Module 3a - Haemodialysis - 4th Edition PDF

CLINICAL PRACTICE GUIDELINES
MODULE 2: HAEMODIALYSIS
UK Renal Association,
4th Edition, 2007
Final Version
Dr. Robert Mactier,

Consultant Nephrologist,
Glasgow Royal Infirmary,
NHS Greater Glasgow & Clyde
MODULE 2: CLINICAL PRACTICE GUIDELINES FOR HAEMODIALYSIS 2
Contents
INTRODUCTION 3
SUMMARY OF CLINICAL PRACTICE GUIDELINES FOR HAEMODIALYSIS

5
1 Haemodialysis facilities (Guidelines DIAL-HD 1.1 – 1.6) 5
2 Haemodialysis equipment and disposables (Guidelines DIAL-HD 2.1 – 2.3) 6
3 Concentrates and water for haemodialysis (Guidelines DIAL-HD 3.1 – 3.4) 6
4 Haemodialysis membranes (Guidelines DIAL-HD 4.1 – 4.6) 7
5 Haemodialysis dose, frequency and duration (Guidelines DIAL-HD 5.1 – 5.7) 8
6 Laboratory and clinical indices of dialysis adequacy other than dialysis dose (Guidelines
DIAL-HD 6.1 – 6.11) 9
7 Vascular access (Guidelines DIAL-HD 7.1 – 7.16) 11
8 Access to and withdrawal from dialysis (Guidelines DIAL-HD 8.1 – 8.5) 13
SUMMARY OF AUDIT MEASURES FOR HAEMODIALYSIS 14
FULL CLINICAL PRACTICE GUIDELINES FOR HAEMODIALYSIS 16

1 Haemodialysis facilities (Guidelines DIAL-HD 1.1 - 1.6) 16
2 Haemodialysis equipment and disposables (Guidelines DIAL-HD 2.1 – 2.3) 23
3 Concentrates and water for haemodialysis (Guidelines DIAL-HD 3.1 – 3.4) 25
4 Haemodialysis membranes (Guidelines DIAL-HD 4.1 - 4.6) 34
5 Haemodialysis dose, frequency and duration (Guidelines DIAL-HD 5.1 – 5.7) 42
6 Laboratory and clinical indices of dialysis adequacy other than dialysis dose (Guidelines
DIAL-HD 6.1 – 6.11) 55
7 Vascular access (Guidelines 7.1 - 7.16) 63
8 Access to and withdrawal from dialysis (Guidelines DIAL-HD 8.1 - 8.5) 76
ACKNOWLEDGEMENTS AND DECLARATIONS OF INTEREST 83

Introduction
The basis for the management of advanced chronic kidney disease is the
seamless integration of renal replacement therapy (HD, peritoneal dialysis,
and transplantation) with evidence based medical treatment of its
complications. The National Service Framework Part 1: Dialysis and
Transplantation has stressed the need for a patient-centred approach in the
planning and provision of renal replacement therapy with an emphasis on
patient education and choice as well as the provision of adequate resources for
elective access surgery, dialysis and transplantation (1). It also identified that a
small proportion of patients after counseling may opt for optimal conservative
medical therapy without planning to initiate dialysis.
Innovations and changes in HD practice have seldom been underpinned by

adequately powered randomised trials. Nevertheless, day-to-day clinical
decisions on HD are required and standards need to be set on the best available
evidence. Consequently clinical practice guidelines for HD have been
developed in Australasia, Canada, Europe and the USA as well as the UK (2-
17). These guidelines serve to identify and promote best practice in the
delivery of HD and have set clinical standards to allow comparative audit of
the key aspects of the HD prescription, laboratory data and patient outcomes.
The reports of the UK Renal Registry, Scottish Renal Registry and NHS
Quality Improvement Scotland have demonstrated the benefits of performing
regular audit to improve clinical standards in HD (2-4).
This module provides an expansion on the 2002 guidelines in HD to

incorporate sections on patient-centred HD facilities and initiation of dialysis,
an expansion on the section on vascular access and, most importantly, an
update on the current guidelines based on evidence from new studies. The
USA (NKF-KDOQI) and European (EBPG) guidelines have also been
updated recently (9, 11, 13) and standardisation with these and other
international guidelines has been attempted whenever possible. This module
promotes the adoption of a range of standardized audit measures in HD and
has been designed to permit easy modification on the website to incorporate
future changes in practice recommendations based on evidence from new
research. The proportions of patients who should achieve clinical and
laboratory targets have not been specified for most of the clinical practice
guidelines. This approach is designed to allow for greater achievement of audit
measures in parallel with improvements in clinical practice.
References
1 The National Service Framework for Renal Services Part 1: Dialysis and
Transplantation, Department of Health, London, UK, January 2004.
(www.doh.gov.uk/nsf/renal/index.htm)
2 Clinical Standards for Adult Renal Services, NHS Quality Improvement
Scotland, March 2003. (www.clinicalstandards.org)
3 Renal Association Standards & Audit Subcommittee "Treatment of adults &

children with renal failure - Standards and audit measures". 3rd Edition, London:
Royal College of Physicians 2002. (www.renal.org/Standards/standards.html)
4 Report of NHS Quality Improvement Scotland (www.nhshealthquality.org)
5 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for chronic
kidney disease: Evaluation, classification and stratification. Am J Kidney Dis
2002; 39: 2 Supplement 1 S1-S266.
(www.kidney.org/professionals/kdoqi/guidelines.cfm)
6 Canadian Society of Nephrology Clinical Practice Guidelines. JASN 1999; 10:
Supplement 13 (http://csnscn.ca)
7 CARI (Caring for Australians with Renal Impairment) Guidelines Part 1 -
Dialysis Guidelines. Eds: Knight J and Vimalachandra D, Excerpta Medica
Communications, 2000 (www.kidney.org.au/cari/)
8 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for
hemodialysis adequacy, Update 2000. Am J Kidney Dis 2000; 37:1 Supplement 1
S7-S62 (www.kidney.org/professionals/kdoqi/guidelines.cfm)
9 National Kidney Foundation-K/DOQI Clinical Practice Guidelines and Clinical
Practice Recommendations for Haemodialysis Adequacy, 2005 (in press)
10 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for vascular
access 2000. Am J Kidney Dis 2000; 37: 1 Supplement 1 S137-S180
access, Update 2005. (in press)
12 European Best Practice Guidelines for haemodialysis Part 1. Nephrol Dial
Transplant 2002; 17: Supplement 7 S1-S111.
(http://ndt.oupjournals.org/content/vol17/suppl_7/index.shtml)
Transplant 2005;20 (suppl 5) 148-155
(http://ndt.oupjournals.org/content/vol17/suppl_7/index.shtml).
14 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for bone
metabolism and disease in chronic kidney disease. Am J Kidney Dis 2002;39:2
Supplement (www.kidney.org/professionals/kdoqi/guidelines.cfm)
15 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for managing
dyslipidaemias in chronic kidney disease. Am J Kidney Dis 2003; 41: 4
Supplement 3 S1-S92. (www.kidney.org/professionals/kdoqi/guidelines.cfm)
16 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for nutrition
of chronic renal failure. Am J Kidney Dis 2001; 37: 1 Supplement 2 S66-S70.
17 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for anaemia of
chronic kidney disease. Am J Kidney Dis 2001; 37: 1 Supplement 1 S182-S236.
Summary of clinical practice guidelines

for haemodialysis
1 Haemodialysis facilities (Guidelines DIAL-

HD 1.1 – 1.6)
Guideline 1.1 DIAL-HD

The specification of new or refurbished haemodialysis facilities should adhere
to the guidelines that are described in the NHS Estates Health Building Note
53: Volumes 1 & 2.

The haemodialysis facility should have sufficient specialist support staff to
fulfill the criteria listed by the Renal Workforce Planning Group 2002.

Except in remote geographical areas the travel time to a haemodialysis
facility should be less than 30 minutes or a haemodialysis facility should be
located with 25 miles of the patient’s home.
In inner city areas travel times over short distances may exceed 30 minutes at peak
traffic flow periods during the day.

Haemodialysis patients who require transport should be collected from home
within 30 minutes of the allotted time and be collected to return home within
30 minutes of finishing dialysis.

All patients who may be suitable for home dialysis should receive full
information and education about home haemodialysis.
Home haemodialysis training is not available in all renal units and some patients
may need to travel to a sub-regional or regional centre to pursue their choice to
train for home haemodialysis.

Haemodialysis capacity in satellite and main renal units within a
geographical area should increase in step with predicted need. To allow for
patient choice regarding out of hours haemodialysis schedules, provision of
holiday haemodialysis and expansion in patient numbers calculation of the
required number of haemodialysis stations should be based on using each
station for 2 patients per day three times per week.
The national average number of hospital haemodialysis patients per million
catchment population reported for the previous year by the UK Renal Registry
should be regarded as the minimum capacity for haemodialysis in each
geographically based renal service. Alternatively up-to-date regional data may be
used. For example the national average provision for 312 hospital haemodialysis
patients (78 stations) per million catchment population in Scotland at the end of
2005 may be regarded as a minimum haemodialysis capacity in all regions in
2006. The level of hospital haemodialysis provision will need to be higher in areas
with a high ethnic and/or elderly population and increase nationwide over the next
10 years.
2 Haemodialysis equipment and disposables

(Guidelines DIAL-HD 2.1 – 2.3)

All equipment used in the delivery and monitoring of therapy should comply
with the relevant standards for medical electrical equipment. General safety
standards are covered by BS EN 60601-1: 2006 and specific dialysis machine
requirements are covered by BS-EN 60601-2-16: 1998 (Medical electrical
equipment: Particular requirements for the safety of haemodialysis (HD),
haemodiafiltration and haemofiltration equipment).

Disposables such as dialysers and associated devices are classified as medical
devices and should display the CE mark.
The presence of such a mark signifies compliance with the requirements of the
statutory Medical Device Directive and also national and international standards
where they exist for new products: BS-EN 1283: 1996 (haemodialysers,
haemodiafilters, haemofilters, haemoconcentrators and their extra corporeal
circuits), ISO 8638:2004 (Extracorporeal blood circuit for haemodialysers,
haemodiafilters and haemofilters) or ISO 13960: 2003 (Plasma filters).

Machines should be replaced after between seven and ten years’ service or
after completing between 25,000 and 40,000 hours of use for haemodialysis,
depending upon an assessment of machine condition.
3 Concentrates and water for haemodialysis


Ready made concentrates are classified as medical devices and should display
the CE mark. Concentrates that are manufactured ‘in house’ should meet the
requirements of BS EN 13867: 2002 (Concentrates for haemodialysis and
related therapies) (good practice).

Water used in the preparation of dialysis fluid should, as a minimum, meet
the requirements stated in Table 1 for chemical and microbiological
contaminants.
After consultation within the UK the limits for chemical contaminants are derived
from AAMI RD-52 2004 (1), ISO 13959:2002 (2) and the European
Pharmacopoeia (3), whilst the limits for bacterial counts (100 cfu/ml) and
endotoxin (0.25 IU/ml) are based on the European Pharmacopoeia (3) and the
European Renal Association Best Practice Guidelines (4). New equipment should
be capable of producing ‘ultrapure’ dialysis fluid (bacterial counts <0.1 cfu/ml
and endotoxin <0.03 IU/ml) in order to meet the best practice guidelines. Ideally
this should be achieved using ultrapure water; however water that meets the
minimum standard in Table 1 can be used together with point of use filtration of
the dialysis fluid. If routine monitoring demonstrates continuous contamination in
excess of the desired levels, a programme to improve this should start
immediately (good practice).

A routine testing procedure for water for dialysis should form part of the
renal unit policy.

