

1.9.2 Internet sites

Page 9 of 11 Ian Farrance

1.9.3 Text books and general references

Some useful general text books and references are …

1.9.4 POCT course electronic library

Page 10 of 11 Ian Farrance

Additional reading and supplementary references have been organized as follows:

1.11 Suggested reading

Suggested reading for topic 1

I Farrance. Review – Policies, procedures and guidelines for point-of-care testing.

Course Code: ONPS2425 / ONPS2426

Market segments for point-of-care testing devices

In vitro diagnostic devices

Global market segments for in vitro diagnostic and POCT devices

Page 1 of 4 Ian Farrance

Page 2 of 4 Ian Farrance

European POCT market segments, 2009

Segments of the Western European POC market by discipline in 2009

A useful feature provided by this type of information, is an understanding of the relative

Page 3 of 4 Ian Farrance

 JA DuBois. Point-of-care testing recognised as a distinct domain.

 JF Dooley. Point-of-care diagnostic testing markets.

Page 4 of 4 Ian Farrance

Course Code: ONPS2425 / ONPS2426

Topic 2 – Pre-Analytical considerations

• Topic 2 learning objectives

2.1 Topic 2 learning objectives