The dialysate should contain bicarbonate as the buffer.
4 Haemodialysis membranes (Guidelines

DIAL-HD 4.1 – 4.6)

The balance of evidence supports the use of low flux synthetic and modified
cellulose membranes instead of unmodified cellulose membranes.
The benefits of low flux synthetic and modified cellulose membranes over
unmodified cellulose membranes are limited to advantages arising from different
aspects of improved biocompatibility rather than better patient outcomes.

The balance of evidence supports the use of a dialysis regimen with enhanced
removal of middle molecules in incident patients who are predicted to remain
on haemodialysis for several years and prevalent patients who have been on
haemodialysis for more than 3.7 years. Such patients are at risk of developing
symptoms of dialysis-related amyloidosis.
Treatments with better clearance of middle molecules include haemodialysis with
high flux synthetic membranes and haemodiafiltration. The proven benefits of
high flux synthetic membranes in randomized trials are limited to advantages
arising from improved biocompatibility and enhanced removal of middle
molecules, such as beta-2-microglobulin, rather than better patient survival rates.
Chronic high flux dialysis in the HEMO study did not affect the primary outcome
of all cause mortality or any of the secondary composite outcome measures
including the rates of first cardiac hospitalization or all cause mortality, first
infectious hospitalization or all cause mortality, first 15% decrease in serum
albumin or all cause mortality, or all non-vascular access-related hospitalizations.
.
Patients without increased bleeding risk should be given low-dose
unfractionated heparin or LMWH during haemodialysis to reduce the risk of
clotting of the extracorporeal system.
For patients with a risk of bleeding anticoagulation should be avoided or kept to a
minimum by using a high blood flow rate and regular flushing of the
extracorporeal circuit with saline every 15-30 minutes.

If it is planned to reuse dialysers that are marked ‘for single use only’ the
implications of dialyser reuse need to be considered carefully after reading
MDA Device Bulletin DB 2000(04) Single-use medical devices: implications
and consequences of reuse.

The use of dialysers sterilized with ethylene oxide should be avoided.

Haemodialysis patients should not be treated with ACE inhibitor drugs and
AN 69 dialyser membranes at the same time.
5 Haemodialysis dose, frequency and

duration (Guidelines DIAL-HD 5.1 – 5.7)

HD should take place at least three times per week in nearly all patients with
end-stage chronic renal failure.
Reduction of dialysis frequency to twice per week because of insufficient dialysis
facilities is unacceptable.

Every patient with end-stage chronic renal failure receiving thrice weekly HD
should have consistently:
either urea reduction ratio (URR) > 65%
or equilibrated Kt/V of >1.2 (or sp Kt/V of > 1.3) calculated from pre-
and post-dialysis urea values, duration of dialysis and weight loss during
dialysis.
To achieve a URR above 65% or eKt/V above 1.2 consistently in the vast majority
of the haemodialysis population clinicians should aim for a minimum target URR
of 70% or minimum eKt/V of 1.4 in individual patients. Aiming for these target
doses also addresses the concerns raised by recent data which suggest that women
and patients of low body weight may have improved survival rates if the URR is
maintained above 70% or eKt/V is at least 1.4.

The duration of thrice weekly HD in adult patients with minimal residual
renal function should not be reduced below 4 hours without careful
consideration.

Patients receiving dialysis twice weekly for reasons of geography should
receive a higher sessional dose of dialysis.
If this cannot be achieved, then it should be recognised that there is a compromise
between the practicalities of dialysis and the patient’s long-term health.

Measurement of the ‘dose’ or ‘adequacy’ of HD should be performed
monthly in all hospital HD patients and may be performed less frequently in
home HD patients.
All dialysis units should collect and report this data to their regional network and
the UK Renal Registry.

Standardisation of the method of post-dialysis blood sampling is essential
since all measurements of dialysis dose require the measurement of the post-
dialysis blood urea concentration. Post-dialysis blood samples should be
collected either by the stop-dialysate flow method, the slow-flow method or
the simplified stop-flow method.
The method used should remain consistent within renal units and should be
reported to the Registry.

Patients with acute renal failure should initially receive daily renal
replacement therapy.
The frequency of renal replacement therapy may be reduced once the metabolic
syndrome and fluid status of patients with acute renal failure is stable.
6 Laboratory and clinical indices of dialysis

adequacy other than dialysis dose

Blood sampling for biochemical and haematological measurements should be
performed before a mid-week HD session using a dry needle or syringe.

Monitoring of pre-dialysis biochemical and haematological parameters
should be performed monthly in hospital HD patients and at least 3 monthly
in home HD patients.

Pre-dialysis serum bicarbonate concentrations measured with minimum
delay after venepuncture should be between 20 and 26mmol/l.

Pre-dialysis serum potassium should be between 3.5 and 6.5 mmol/l in HD
patients.

Pre-dialysis serum phosphate should be between 1.1 and 1.8mmol/l.

Pre-dialysis serum calcium, adjusted for serum albumin, should be within the
normal range.

Pre-dialysis serum albumin corrected calcium x phosphate product should be
less than 4.8 mmol2/l2.

Serum PTH levels should be more than twice and less than 4 times the upper
limit of normal for the intact PTH assay used.
Serum PTH levels do not need to be performed routinely more often than every 3
months.

Serum aluminium concentration should be measured every three months in
all patients receiving oral aluminium containing phosphate binders.

Pre-dialysis haemoglobin concentration should be 10.5-12.5g/dl.

Data on the frequency of dialysis-related hypotension, defined as an acute
symptomatic fall in blood pressure during dialysis requiring immediate
intervention to prevent syncope, should be collected and audited.
7 Vascular access (Guidelines DIAL-HD 7.1 –

7.16)

To preserve veins for creation of vascular access venepuncture or insertion of
peripheral venous cannulae should be avoided in the forearm or arm of all
patients with advanced renal failure whenever possible.

The preferred mode of vascular access for HD patients is a native
arteriovenous fistula.

There should be enough dedicated theatre sessions for access surgery to
provide one session per week for every 120 patients on dialysis. With this
level of access surgery provision no patient on dialysis, including those
patients who present late, should wait more than four weeks for fistula
construction.

Patients should undergo fistula creation between 6 and 12 months before
haemodialysis is expected to start to allow time for adequate maturation of
the fistula or time for a revision procedure if the fistula fails or is inadequate
for use.

The time to first cannulation of an AVF should be a minimum of 1 month
and preferably at least 2 months after creation.
First cannulation may be considered between 2 and 4 weeks after creation of an
AVF if this is the alternative to insertion of a central venous catheter and a
nephrologist or experienced haemodialysis nurse has assessed that the fistula has
matured adequately for use for dialysis.

At least 65% of patients presenting more than three months before initiation
of dialysis should start HD with a usable native arteriovenous fistula.

Investigation of the AVF or graft to assess for evidence of arterial or venous
stenoses or access recirculation is required if there is a significant fall in the
blood flow rate that can be achieved, a reduction in delivered dialysis dose or
a persistent rise in venous pressure in sequential dialysis sessions.

All patients should be evaluated for a secondary arteriovenous access after
each episode of access failure.

As few HD patients as possible should rely on central venous catheters for
vascular access.
As an audit measure less than 20% of patients on long-term HD should use
tunneled or non-tunneled central venous catheters as the form of vascular access.

Cuffed, tunneled double-lumen central venous catheters are preferred if
temporary vascular access is likely to be needed for more than 3 weeks. Non-
cuffed double-lumen catheters may be used if temporary vascular access for
haemodialysis is predicted to be required for less than 3 weeks.

The preferred insertion site for central venous catheters is the internal
jugular vein and the catheter should not be placed on the same side as a
planned or maturing upper limb arterio-venous access, whenever possible.

All renal units should use real-time ultrasound to guide insertion of central
venous catheters.

All renal units should have protocols to ensure that full barrier precautions
are followed during insertion of temporary and tunneled central venous
dialysis catheters.

All central venous catheter connections and disconnections should be
performed under aseptic conditions by trained staff.

Peripheral and central line blood cultures should be taken prior to starting
antibiotics in all cases of suspected catheter-related infection.

All HD units should collect and audit data on the form of vascular access in
use in incident and prevalent haemodialysis patients and the rates of
bacteraemia per 1000 patient days using central venous catheters, arterio-
venous grafts and arterio-venous fistulae.
8 Access to and withdrawal from dialysis


All patients with advanced renal failure (eGFR < 30ml/min), who have a life
expectancy of more than 3 months, should be considered for renal
replacement therapy and should be referred to a nephrologist.

If there is no medical contraindication the choice of initial dialysis modality
should be based on patient choice.

After full education and counseling a small proportion of patients may opt
for active non-dialytic management of advanced chronic kidney disease,
including nutritional, medical and psychological support, rather than plan to
initiate dialysis.
The numbers of patients not taken on to dialysis and the reasons for this decision
should be subject to audit.

Renal replacement therapy should be commenced when a patient with an
eGFR < 15ml/min/1.73m2 has symptoms or signs of uraemia, fluid overload
or malnutrition in spite of medical therapy and considered carefully when an
asymptomatic patient has an eGFR < 6ml/min/1.73m2.

Any decision to discontinue haemodialysis should be made jointly by the
patient and the responsible consultant nephrologist after consultation with
relatives, the family practitioner and members of the caring team.
The decision, and the reasons for it, must be recorded in the patient’s notes. Renal
units should develop guidelines for palliative care of such patients, including
liaison with community services.
Summary of audit measures for

haemodialysis
1 The distance and travel time between the patient’s home and the nearest
satellite or main haemodialysis unit.
2 The waiting time after arrival before starting dialysis and the waiting
time for patient transport after the end of haemodialysis.
3 The proportion of dialysis patients in the main renal unit and its satellite
units who are on home haemodialysis.
4 The number of haemodialysis patients and number of haemodialysis

stations in the main renal unit and its satellite units expressed per million
catchment population.
5 The proportion of patients in the main renal unit and its satellite units
who are on twice weekly haemodialysis.
6 Cumulative frequency curves of urea reduction ratio measured using a

standard method of post-dialysis sampling.
7 The proportion of patient non-attendances for haemodialysis sessions and

the proportion of dialysis sessions shortened at the patient’s request.
8 Cumulative frequency curves of pre-dialysis serum potassium

concentration.
9 Cumulative frequency curves of pre-dialysis serum calcium, phosphate

calcium x phosphate product and PTH concentrations.
10 Cumulative frequency curves of pre-dialysis haemoglobin concentration.
11 The incidence of symptomatic hypotensive episodes during dialysis

sessions.
12 The proportion of prevalent patients on long-term haemodialysis who use

an arterio-venous fistula, arterio-venous graft and tunneled or non-
tunneled central venous catheters as the mode of vascular access.
13 The number of dedicated renal failure access surgery sessions per 120
dialysis patients.
14 The dates of first referral to nephrology, referral for creation of vascular

access and creation of first vascular access and the date and mode of
vascular access at the initiation of dialysis should be recorded and audited
in all incident chronic haemodialysis patients.
15 The rates of bacteraemia (and specifically the rates of MRSA

bacteraemia) observed per 1000 patient days using central venous
catheters, arterio-venous grafts and arterio-venous fistulae.
16 The proportion of patients with advanced renal failure (CKD stage 5)

who are treated with conservative medical therapy.
17 Record of the serum creatinine, estimated GFR and co-morbidity at

initiation of chronic renal replacement therapy (dialysis or
transplantation).
Full clinical practice guidelines for

haemodialysis
1 Haemodialysis facilities (Guidelines DIAL-

HD 1.1 - 1.6)

The specification of new or refurbished haemodialysis facilities should adhere
to the guidelines that are described in the NHS Estates Health Building Note
53: Volumes 1 & 2 (good practice).
Rationale
The specification that is required for a modern haemodialysis (HD) unit has been
detailed by NHS Estates and should be followed in all new and refurbished
satellite and main renal unit HD facilities (1,2).
References
1 NHS Estates, Facilities for Renal Services, Health Building Note 53: Volume 1,
Satellite dialysis unit & Volume 2, Main renal unit

The haemodialysis facility should have sufficient specialist support staff to
fulfill the criteria listed by the Renal Workforce Planning Group 2002 (good
practice).
Rationale
The number of medical, specialist nursing, technical and allied health
professionals that are required to provide high quality HD therapy has been
standardized by the Renal Workforce Planning Group (1). There should be great
emphasis on teamwork, quality assurance and audit, health and safety and
continuing professional development for all members of the multidisciplinary
team (2).
References
1 Section 5 Workforce Planning Projections. National Renal Workforce Planning
Group Recommendations 2002

Except in remote geographical areas the travel time to a haemodialysis
facility should be less than 30 minutes or a haemodialysis facility should be
located with 25 miles of the patient’s home.
In inner city areas travel times over short distances may exceed 30 minutes at peak
traffic flow periods during the day (good practice).
Audit measure 1
The distance and travel time between the patient’s home and the nearest
satellite or main haemodialysis unit.
Rationale
Equity of access to HD is self evident in a patient-centred service. Lack of local
HD provision and the inadequacy of patient transport services are the commonest
concerns cited by HD patients and Kidney Patient Associations. The acceptance
rate for dialysis declines with increasing distance and travel time from the nearest
dialysis unit and patients are less likely to be offered dialysis if the travel time
from home to the dialysis unit is more than 37 minutes (1,2). The prevalence rate
of HD patients remains significantly lower in the areas of Wales with travel times
greater than a 30 minute drive to the nearest current dialysis unit (3). To reverse
the inverse relationship between acceptance rates for HD and travel time to the
nearest HD facility patients should not need to spend more than 30 minutes
traveling to and from dialysis unless they live in a remote geographical area. NHS
Quality Improvement Scotland has adopted 30 minutes as the maximum routine
travel time to and from HD facilities in Scotland except in remote areas (4) but
this guideline may be viewed as impractical in some urban areas because of
transport delays due to traffic congestion.
Small satellite units should be established also in rural areas or islands to provide
more local access to HD and permit travel distances or times that make thrice
weekly HD acceptable to patients. Many of the prevalent HD population are
elderly, have diabetes and/or overt cardiovascular disease and have suboptimal
vascular access in the form of central venous catheters. Some of these patients
therefore may not be medically suitable for treatment at a local satellite HD unit
and may need to travel further to a main renal unit for dialysis. A comparison of
the costs, quality of dialysis, quality of life and frequency of adverse events of HD
in satellite and main renal units in England and Wales showed no major
differences except the adequacy of HD, as assessed by measurement of the urea
reduction ratio, was better in the patients treated in satellite units (5, 6). The
provision of dialysis treatment at the 12 renal satellite units in the study
potentially saved the HD patients an additional 19 minutes travel time for each
dialysis session (5). This study has confirmed that HD in a satellite unit is an
effective alternative to treatment in a main renal unit and provides support for a
national network of HD facilities with adequate capacity to enable all medically
suitable patients to receive chronic HD without having routine travel times in
excess of 30 minutes. The location of satellite units should provide maximum
geographic access to patients within the local catchment population and a centre
of population based approach has been used in the planning of small satellite HD
units in some regions of the UK (7).
Better local access to HD can only be achieved if there are improvements in

patient transport as well as the development of an extensive network of HD
facilities. The Cross Party Group on Kidney Disease Report, 2004 reinforces this
point since it identified that 49% of HD patients in Scotland had travel times in
excess of 30 minutes even though only 10% patients lived more than a 30 minute
drive from the nearest HD facility (8). The development of patient transport
services that avoid the need to collect and drop off other patients at the dialysis
centre or at other healthcare facilities would help keep travel times to a minimum.
References
1 Roderick P, Clements S, Stone N et al. What determines geographical variation in
rates of acceptance onto renal replacement therapy in England? J Health Service Res
Policy 1999; 4:139-146
2 Boyle PJ, Kudlac H, Williams, AJ. Geographical variation in the referral of patients
with chronic end-stage renal failure for renal replacement therapy. QJM 1996; 89:
151-157
3 White P, James V, Ansell D et al. Equity of access to dialysis facilities in Wales.
QJM 2006; 99:445-452
4 Clinical Standards for Adult Renal Services, NHS Quality Improvement Scotland,
March 2003. (www.clinicalstandards.org)
5 Roderick P, Armitage A, Nicholson T et al. A clinical and cost evaluation of
haemodialysis in renal satellite units in England and Wales. Am J Kidney Dis 2004;
44: 121-131
6 Roderick P, Nicholson T, Armitage A et al. An evaluation of the costs, effectiveness
and quality of renal replacement therapy provision in renal satellite units in England
and Wales. Health Technol Assess 2005; 9:1-178
7 MacGregor MS, Campbell J, Bain M et al. Using geographical information systems
to plan dialysis facility provision. Nephrol Dial Transplant 2005; 20:1509-1511
8 Cross Party Group on Kidney Disease, April 2004
9 (www. show.scot.nhs.uk/srr/Publications/Cross party report renal disease in
Scotland.pdf)

Haemodialysis patients who require transport should be collected from home
within 30 minutes of the allotted time and be collected to return home within
30 minutes of finishing dialysis (good practice).
Audit measure 2
The waiting time after arrival before starting dialysis and the waiting time
for patient transport after the end of haemodialysis
Rationale
Patient travel to and from hospital is the main source of complaint of hospital HD
patients (1). Reduction in the waiting times before traveling to or from the HD
unit would significantly shorten the “dialysis day” for many patients (1).
Provision of designated parking adjacent to the dialysis area would encourage

patients to organize their own transport to and from dialysis and so reduce the
need for hospital provision of patient transport. Specialised, fully funded transport
for dialysis patients is the gold standard and should be developed to facilitate
timely transport by car or ambulance to meet these guidelines. The provision of
dedicated or individualized HD patient transport services, which can avoid the
need to collect and drop off other patients, and the use of staggered starting times
for HD would help to reduce patient waiting times before starting and after
completing dialysis. Audit of this patient-centred index of quality of HD
provision has been reported in the Scottish HD population by Quality
Improvement Scotland (QIS) (2).
References
1 Clinical Standards for Adult Renal Services, NHS Quality Improvement Scotland,
March 2003. (www.clinicalstandards.org)

All patients who may be suitable for home dialysis should receive full
information and education about home haemodialysis.
Home haemodialysis training is not available in all renal units and some patients
may need to travel to a sub-regional or regional centre to pursue their choice to
train for home haemodialysis (good practice).
Audit measure 3
The proportion of dialysis patients in the main renal unit and its satellite
units who are on home haemodialysis
Rationale
HD may be performed in a variety of settings, including hospital-based units, free-
standing units, and in the home. Patient survival and quality of life adjusted for
co-morbid risk factors has been reported to be higher on home than hospital HD
(1, 2). Home HD is more cost-effective than hospital HD if patients remain on
dialysis for more than 14 months to offset training and setup costs (3). The choice
between home and hospital HD for patients assessed as able to perform dialysis at
home should be determined mainly by patient preference rather than economic
grounds. Nevertheless the number of patients on home HD in the UK has
continued to decline. Not all UK units provide home HD and, based on a review
of the clinical-effectiveness and cost-effectiveness of home, satellite and hospital
HD, the National Institute of Clinical Excellence (NICE) has recommended that
the option to train to perform home HD should be available to all patients (4,5).
NICE recommended that more than 10% of dialysis patients should be treated by
home HD and, whilst this recommendation is achieved in Australasia (6), very
few centres in the UK have more than 5% of dialysis patients on home HD (7).
Higher prevalence rates of home HD may be achieved by having a designated
home HD training centre serving several renal units within a region akin to current
service provision for renal transplantation.
References
1 Woods JD, Stannard D, Blagg CR et al. Comparison of mortality with home
haemodialysis and centre haemodialysis: A national study. Kidney Int 1996; 49:
1464- 1470
2 Saner E, Nitsch D, Descourdes C et al. Outcome of home haemodialysis patients: A
case-control study. Nephrol Dial Transplant 2005, 20: 604- 610
3 Mackenzie P, Mactier RA. Home haemodialysis in the 1990's.
4 Nephrol Dial Transplant 1998; 13: 1944-1948
5 Mowatt G, Vale L, Perez J et al. Systematic review of the effectiveness and cost-
effectiveness and economic evaluation of home versus hospital or satellite
haemodialysis for people with end-stage renal failure. Health Technol Assess 2003;
7: 1-174
6 National Institute of Clinical Excellence.
7 Full guidance on home compared with hospital haemodialysis for patients with end-
stage renal failure October 2002. (www.nice.org.uk)
8 MacGregor MS, Agar JW, Blagg CR. Home haemodialysis - international trends and
variation. Nephrol Dial Transplant 2006; 21: 1934-45
9 The Renal Association UK Renal Registry, The Seventh Annual Report, December
2004. (www.renalreg.com Renal Association Standards & Audit Subcommittee)

Haemodialysis capacity in satellite and main renal units within a
geographical area should increase in step with predicted need. To allow for
patient choice regarding out of hours haemodialysis schedules, provision of
holiday haemodialysis and expansion in patient numbers calculation of the
required number of haemodialysis stations should be based on using each
station for 2 patients per day three times per week.
The national average number of hospital haemodialysis patients per million
catchment population reported for the previous year by the UK Renal Registry
should be regarded as the minimum capacity for haemodialysis in each
geographically based renal service. Alternatively up-to-date regional data may be
used. For example the national average provision for 312 hospital haemodialysis
patients (78 stations) per million catchment population in Scotland at the end of
2005 may be regarded as a minimum haemodialysis capacity in all regions in
2006. The level of hospital haemodialysis provision will need to be higher in areas
with a high ethnic and/or elderly population and increase nationwide over the next
10 years (good practice).
Audit measure 4
The number of haemodialysis patients and number of haemodialysis stations
in the main renal unit and its satellite units expressed per million catchment
population.
Rationale
HD treatment has evolved rapidly since its introduction and HD is the main mode
of dialysis in most developed countries. HD was the established mode of dialysis
at 90 days in 67.5% of the UK patient cohort in 2003 compared with 59% in 1998
(1). About 40% of patients starting renal replacement therapy (RRT) are referred
as late uraemic emergencies with no time for the planning of, or counseling on,
the options for dialysis, and such patients are more likely to remain on HD (2,3).
HD is also the default therapy for all end stage renal disease (ESRD). Despite the
success of transplantation and peritoneal dialysis (PD), HD continues to have the
highest rate of growth of all treatment modalities. Many patients are maintained
by HD after failure of renal transplants or because they have had to abandon PD.
After the first 3 years of dialysis 3% of the 1998-2000 cohort of HD patients in the
UK had converted to peritoneal dialysis, mostly within the first year, whereas
almost 11% of the PD patients had switched to HD each year (1).
The provision of HD capacity within the UK has tended to lag behind patient
demand and this has restricted both patient choice and access to hospital HD (4).
UK Registry data from the end of 2004 showed that there were 261 patients per
million population on hospital or satellite HD (5). 40.9% of the estimated 638
prevalent end-stage renal failure patients per million population were receiving
hospital HD and only 1.2% were on home HD at the end of 2004 (5). Scottish
Renal Registry data from the end of 2004 showed that 76% of dialysis patients
were receiving hospital HD, 299 patients per million were receiving hospital HD
and 725 prevalent patients per million were on chronic RRT (6). At the end of
2005 the Scottish Renal Registry data showed that 77% of dialysis patients were
receiving hospital HD, 312 patients per million were receiving hospital HD and
758 prevalent patients per million were on chronic RRT (7). Hospital HD
provision in Scotland increased by an average of 18 patients per million
population each year between 2000 and 2005. Regional variation in the level of
provision of HD within the UK continues and this needs to be addressed to permit
equity of access to HD throughout the country (8).
Additional capacity is needed to allow for patient choice of HD schedule, holiday

HD and anticipated expansion in patient numbers. For these reasons the calculated
number of dialysis stations that are required in each geographical area should be
based on using each machine only for two patients per day three days per week.
The degree of flexibility in HD capacity and scheduling then depends on the
proportion of HD patients who are on a third shift each day. The national average
number of hospital HD patients per million catchment population reported for the
previous year by the UK Renal Registry may be regarded as the minimum
capacity for HD in each geographically based renal service. This approach should
drive the provision of HD upwards in the areas with below average HD capacity.
For example the provision for an average of 312 hospital HD patients (or 78
stations) per million catchment population in Scotland at the end of 2005 could be
regarded as a minimum HD capacity in all regions in 2006. The required capacity
for HD will be greater in areas with a high ethnic or elderly population due to their
higher prevalence of ESRD and these areas will need proportionately greater HD
capacity than the national average. HD capacity will need to expand greatly over
the next 10 years as the number of prevalent ESRD patients rises progressively
and the proportion of the patients who are elderly and/or have co-morbidity also
increases (9). Regional and national audit of HD capacity will highlight if there is
inequity of access to HD and provide support for the development of HD facilities
in such geographical areas. Meeting the need for HD will be a major challenge
and regular audit should be used to raise HD capacity across the UK in step with
the projected increase in demand over the next decade.
References
2 Metcalfe W, Khan IH, Prescott GJ et al. Can we improve early mortality in patients
receiving renal replacement therapy? Kidney Int 2000; 57: 2539–45
3 Little J, Irwin A, Marshall T et al. Predicting a patient’s choice of dialysis modality:
experience in a United Kingdom renal department. Am J Kidney Dis 2001; 37: 981–6
4 Treatment of adults & children with renal failure - Standards and audit measures. 3rd
Edition, London: Royal College of Physicians 2002.
(www.renal.org/Standards/standards.html)
5 The Renal Association UK Renal Registry, The Eighth Annual Report, December
6 Report of the Scottish Renal Registry 2004 (www.show.scot.nhs.uk/srr)
7 Report of the Scottish Renal Registry 2005 (www.show.scot.nhs.uk/srr)
8 Blank L, Peters J, Lumsdon A et al. Regional differences in the provision of adult
renal dialysis services in the UK. QJM 2005; 98:183-190
9 Feest TG, Rajamahesh J, Byrne C et al. Trends in adult renal replacement therapy in
the UK: 1982-2002. QJM 2005: 98: 21-28
2 Haemodialysis equipment and disposables


All equipment used in the delivery and monitoring of therapy should comply
with the relevant standards for medical electrical equipment. General safety
standards are covered by BS EN 60601-1: 2006 and specific dialysis machine
requirements are covered by BS-EN 60601-2-16: 1998 (Medical electrical
equipment: Particular requirements for the safety of haemodialysis (HD),
haemodiafiltration and haemofiltration equipment (good practice).
The equipment used in renal units represents a substantial asset that must be
carefully maintained. The selection of equipment should be in accordance with a
policy that conforms to the recommendations of the MHRA (MHRA DB2006
(05): - Managing Medical Devices – Guidance for Healthcare & Social Services
Organisations – Nov. 2006 (personal communication, Ged Dean, Nottingham
University Hospitals Trust) and National Audit Office (The management of
medical equipment in NHS acute trusts in England, National Audit Office, 1999).
The above BS-EN 60601-2-16 standard for electrical equipment for renal
replacement therapy was defined in 1998, superceded BS5724-2-16:1998 and IEC
60601-2-16:1998 and remains applicable in 2006 (personal communication, Andy
Mosson, Association of Renal Technologists).

Disposables such as dialysers and associated devices are classified as medical
devices and should display the CE mark (good practice).
All disposable equipment such as haemodialysers, blood tubing sets and related
devices should display the CE mark. The presence of such a mark signifies
compliance with the requirements of the statutory Medical Device Directive and
also national and international standards where they exist for new products: BS-
EN 1283: 1996 (haemodialysers, haemodiafilters, haemofilters,
haemoconcentrators and their extra corporeal circuits), ISO 8638:2004
(Extracorporeal blood circuit for haemodialysers, haemodiafilters and
haemofilters) or ISO 13960: 2003 (Plasma filters).

Machines should be replaced after between seven and ten years’ service or
after completing between 25,000 and 40,000 hours of use for haemodialysis,
depending upon an assessment of machine condition (Good practice).
The routine maintenance of the equipment used for renal replacement therapy is
essential and the service history of each machine should be documented fully
throughout its use-life by the renal unit technicians. Renal units should endeavour
to adopt a programme of phased replacement of older HD machines. Although it
is possible to keep a dialysis machine operating safely for many years, practical
considerations of obsolescence and maintenance costs require a more structured

approach. When a particular model of a machine becomes obsolete, companies
generally only undertake to supply replacement parts for seven years. Intensive
use of HD machines for three 4 hour shifts per day, 6 days per week would
complete 26208 hours of use after 7 years. We accept that there is no firm
evidence that replacement, as suggested above, is the most cost-effective strategy.
3 Concentrates and water for haemodialysis

Ready made concentrates are classified as medical devices and should display
the CE mark. Concentrates that are manufactured ‘in house’ should meet the
requirements of BS EN 13867: 2002 (Concentrates for haemodialysis and
related therapies) (good practice).
Rationale
The presence of the CE mark signifies compliance with the requirements of the
statutory Medical Device Directive and also national and international standards
where they exist. Dialysis units that manufacture concentrates in-house should
ensure that the fluid fulfils the requirements stated in BS EN 13867: 2002 (1).
References
1 BS EN 13867: 2002 Concentrates for haemodialysis and related therapies.
(http://www.bsonline.bsi-global.com)
Note: A revision of this standard is currently at the Committee Draft stage and should
be available by 2008.

Water used in preparation of dialysis fluid should, as a minimum, meet the
requirements stated in Table 1 for chemical and microbiological
contaminants.
Rationale
The limits for chemical contaminants are derived from AAMI RD-52 2004 (1),
ISO 13959:2002 (2) and the European Pharmacopoeia (3) after consultation
within the UK, whilst the limits for bacterial counts (100 cfu/ml) and endotoxin
(0.25 IU/ml) are based on the European Pharmacopoeia (3) and the European
Renal Association Best Practice Guidelines (4). New equipment should be
capable of producing ‘ultrapure’ dialysis fluid (bacterial counts <0.1 cfu/ml and
endotoxin <0.03 IU/ml) in order to meet the best practice guidelines. Ideally this
should be achieved using ultrapure water, however water that meets the minimum
standard in Table 1 can be used together with point of use filtration of the dialysis
fluid. If routine monitoring demonstrates continuous contamination in excess of
the desired levels, a programme to improve this should start immediately (good
practice).
HD exposes the blood of the patient to in excess of 300 litres of water per week
through a non-selective membrane, in contrast to an average 12 litres per week
through a highly selective membrane (intestinal tract) in healthy individuals. It is
essential for the water used to produce dialysis fluid to have appropriate chemical
and microbiological purity. Achieving this standard of purity usually requires
softening, carbon filtration, reverse osmosis and an effective disinfection
programme for the pipework between the treatment plant and the dialysis
machines.
The dialysis membrane was regarded as an effective barrier against the passage of
bacteria and endotoxin (potent pyrogenic materials arising from the outer layers of
bacterial cells) from dialysis fluid to blood. This produced a complacent attitude
towards the purity of dialysis fluid. About 10 years ago, several in vitro studies
showed that intact membranes used in dialysers are permeable to bacterial
contaminants (5-7). The pore size of the membrane appears to be less important
than the thickness and the capacity of the membrane to adsorb bacterial products.
Consequently low flux (standard) dialysis does not necessarily translate into
higher microbiological safety than high flux dialysis or HDF. Patients receiving
standard dialysis treatment with low flux cellulose-based membranes (thickness
6–8 microns), may therefore be at greater risk of pyrogenic reactions (see below)
than those treated using thicker synthetic membranes which have the capacity to
adsorb bacterial endotoxin.
In patients treated with high flux membranes, a risk of pyrogen transfer due to
backfiltration (a movement of dialysis fluid into the blood pathway of the device
due to an inverted pressure gradient rather than the diffusion gradient discussed
above) may exist. Lonneman et al, however, concluded that diffusion rather than
convection is the predominant mechanism of transmembrane transport of
pyrogens, and backfiltration across pyrogen adsorbing membranes does not
necessarily increase their passage (5). It should be emphasised that the adsorption
capacity of the synthetic membranes is not infinite and that a breakthrough of
pyrogenic substances can occur in the event of excessive water contamination.
A raised C-reactive protein (a sensitive marker of activation of the acute phase

response) is associated with a significantly increased risk of death (8, 9), which
has led to speculation that impure dialysis fluid may contribute to an increased
risk of death in dialysis patients. The use of ultrapure water in a randomized study
of 30 incident HD patients was associated with a reduction in CRP levels and a
decrease in the rate of loss of residual renal function (10). Impure dialysis fluid
has also been implicated in the pathogenesis of dialysis-related amyloidosis (11-
13). While this suggestion has not been tested in clinical practice, it would seem
prudent to ensure that water is as pure as reasonably possible.
Ultrapure water (< 0.1 cfu/ml and bacterial endotoxins < 0.03 IU/ml) is readily
achievable using modern water treatment techniques and should be regarded as
the standard for all newly installed water treatment plants (14). The European
Best Practice Guidelines recommend the use of ultrapure water for conventional
as well as high flux HD (4).
Reinfusion fluid, used in haemofiltration and haemodiafiltration, must be sterile

(<1 cfu/1000 litres) and, particularly where large exchange volumes are required,
have an endotoxin level of <0.03 IU/ml (15). Even with ultrapure water, this
standard of purity can only be achieved with ‘on-line’ fluid production with
multiple filtration of the dialysis fluid. Machines designed to produce reinfusion
fluid usually require a water supply that meets the microbiological requirements
of table 1.
Knowledge of the potentially harmful effects of bacterial products, trace elements

and chemicals is still growing, and techniques of water treatment are continuously
being developed. The maximum acceptable levels of chemical contaminants in
water for dialysis have been established by AAMI (1), ISO 13959 (2) and the
European Pharmacopoeia (3). These standards differ in the number and limits of
the contaminants specified. Table 1 lists all the contaminants for which a limit is
defined for water for dialysis in one or more of the standards. With the exception
of nitrate, where the standards differ in their recommendations, the most stringent
limit has been adopted. These limits should not be difficult to meet with a
correctly specified and maintained water plant. For nitrate, the European
Pharmacopoeia specifies a maximum of 2 mg/l nitrate (NO3) compared to the
AAMI standard and ISO 13959 which recommend a limit of 2 mg/l of nitrate as
nitrogen (N) which equates to approximately 9 mg/l of NO3. The more stringent
EP limit is often difficult to meet in home HD installations and since the
development of the AAMI standard, no new evidence for a lower limit for nitrate
has emerged. As the benefits of home HD are well established, the less stringent
recommendation has been adopted for nitrate.
Table 1 defines a subset of contaminants that should always be included in routine

testing because they occur in relatively high levels and are not restricted in
drinking water (chlorine, calcium, magnesium and potassium), or where the
drinking water limit (16) is more than five times the recommended limit for water
for dialysis. Sodium is included in the ‘mandatory’ group because, although the
drinking water limit is 200 mg/l, additional sodium is introduced by softening.
Table 1 also defines a group of contaminants for which the drinking water limit is
2 to 5 times the recommended limit for dialysis. In water treated by reverse
osmosis, these contaminants will only exceed the limits in table 1 if they occur at
relatively high levels in the water supplied to the unit. These contaminants can be
omitted from routine tests if data is available to show that the levels in the water
supplied to the unit rarely exceed the limit in the table. These data should be
obtained from the municipal water supplier, or from tests on the raw water if it is
obtained from a private source.
Table 1: Maximum recommended concentrations for chemical and microbial

contaminants in water for dialysis
Contaminant Criteria for Maximum Standards on Initial Test

inclusion in recommended which limit is Frequency
routine tests concentration based (if not omitted)
(mg/l = ppm)
Aluminium Mandatory 0.01 EP, AAMI, 3 monthly
ISO
Calcium Mandatory 2 (0.05 EP, AAMI, 3 monthly
mmol/l) ISO
Total chlorine Mandatory 0.1 EP Not less than
weekly
Copper Mandatory 0.1 AAMI, ISO 3 monthly
Fluoride Mandatory 0.2 EP, AAMI, 3 monthly

ISO
Magnesium Mandatory 2 (0.08 EP 3 monthly
mmol/l)
Nitrate (as N) Mandatory 2 (equates to 9 AAMI, ISO 3 monthly
mg/l NO3)
Potassium Mandatory 2 (0.05 EP 3 monthly
mmol/l)
Sodium Mandatory 50 (2.2 EP 3 monthly
mmol/l)
Bacteria Mandatory 100 cfu/ml EP, ISO Not less than
(TVC) monthly
Endotoxin Mandatory 0.25 IU/ml EP Not less than
monthly
Ammonium Omit if evidence 0.2 EP 3 monthly
permits
Arsenic Omit if evidence 0.005 AAMI, ISO 3 monthly
permits
Cadmium Omit if evidence 0.001 AAMI, ISO 3 monthly
permits
Chloride Omit if evidence 50 EP 3 monthly
permits
Chromium Omit if evidence 0.014 AAMI, ISO 3 monthly
permits
Lead Omit if evidence 0.005 AAMI, ISO 3 monthly
permits
Mercury Omit if evidence 0.0002 AAMI, ISO 3 monthly
permits
Sulphate Omit if evidence 50 EP 3 monthly
permits
Barium Include on 0.1 AAMI, ISO As indicated
indication only
Beryllium Include on 0.0004 AAMI As indicated
indication only
Silver Include on 0.005 AAMI, ISO As indicated
indication only
Thallium Include on 0.002 AAMI As indicated
indication only
Tin Include on 0.1 ISO As indicated
indication only
Zinc Include on 0.1 EP, AAMI, As indicated
indication only ISO
Notes: Antimony (AAMI limit 0.006 mg/l) and selenium (AAMI and ISO limit 0.09 mg/l)
have been excluded from this table as the limit for drinking water in the UK is lower than the
limit for water for dialysis. The limit for tin may be omitted from the next revision of ISO
13959.
The final group of contaminants (barium, beryllium, silver, thallium, tin and zinc)
are those for which a limit has been defined for water for dialysis and there is no
limit specified for drinking water in the UK. These trace elements are not
considered to occur in levels that give cause for concern and, if low levels are
present, they are removed effectively by reverse osmosis. Testing is only required
if there is evidence of high levels in the local water supply (zinc, for example, can
be introduced in the pipework).
Although the standards specify laboratory test methods to be used for chemical
contaminant, any test method validated by the United Kingdom Accreditation
Service is acceptable provided the detection limit is not less than 50% of the limit
given in Table 1.
The tests used for monitoring microbial contamination of water for dialysis should
be appropriate to the type of organisms found in water. A low nutrient agar, such
as Tryptone Glucose Extract Agar or Reasoner’s 2A, should be used [17-19] and
samples should be incubated for at least 7 days at 20-22ºC [20]. These conditions
have been shown to give good recovery for most environmental bacteria found in
purified water. Some species are better adapted for growth at a higher temperature
and/or on richer media, but the long incubation time will allow most of these to
grow. Details of methods for sampling and culturing of water for dialysis are
available in the Appendix of European Best Practice Guidelines for
Haemodialysis Part 1 (3) and in the EDTNA/ERCA Guidelines on Control and
Monitoring of Microbiological Contamination in Water for Dialysis (21), which
also gives specific test conditions for fungi.
Procedures for providing quality assurance of the reinfusion fluid produced 'on-
line' for haemofiltration or haemodiafiltration will be reviewed before the next
edition of these guidelines as dialysis units gain experience managing this
technology.
References
2 Association for the Advancement of Medical Instrumentation. Standard RD52:2004 -
- Dialysate for haemodialysis. AAMI, Arlington, VA.
3 International standard reference number ISO 13959: 2002 (E). Water for
haemodialysis and related therapies. (www.iso.org).
Note: A revision of this standard is currently at the Committee Draft stage and should
be available by 2008.
4 Monograph 1167:1997 (corrected 2000, republished 2001) Haemodialysis solutions,
concentrated, water for diluting. European Pharmacopoeia Supplement 2001.
5 European Best Practice Guidelines for haemodialysis Part 1. Section IV. Dialysis
fluid purity. Nephrol Dial Transplant 2002; 17: Supplement 7 S45-S62
6 Lonnemann G, Behme TC, Lenzner B et al. Permeability of dialyzer membranes to
TNF alpha-inducing substances derived from water bacteria. Kidney Int 1992; 42:61–
68
7 Evans RC, Holmes CJ. In vitro study of the transfer of cytokine inducing substances
across selected high flux hemodialysis membranes. Blood Purif 1991; 9: 92–101
8 Laude-Sharp M, Caroff M, Simard L et al. Induction of IL-1 during hemodialysis:
transmembrane passage of intact endotoxins (LPS). Kidney Int 1990; 38: 1089–94
9 Owen WF, Lowrie EG. C-reactive protein as an outcome predictor for maintenance
hemodialysis patients. Kidney Int 1998; 54:627–36
10 Zimmermann J, Herrlinger S, Pruy A et al. Inflammation enhances cardiovascular

risk and mortality in hemodialysis patients. Kidney Int 1999; 55:648–58
11 Schiffl H, Lang SM, Fischer R. Ultrapure dialysis fluid slows loss of residual renal
function in new dialysis patients. Nephrol Dial Transplant 2002; 17: 814-1818
12 Lonnemann G. The quality of dialysate: an integrated approach. Kidney Int Suppl
2000; 58(Suppl 76):S112–19
13 Lonnemann G. Chronic inflammation in haemodialysis: the role of contaminated
dialysate. Blood Purif 2000; 18: 214–23
14 Lonnemann G. Should ultra-pure dialysate be mandatory? Nephrol Dial Transplant
2000; 15(Suppl 1): 55–9
15 Ward RA. Ultrapure dialysate: a desirable and achievable goal for routine
haemodialysis. Semin Dial 2000; 13: 378–80
16 Nystrand R. Endotoxin and 'on line' production of substitution fluid in
haemodiafiltration and haemofiltration. EDTNA-ERCA J 2002; 28: 127-9
17 The Water Supply (Water Quality) (England and Wales) Regulations 2000. Statutory
Instrument No. 3184. Prescribed concentrations and values.
http://www.dwi.gov.uk/regs/si3184/3184.htm#sch1
18 Van der Linde K, Lim BT, Rondeel JM, Antonissen LP, de Jong GM. Improved
bacteriological surveillance of haemodialysis fluids: a comparison between Tryptic
soy agar and Reasoner's 2A media. Nephrol Dial Transplant 1999;14: 2433-7
19 Ledebo I, Nystrand R. Defining the microbiological quality of dialysis fluid. Artif
Organs 1999;23: 37-43
20 Harding GB, Pass T, Million C, Wright R, DeJarnette J, Klein E. Bacterial
contamination of haemodialysis center water and dialysate: are current assays
adequate? Artif Organs 1989;13:155-9
21 Pass T, Wright R, Sharp B, Harding GB. Culture of dialysis fluids on nutrient-rich
media for short periods at elevated temperatures underestimate microbial
contamination. Blood Purif 1996; 14:136-45
22 Guidelines for the control and monitoring of microbiological contamination in water
for dialysis. EDTNA-ERCA J 2002; 28: 107-115
http://www.associationhq.com/edtna/pdf/WTguidelinesmicrob.pdf

A routine testing procedure for water for dialysis should form part of the
renal unit policy (good practice).
Rationale
The manufacturer of the water treatment plant and distribution system should
demonstrate that the requirements for microbial contamination are met throughout
the complete system at the time of installation.
The frequency for testing of water for dialysis for microbial contamination should
not fall below monthly in main units and should be sufficiently frequent to detect
trends. This is particularly important after installation and when alterations to the
system are carried out. For home installations, it is usually impractical to maintain
a monthly testing programme. In this situation, the dialysis machine should be
fitted with point of use filtration.
The absence of any type of bacteriostat in the water following treatment makes it
susceptible to bacterial contamination down stream of the water treatment plant.
Such contamination may be further enhanced by stagnant areas within the

distribution network as well as irregular cleaning all of which contribute to the
development of a biofilm which may also be present in the dialysate pathway of
the proportionating system, particularly when non sterile liquid bicarbonate
concentrate is used. Such biofilm is difficult to remove and results in the release
of bacteria, bacterial fragments (endotoxins, muramylpeptides, and
polysaccharides). The dialysis membrane prevents intact bacteria from crossing,
however fragments have molecular weights that allow them to pass across the
membrane into the bloodstream. Considerable differences exist in the adsorption
capacity of such membranes, which may permit the passage of short bacterial
DNA fragments (1,2). Current proportionating systems incorporate filters for the
removal of such fragments on the basis of size exclusion and hydrophobic
interaction. Monitoring as well as disinfection should be scheduled to prevent
formation rather than to eliminate biofilm that may be present and a routine
testing procedure for dialysate and feed water should form part of the renal unit
policy.
Testing for chemical contaminants will normally include continuous conductivity

monitoring of the water leaving the reverse osmosis system, and regular in-house
checks of hardness and total chlorine (3). There is increasing use of chlorine
dioxide to prevent growth of Legionella bacteria in hospital water systems. There
is currently no guidance on the control and monitoring of chlorine dioxide in
water for dialysis. Confirmation that the standard DPD test used to monitor
chlorine and chloramines gives an accurate measure of the levels of chlorine
dioxide and its breakdown products (chlorite and chlorates) is needed as is data on
the carbon filter empty bed contact time required for effective removal. New data
will be reviewed before the next edition of these guidelines.
The laboratory tests required to demonstrate compliance with the

recommendations in Table 1 should be carried out during commissioning and
within 3 months of setting up a new water system or making alterations to the
water treatment plant. The frequency of laboratory tests may be modified once
local trends have been established, but should not fall below annually. An initial
full test on the supply water may be advisable and regular monitoring of water
quality data from the supplier is essential when tests are omitted based on low
levels of contamination in the water supply.
Records should be kept of all chemical and microbiological test results and
remedial actions.
References
1 Differences in the permeability of high-flux dialyzer membranes for bacterial
pyrogens. Schindler R, Christ-Kohlrausch F, Frei U, Shaldon S. Clin Nephrol 2003;
59:447-54
2 Short bacterial DNA fragments: detection in dialysate and induction of cytokines.
Schindler R, Beck W, Deppisch R, Aussieker M, Wilde A, Gohl H, Frei U. J Am Soc
Nephrol. 2004; 15:3207-14
3 Guidelines for the control of chlorine and chloramine in water for haemodialysis
using activated carbon filtration. Morgan I. EDTNA ERCA J. 2004; 30:106-12

The dialysate should contain bicarbonate as the buffer. (Evidence)
Rationale
One of the critical functions of dialysis is the correction of the metabolic acidosis
caused by the failure of the diseased kidneys to excrete non-volatile acids and to
regenerate bicarbonate. Bicarbonate is the natural buffer normally regenerated by
the kidneys and was the initial choice as dialysate buffer. If, however, sodium
bicarbonate is added to a calcium- or magnesium-containing dialysate, their
respective carbonate salts will precipitate unless the dialysate is maintained at a
low pH level. Since it does not precipitate calcium or magnesium, acetate was
used as an alternative buffer (1) because of its rapid conversion to bicarbonate in
the liver. In the late 1970s and early 1980s, a number of studies suggested that
some of the morbidity associated with HD could be attributed to the acetate
component of the dialysate (2,3). This appears to have been unmasked by the
introduction of high-efficiency and short-duration dialysis, using membranes with
large surface areas. Acetate intolerance led to the reappraisal of bicarbonate as a
dialysis buffer in the early 1980s and, following the solving of the issue of
precipitation, to its reintroduction. A systematic review of 18 randomised trials
indicated a reduction in the number of treatments complicated by headaches,
nausea/vomiting and symptomatic hypotension when bicarbonate was used (4).
Economic evaluations showed the cost of self-mix bicarbonate buffer to be similar
to that of acetate. It should be noted, however, that even ‘bicarbonate’ dialysate
contains moderate amounts of acetate (5)
It is not possible to set evidence-based standards for other components of the

dialysate. However there is recent evidence that non-diabetic HD patients using
glucose-free dialysate have a surprisingly high rate of asymptomatic
hypoglycaemia without an associated counter-regulatory response (6,7) The long-
term effects of repeated dialysis-induced hypoglycaemia are uncertain.
Hypoglycaemia is not observed if the dialysate contains glucose, but glucose-
containing dialysate is slightly more expensive. Individualisation of dialysate
potassium may be required in patients with hypokalaemia and adjustment of
dialysate sodium concentrations during HD (sodium profiling) may be beneficial
in some patients with haemodynamic instability.
References
1 Mion CM, Hegstrom RM, Boen ST et al. Substitution of sodium acetate for sodium
bicarbonate in the bath fluid for hemodialysis. Trans Am Soc Artif Intern Organs
1964; 10:110–15
2 Novello A, Kelsch RC, Easterling RE. Acetate intolerance during haemodialysis. Clin
Nephrol 1976; 5:29–32
3 Aizawa Y, Ohmori T, Imai K et al. Depressant action of acetate upon the human
cardiovascular system. Clin Nephrol 1977; 8:477–80
4 MacLeod A, Grant A, Donaldson C et al. Effectiveness and efficiency of methods of
dialysis therapy for end-stage renal disease: systematic reviews. Health Technol
Assess 1998; 2:1–166
5 Veech RL. The untoward effects of the anions of dialysis fluids. Kidney Int 1988;
34:587–97
6 Jackson MA, Holland MR, Nicholas J et al. Occult hypoglycemia caused by
hemodialysis. Clin Nephrol 1999; 51:242–7
7 Catalano C, Bordin V, Fabbian F et al. Glucose-free standard haemodialysis and
occult hypoglycemia. Clin Nephrol 2000; 53:235–6
4 Haemodialysis membranes (Guidelines

DIAL-HD 4.1 - 4.6)

The balance of evidence supports the use of low flux synthetic and modified
cellulose membranes instead of unmodified cellulose membranes (good
practice).
The benefits of low flux synthetic and modified cellulose membranes over
unmodified cellulose membranes are limited to advantages arising from different
aspects of improved biocompatibility rather than better patient outcomes.
Rationale
Synthetic membranes, which can have more porous characteristics (high flux)
than standard cellulose membranes, started to be used in the mid-1980s with a
view to increasing the depurative capacity of HD. Interest was heightened by the
subsequent discovery that a number of these membranes (eg polysulphone,
polyamide, polyacrylonitrile) had markedly less ability to activate complement,
leucocytes and other cellular elements than standard cellulose and hence decrease
the inflammatory response, i.e. they were more biocompatible. Cellulose
membranes have been modified to make them both more biocompatible and of
slightly higher flux (semi-synthetic membranes e.g. haemophan or cellulose
triacetate), and synthetic membranes with lower flux properties have also been
produced (e.g. low-flux polysulphone). The more biocompatible membranes may
have other advantages as a result of reduced activation of the systemic
inflammatory response during dialysis but this is less certain (1).
A systematic Cochrane review showed no evidence of benefit when synthetic

membranes were used compared with cellulose/modified cellulose membranes in
terms of reduced mortality or reduction in dialysis-related adverse symptoms (2).
Comparison of unmodified cellulose and modified cellulose membranes was not
undertaken. Despite the relatively large number of randomised controlled trials
undertaken in this area, none of the studies that were included in the review
reported any measures of quality of life. Plasma triglyceride values were lower
with synthetic membranes in the single study that measured this outcome in this
systematic review but a recent randomized study has shown no difference in
serum lipid levels in the patient group treated with high-flux biocompatible
membranes (3). Serum albumin was slightly higher at certain time points in some
studies when synthetic membranes of both high and low flux were used and this
may be an important finding given the adverse prognostic impact of
hypoalbuminaemia in dialysis patients (4,5). The lower complement and
leucocyte activation and greater adsorptive capacity for cytokines and beta-2-
microglobulin of the more biocompatible dialysis membranes have potentially
beneficial biological effects but have not been shown so far to provide better
patient survival rates than unmodified cellulose membranes (2,6). Low-flux
synthetic and modified cellulose dialysers are now no more expensive than
unmodified cellulose dialysers and the use of these more biocompatible dialysers
instead of unmodified cellulose therefore seems justifiable on the basis of

evidence of biological benefits and equivalent costs.
References
1 Ikizler TA, Wingard RL, Harvell J et al. Association of morbidity with markers of
nutrition and inflammation in chronic haemodialysis patients: A prospective study.
Kidney Int 1999; 55: 1945–51
2 Macleod AM, Campbell M, Cody JD et al. Cellulose, modified cellulose and
synthetic membranes in the haemodialysis of patients with end-stage renal disease.
The Cochrane Database of Systematic Reviews 2005 Issue 3. Art No:
CD003234.DOI: 10.1002/14651858.CD003234.pub2
3 House AA, Wells GA, Donnelly JG et al. Randomised trial of high-flux versus low-
flux haemodialysis: effects on homocysteine and lipids. Nephrol Dial Transplant
2000; 15:1029-1034
4 Foley RN, Parfrey PS, Harnett JD et al. Hypoalbuminemia, cardiac morbidity, and
mortality in end-stage renal disease. J Am Soc Nephrol 1996; 7:728–36
5 Goldwasser P, Mittman N, Antignani A et al. Predictors of mortality in hemodialysis
patients. J Am Soc Nephrol 1993; 3:1613–22
6 Grooteman MPC, Nube MJ. Impact of the type of dialyser on clinical outcome in
chronic haemodialysis patients: does it really matter? Nephrol Dial Transplant 2004;
19:2965-2970

The balance of evidence supports the use of a dialysis regimen with enhanced
removal of middle molecules in incident patients who are predicted to remain
on haemodialysis for several years and prevalent patients who have been on
haemodialysis for more than 3.7 years. Such patients are at risk of developing
symptoms of dialysis-related amyloidosis (good practice).
Treatments with better clearance of middle molecules include haemodialysis with

high flux synthetic membranes and haemodiafiltration. The proven benefits of
high flux synthetic membranes in randomized trials are limited to advantages
arising from improved biocompatibility and enhanced removal of middle
molecules, such as beta-2-microglobulin, rather than better patient survival rates.
Chronic high flux dialysis in the HEMO study did not affect the primary outcome
of all cause mortality or any of the secondary composite outcome measures
including the rates of first cardiac hospitalization or all cause mortality, first
infectious hospitalization or all cause mortality, first 15% decrease in serum
albumin or all cause mortality, or all non-vascular access-related hospitalizations.
Rationale
Dialysis-related amyloidosis is a disabling, progressive condition caused by the
polymerisation within tendons, synovium, and other tissues of beta-2-
microglobulin, a large (molecular weight (MW) 11,600) molecule, which is
released into the circulation as a result of normal cell turnover but is not excreted
in renal failure and is not removed by cellulose membranes. Exposure to bio-
incompatible membranes may increase beta-2-microglobulin generation.
Symptoms are typically first reported 7–10 years after commencing HD although
tissue accumulation of dialysis-related amyloid can be demonstrated much earlier.
A systematic review of 27 randomised trials comparing cellulose, modified
cellulose and synthetic membranes, showed a significant reduction in end of study
beta-2-microglobulin values when high flux synthetic membranes were used and
one small study showed amyloid occurred less frequently with this treatment (1).
High flux HD membranes remove beta-2-microglobulin by a combination of
diffusive clearance and adsorption and haemodiafiltration removes substantially
more as a result of convective clearance. Both treatments are thought to reduce
the risk of developing dialysis-related amyloid.
The effect of dialyser membrane flux was examined in the HEMO study, which
was a prospective randomized trial of prevalent HD patients who had been on
dialysis for a median of 3.7 years at the time of recruitment to the study (2,3).
After a mean follow-up period of 2.8 years, during which 871 of the 1846
randomised patients died, no significant difference was observed in all cause
mortality or secondary endpoints between the high and low flux treatment groups
in spite of a ten fold increase in beta-2-microglobulin clearances in the high flux
group (beta-2-microglobulin clearances of at least 20ml/min). Secondary analyses
of the patients who had been on HD for greater than the median of 3.7 years
before enrolment showed that the patients on high flux dialysis membranes had a
32% reduction in all cause mortality (CI 14-47% ; p = 0.001) and 37% reduction
in cardiac death ( CI 37-57% ; p = 0.016) compared with the low flux patients (4).
However, when the number of prevalent years on HD was analysed as a
continuous variable, the interaction of flux and years of dialysis on patient
survival was not significant. The HEMO study was designed to have adequate
power to detect a 25% reduction in the predicted baseline all cause mortality rate
of the interventions (5). However the limited benefit observed with high flux
membranes has been attributed to several factors in the design of the HEMO study
such as the inclusion of prevalent rather than incident patients, the exclusion of
patients with major co-morbidity, the failure to utilize ultra-pure water whilst
using dialyser reuse and the high and low flux groups may have been separated
inadequately since pre-dialysis beta-2-microglobulin levels were only 19% lower
in the high flux group. Most of these confounding factors have been addressed in
the Membrane Permeability Outcome (MPO) study which is a randomized,
multicentre European study of high flux membranes in incident HD patients who
have few exclusion criteria and do not reuse dialysers (6). In addition a
multicentre, randomized controlled trial has failed to show a beneficial effect on
anaemia in stable HD patients using a high flux biocompatible membrane
compared with conventional cellulose membranes over a 12 week study period
(7). A multivariate Cox proportional hazards analysis of a prospective non-
randomised study of 1610 prevalent HD patients from 20 centres in France
showed that age, diabetes, lower serum albumin and the use of low-flux dialyser
membranes were associated with poorer survival (8). The patients on high-flux
dialysers had a 38% lower risk of death (p=0.01) than patients on low-flux
membranes. This non-randomised study (8) and post hoc analysis of the HEMO
study (4) provide some evidence that long-term HD patients may have better
survival from the use of high flux dialysers but this observation needs to be
confirmed in a large prospective randomized study, such as the MPO study which
has completed recruitment (6). One small prospective study has shown better
preservation of residual renal function when using high flux membranes
combined with ultrapure water (9). Preservation of residual renal function is
desirable as residual renal function is a predictor of survival in HD patients (10),
decreases beta-2-microglobulin levels and lessens the need for ultrafiltration.
As long as the cost of high flux membranes is significantly higher than low flux
synthetic and modified cellulose membranes and the single use of dialysers
remains routine practice the use high flux membranes should be a higher priority
in patients who are likely to remain on or have been on HD for at least 3.7 years
as this group of patients is at the greatest risk of developing dialysis-related
amyloidosis. Appropriate incident patients include patients who are unlikely to
receive a transplant either as a result of human major histocompatibility complex
(HLA) sensitisation, high risk of recurrent disease, rare tissue type or other
contra-indications (including personal choice and age) and prevalent patients with
high risk of dialysis-related amyloidosis because of long-term dialysis or absence
of residual renal function at the start of dialysis.
Since so-called ‘middle molecules’ (MW 200–20,000) diffuse only slowly into
dialysis fluid, shortened treatment times have a proportionately greater deleterious
effect on their clearance and this may have implications for the long-term health
of dialysis patients. Theoretically, reductions in sessional dialysis time can be
more safely pursued if there is a concomitant improvement in middle molecular
(MM) clearance, a goal which cannot be achieved by high blood flow rate or
dialysis fluid flow rate and large surface areas of membranes impermeable to
middle molecules. While the use of high flux membranes can increase this, a more
effective way of promoting MM clearance is to superimpose convection upon
standard diffusive blood purification technique using haemodiafiltration. In this
technique approximately 20 litres of ‘extra’ fluid, over and above the patients’
interdialytic fluid gain, is removed through the dialyser and an equal volume of
physiological ‘replacement’ fluid is returned to the blood before (pre-dilutional)
or after (post-dilutional) the dialyser. However a recent systematic review of the
existing 18, albeit mainly small, randomized trials showed no difference in patient
outcomes between HD, haemodiafiltration and haemofiltration (11). The authors
have acknowledged that there was a small arithmetic error in this systematic
review although this did not alter its main conclusion (12). Haemodynamic
variables were found to be similar in a further recent study comparing
haemodiafiltration and low-flux HD under conditions of equivalent dialysis dose,
ultrafiltration volume and core temperature (13). In a retrospective observational
study of 2165 patients from 1998-2001 in five European countries, stratified into
4 groups (low-flux HD, high-flux HD, low-efficiency haemodiafiltration and
high-efficiency haemodiafiltration), the subgroup on high-efficiency
haemodiafiltration had a 35% lower mortality risk compared with patients on low-
flux HD after adjusting for the dialysis dose and co-morbidity (p = 0.01) (14). In
view of the potential influence of selection bias and other confounding factors the
authors of this study stated that a controlled clinical trial was required to
document the benefits of haemodiafiltration before it can be recommended in
clinical practice guidelines (13). The Dutch CONvective TRAnsport STudy
(CONTRAST) is a 3 year randomized study that addresses if all cause mortality
and/or fatal and non-fatal cardiovascular events differ between haemodiafiltration
and low flux HD in almost 800 HD patients (15). At present there is no objective
evidence to support the routine use of haemofiltration or haemodiafiltration
instead of HD in the management of end-stage chronic renal failure.
References
1 Macleod AM, Campbell M, Cody JD et al. Cellulose, modified cellulose and
synthetic membranes in the haemodialysis of patients with end-stage renal disease.
The Cochrane Database of Systematic Reviews 2005 Issue 3. Art No:
CD003234.DOI:10.1002/14651858.CD003234.pub2
2 Eknoyan G, Levey AS, Beck GJ et al. The haemodialysis (HEMO) study:
rationale for selection of interventions. Semin Dial 1996;9:24–33
3 Eknoyan G, Beck GJ, Cheung AK et al. Effect of dialysis dose and flux on
mortality and morbidity in maintenance hemodialysis patients: Primary results
of the HEMO study. N Engl J Med 2002; 347:2010-2019
4 Cheung AK, Levin NW, Greene T et al. Effects of high-flux haemodialysis on
clinical outcomes: Results of the HEMO study. J Am Soc Nephrol 2003;
14:3251-3263
5 Greene T, Beck GJ, Gassman JJ et al. Design and statistical issues of the
haemodialysis (HEMO) study. Control Clin Trials 2000; 21: 502-525
6 Locatelli F, Pozzoni P, Di Filippo S. What are we expecting to learn from the
MPO study? Contrib Nephrol 2005; 149: 83-89
7 Locatelli F, Andrulli S, Pecchini F et al. Effect of high-flux dialysis on the
anaemia of haemodialysis patients. Nephrol Dial Transplant 2000; 15: 1399-
1409
8 Chauveau P, Nguyen H, Combe C et al. Dialyser membrane permeability and
survival in haemodialysis patients. Am J Kidney Dis 2005; 45:564-571
9 Schiffl H, Lang SM, Fischer R. Ultrapure dialysis fluid slows loss of residual
renal function in new dialysis patients. Nephrol Dial Transplant 2002; 17:
1814-1818
10 Termorshuizen F, Dekker FW, van Manen JG et al. Relative contribution of
residual renal function and different measures of adequacy to survival in
hemodialysis patients: an analysis of the Netherlands Cooperative Study on
the Adequacy of Dialysis (NECOSAD)-2. J Am Soc Nephrol 2004; 15: 1061-
1070
11 Rabindranath KS, Strippoli GF, Roderick P et al. Comparison of
haemodialysis, hemofiltration, and acetate free biofiltration for ESRD:
systematic review. Am J Kidney Dis 2005; 45:437-447
12 Locatelli F. Comparison of haemodialysis, haemodiafiltration and
haemofiltration: systematic review or systematic error. Am J Kidney Dis
2005; 46: 787-788 (and reply from the authors of reference 9, 788-789)
13 Karamperis N, Sloth E, Jensen JD. Predilution haemofiltration displays no
hemodynamic advantage over low-flux haemodialysis under matched
conditions. Kidney Int 2005;67:1601-1608
14 Canaud B, Bragg-Gresham JL, Marshall MR et al. Mortality risk for patients
receiving haemodiafiltration versus haemodialysis: European results from the
DOPPS. Kidney Int 2006; 69: 2087-2093
15 Penne EL, Blankestijn PJ, Bots ML et al. Effect of increased convective
clearance by on-line haemodiafiltration on all cause mortality in chronic
haemodialysis patients - the Dutch CONvective TRAnsport STudy
(CONTRAST): rationale and design of a randomised controlled trial. Curr
Control Trail Cardiovasc Med 2005; 6: 8

Patients without increased bleeding risk should be given low-dose
unfractionated heparin or LMWH during HD to reduce the risk of clotting of
the extracorporeal system. (Evidence) For patients with a risk of bleeding

anticoagulation should be avoided or kept to a minimum by using a high
blood flow rate and regular flushing of the extracorporeal circuit with saline
every 15-30 minutes.
Rationale
Extracorporeal anticoagulation is usually required to prevent thrombosis of all
forms of dialyser and extracorporeal circuit. Unfractionated heparin may be used
as the standard anticoagulant in view of its proven efficacy, ease of use and safety
record unless the patient has a history of recent or active bleeding or heparin
induced thrombocytopenia (1, 2). Heparin with a mean half-life of 1.5 hours is
best administered as a loading dose followed by a continuous infusion of 500-
1500 units/hour that is discontinued approximately 30 minutes before the end of
the dialysis session. Monitoring when required can be performed by measuring the
activated partial thromboplastin time or the whole-blood activated clotting time
aiming for around 150% of predialysis or normal values. The dosage of heparin
may need to be increased if there has been a substantive rise in the haematocrit
after correction of renal anaemia or reduced if the patient is on warfarin or
antiplatelet drugs. Low molecular weight heparin (LMWH) is an alternative agent
that has been associated with lower risk of bleeding, less frequent episodes of
hyperkalaemia and an improved lipid profile compared with standard heparin.
However a systematic review of 11 trials comparing the use of LMWH and
unfractionated heparin in HD patients concluded that there was no difference in
the incidence of bleeding complications, bleeding from the vascular access after
HD or thrombosis of the extracorporeal circuit (3).
For patients with a risk of bleeding anticoagulation should be avoided or kept to a

minimum by using a high blood flow rate and regular flushing of the
extracorporeal circuit with saline every 15-30 minutes (2). Alternatively heparin
may be replaced by a prostacyclin infusion or regional citrate anticoagulation. The
former may induce hypotension and is expensive whilst the latter requires careful
replacement of calcium and magnesium, monitoring of serum calcium and
magnesium levels during HD and is too complex for routine use. For patients with
heparin induced thrombocytopenia (HIT) or heparin induced thrombocytopenia
and thrombotic syndrome (HITTS) anticoagulation with either Argatroban,
heparinoids (danaparoid) or hirudin should be utilized instead of heparin.
References
1 Ouseph R, Ward RA. Anticoagulation for intermittent haemodialysis. Semin Dial
2000; 13:181-187
2 European Best Practice Guidelines for haemodialysis Part 1. Nephrol Dial Transplant
2002; 17: Supplement 7 S1-S111
3 Lim W, Cook DJ, Crowther MA. Safety and efficacy of low molecular weight
heparins for haemodialysis in patients with end-stage renal failure: a meta-analysis of
randomised trials. J Am Soc Nephrol 2004; 15:3192-3206

If it is planned to reuse dialysers that are marked ‘for single use only’ the
implications of dialyser reuse need to be considered carefully after reading
MDA Device Bulletin DB 2000(04) Single-use medical devices: implications
and consequences of reuse (good practice).
Rationale
Haemodialysers and their extracorporeal circuits contain sterile non-pyrogenic
pathways. Dialysers are generally marked for single use only although some are
now designed for multiple use in an individual patient. Reprocessing is a
combination of processes aimed at cleaning, disinfection and sterilisation of the
dialyser. Within the UK, reprocessing of items marked ‘for single use’ is
discussed in the Medical Devices Agency Device Bulletin DB 2000(04) Single-
use medical devices: implications and consequences of reuse. This is obtainable
from the Medical Devices Agency, Hannibal House, Elephant and Castle, London
SE1 6TQ. With reuse high flux biocompatible membranes can be used more cost
effectively. In recognition of this an agreement was reached between the Food and
Drug Administration (FDA) in the USA and the manufacturers, requiring that
some dialysers should be labelled ‘for multiple use’ and that manufacturers should
issue protocols for the safe reprocessing of their devices. Currently, manufacturers
have different marketing strategies in different countries and the main suppliers in
the UK do not currently supply ‘for multiple use’ labels with dialysers that are so
labelled in the USA.
Re-use has been shown to be safe in a number of studies and may have benefits,
specifically a reduction in beta-2-microglobulin (1). Some studies report an
overall reduction in mortality among patients treated with re-used dialysers (1)
although this may depend on the type of membrane used and on the agent used for
re-processing, the use of bleach being associated with lower mortality than the use
of formalin (2,3). Changing from multiple to single use of dialysers has been
reported recently to result in a reduction in the mortality rate in a large USA
population (4). It is standard practice to discard the dialyser whenever the hollow
fibre volume (total cell volume) is less than 80% of the initial measured value but
this method may not always be reliable in detecting dialyser dysfunction (5). The
significant costs and health and safety issues associated with reprocessing of
dialysers and the ongoing concerns about patient safety, reduced dialyser
efficiency and patient outcomes with reuse have led to reuse being discontinued in
the UK. The cost of high-flux dialysers is falling gradually and it is anticipated
that mass production will result in similar prices for high flux biocompatible
dialysers thus making them cost-effective without having to consider re-use.
References
1 Pollak VE, Kant KS, Parnell SL et al. Repeated use of dialyzers is safe: longterm
observations on morbidity and mortality in patients with end-stage renal disease.
Nephron 1986; 42:217–23
2 Held PJ, Pauly MV, Diamond L. Survival analysis of patients undergoing dialysis.
Jama 1987; 257:645–50
3 Port FK, Wolfe RA, Hulbert-Shearon TE et al. Mortality risk by haemodialyser reuse
practice and dialyzer membrane characteristics: results from the USRDS dialysis
morbidity and mortality study. Am J Kidney Dis 2001; 37:276–86
4 Lowrie EG, Li Z, Ofsthun N, Lazarus JM. Reprocessing dialysers for multiple uses:
recent analysis of death risks for patients. Nephrol Dial Transplant 2004; 19: 2823-
2830
5 Delmez JA, Weerts CA, Hasamear PD, Windus DW. Severe dialyser dysfunction
undetectable by standard reprocessing validation tests. Kidney Int 1989; 36: 478-484.

The use of dialysers sterilized with ethylene oxide should be avoided (good
practice).
Rationale
Chemical sterilization of dialysers and tubing with ethylene oxide has been
associated with anaphylactoid reactions (1) and this risk can be avoided by using
alternatives, such as steam or gamma radiation, for the sterilization of dialysers
(2).
References
1 Bommer J, WilhelmsOH, Barth HP et al. Anaphylactoid reactions in dialysis
patients: role of ethylene oxide. Lancet 1985; 2:1382-1385
Transplant 2002; 17: Supplement 7 S1-S111

Haemodialysis patients should not be treated with ACE inhibitor drugs and
AN 69 dialyser membranes at the same time. (Evidence)
Rationale
The concurrent use of AN 69 dialyser membranes in patients on ACE inhibitors
has been reported to cause haemodynamic instability attributable to bradykinin
(1). This interaction is preventable by changing the ACE inhibitor to an
angiotensin II antagonist or changing to a different dialysis membrane (2).
References
1 Verresen L, Fink E, Lemke HD, Vanrenterghem Y. Bradykinin is a mediator of
anaphylactoid reactions on AN69 membranes in patients receiving ACE inhibitors.
Kidney Int 1994; 45:1497-1503
5 Haemodialysis dose, frequency and

duration (Guidelines DIAL-HD 5.1 – 5.7)

HD should take place at least three times per week in nearly all patients with
end-stage chronic renal failure. Reduction of dialysis frequency to twice per
week because of insufficient dialysis facilities is unacceptable (good practice).
Audit measure 5
The proportion of patients in the main renal unit and its satellite units who
are on twice weekly haemodialysis.
Rationale
Twice weekly HD as a long-term form of chronic renal replacement therapy
should be discouraged. The most powerful determinant of solute removal is
dialysis frequency rather than duration. Twice per week HD is no longer regarded
as adequate and should be avoided. The frequency of twice weekly dialysis has
decreased world wide, including in the USA where it fell from 12.9% to 3.6% of
new patients between 1990 and 1996 (1). Some patients who live at far distances
from a HD unit remain on twice weekly HD and this small subgroup of patients
should be kept to minimum and receive much longer duration sessions. Twice
weekly HD without an increase in treatment time may be acceptable if patients
have a significant level of residual renal function, such as either a combined
urinary urea and creatinine clearance or eGFR above 5ml/min/1.73m2, provided
that residual renal function is monitored at least every 3 months and the frequency
of dialysis is increased when renal function decreases.
The three times per week HD schedule has evolved from empirical considerations
in the belief that it reconciles adequate treatment with adequate breaks between
treatments to provide the patient with a reasonable quality of life within a 7 day
treatment cycle. Furthermore, all outcome data from randomized prospective trials
have so far been derived from patient groups undergoing such dialysis schedules.
The National Co-operative Dialysis Study (NCDS), an historical US randomised
trial where cellulose membranes and acetate dialysate were used, has addressed
the issue of optimal dialysis time. This study randomised non-diabetic patients to
one of four dialysis regimens, two with short (2.5–3.5 hour) and two with longer
(4.5–5.0 hour) dialysis times, and two different time-averaged urea concentrations
in each arm (2). Longer dialysis gave a better outcome (2, 3). A combination of
better patient tolerance using improved machines and higher efficiency HD,
economic constraints and patient preference for shorter times has resulted in a
gradual reduction in the average length of dialysis sessions around the world.
More recently two approaches to more frequent dialysis sessions have been re-
evaluated. The first is dialysis for around two-three hours per day for five-six days
per week (often termed short daily HD) (4-8). The other approach is a renewed
interest in slow overnight treatment for 5–7 nights per week (often termed
nocturnal daily HD) that can (9-12):
a) deliver very large doses of dialysis (weekly Kt/V of almost 6 and much
greater removal of middle molecules)
b) remove sodium and water so that anti-hypertensive treatment can be reduced
to a minimum
c) permit regression of left ventricular hypertrophy
d) allow patients to follow an unrestricted diet
e) permit phosphate binders to be discontinued
f) improve sleep disturbance and sleep apnoea
Both regimes have been reported to give improved clinical outcomes such as
higher quality of life, fewer hospital admissions and reduced need for
erythropoietin when compared with the more conventional regime of three
sessions per week each of four hours (4-12). Daily HD may also be indicated in
the short term when patients develop an acute intercurrent illness or pericarditis.
On the basis of the successful reports from these observational studies of short
daily and nocturnal daily HD the National Institutes of Health (NIH) has
sponsored 2 prospective randomized studies in 250 patients to compare each form
of “daily” or frequent HD with standard thrice weekly HD. These NIH studies
(Frequent Haemodialysis Network Studies) are due to be completed in
2008/2009 (13).
References
1 Port FK, Orzol SM, Held PJ, Wolfe RA. Trends in treatment and survival for
haemodialysis patients in the United States. Am J Kidney Dis 1998; 32(Suppl 4):
S34–38
2 Lowrie EG, Parker TF, Parker TF, Sargent JA. Effect of the haemodialysis
prescription on patient morbidity: report from the National Cooperative Dialysis
Study. N Engl J Med 1981; 305:1176–1181
3 Lowrie EG, Teehan BP. Principles of prescribing dialysis therapy: implementing
recommendations from the National Cooperative Dialysis Study. Kidney Int Suppl
1983; 13:S113–122
4 Mastrangelo F, Alfonso L, Napoli M et al. Dialysis with increased frequency of
sessions (Lecce dialysis). Nephrol Dial Transplant 1998; 13(Suppl 6): 139–147
5 Twardowski ZJ. Daily dialysis: is this a reasonable option for the new millennium?
Nephrol Dial Transplant 2001; 16:1321–1324
6 Galland R, Traeger J, Arkouche W et al. Short daily haemodialysis and nutritional
status. Am J Kidney Dis 2001; 37(Suppl 2): S95–98
7 Goldfarb-Rumyantzev AS, Leypoldt JK, Nelson N et al. A crossover study of short
daily haemodialysis. Nephrol Dial Transplant 2005; 21: 166-175
8 Woods JD, Port FK, Orzol S et al. Clinical and biochemical correlates of starting
"daily" haemodialysis. Kidney Int 1999; 55: 2467-2476
9 Fagugli RM, Reboldi G, Quintaliani G et al. Short daily haemodialysis: blood
pressure control and left ventricular mass reduction in hypertensive haemodialysis
patients. Am J Kidney Dis 2001; 38: 371-376
10 Pierratos A. Nocturnal home haemodialysis: an update on a 5-year experience.
Nephrol Dial Transplant 1999; 14:2835-2840
11 Mucsi I, Hercz C, Uldall R et al. Control of serum phosphate without any phosphate
binder in patients treated with nocturnal haemodialysis. Kidney Int 1998; 53:1399-
1404
12 Raj DS, Charra B, Pierratos A, Work J. In search of ideal haemodialysis: is prolonged

frequent dialysis the answer? Am J Kidney Dis 1999; 34:597–610
13 Walsh M, Culleton B, Tonelli M, Manns B. A systematic review of the effect of
nocturnal haemodialysis on blood pressure, left ventricular hypertrophy, anaemia,
mineral metabolism and health-related quality of life. Kidney Int 2005; 67:1500-1508
14 www.niddk.nih.gov/patient/hemodialysis/hemodialysis.htm

Every patient with end-stage chronic renal failure receiving thrice weekly HD
should have consistently:
• either urea reduction ratio (URR) > 65%

• or equilibrated Kt/V of >1.2 (or sp Kt/V of > 1.3) calculated from pre-
and post-dialysis urea values, duration of dialysis and weight loss
during dialysis (Evidence)
To achieve a URR above 65% or eKt/V above 1.2 consistently in the vast
majority of the haemodialysis population clinicians should aim for a minimum
target URR of 70% or minimum eKt/V of 1.4 in individual patients. Aiming for
these target doses also addresses the concerns raised by recent data that suggest
that women and patients of low body weight may have improved survival rates if
the URR is maintained above 70% or eKt/V is at least 1.4.
Audit measure 2
Cumulative frequency curves of urea reduction ratio measured using a
standard method of post-dialysis sampling
Rationale
Dialysis adequacy is a global concept that includes the clinical assessment of
general well-being, nutrition, the impact on the patient’s quality of life, anaemia,
blood pressure and fluid status as well as measures of clearance of putative
uraemic toxins by the dialysis process. The molecular weights of the solvent and
solutes to be cleared by dialysis range over three orders of magnitude, from small
(water, urea) to large (beta-2-microglobulin). Adequate clearance of the whole
range of molecules by dialysis is important and in the future monitoring of beta-2-
microglobulin levels may be used to assess dialysis adequacy. For practical
reasons HD adequacy thus far has been measured using small, easily measured
solutes such as urea (1-3)
Three methods of assessing urea removal are in current use (1, 2):
a) The URR (4) is the simplest. The percentage fall in blood urea achieved by a
dialysis session is measured as follows:
{(pre-dialysis [urea] – post-dialysis [urea])}/pre-dialysis (urea) • 100%.
The URR is easy to perform and is the most widely used index of dialysis dose
used in the UK. URR does not take solute removal via ultrafiltration or residual
renal function or urea generation during dialysis into account (5, 6) and hence
adjustment of dialysis dose to achieve a particular target will result in higher

overall urea removal than predicted from the percentage reduction in blood urea.
However these drawbacks are not important if the main aim of measuring small
solute removal by HD is to ensure that a minimum target dialysis dose is
delivered consistently. A number of large observational studies in populations of
HD patients have shown that variations in URR are associated with major
differences in mortality and have led to recommendations that the URR should be
at least 65% (7-10).
b) Kt/V urea can also be predicted from one of several simple formulae requiring
as input data the pre- and post-dialysis urea concentrations, the duration of
dialysis, and the weight loss during dialysis. Kt/V can be calculated using several
formulae giving different results (11) and hence, if Kt/V is being used for
comparative audit, it is important that the raw data are collected to allow
calculation of URR and estimated Kt/V using a single formula. The second
generation formula validated and reported by Daugirdas is recommended (12).
c) Urea kinetic modeling (Formal UKM), the most complex measure, involves
analysis of the fall in (urea) during HD, the rise in (urea) in the interdialytic
period, clearance of urea by residual renal function, and the total clearance
predicted from the dialyser clearance, blood and dialysate flow, time on dialysis,
and fluid removal during dialysis. Therefore UKM requires collection of
additional data on dialyser clearance, an interdialytic urine collection for
measurement of urea concentration and volume, and measurement of pre-dialysis
urea concentration on the subsequent dialysis. These data are fed into a computer
programme which, assuming steady state, calculates Kt/V urea and normalised
protein catabolic rate (5). Kt/V measured by formal UKM is more accurate than
URR, particularly at high values of URR and Kt/V (3). Its use allows accurate
prediction of the effects of changing one particular component of the dialysis
prescription (e.g. dialyser size, dialysis duration, blood flow rate) on the delivered
dialysis dose although this benefit has been overstated given the limited number
of practical options for changing the dialysis prescription. UKM also may give
valuable information on urea generation rate and protein catabolic rate. If the
patient is in a steady state nutritionally, this gives information on current protein
intake, and may be a useful adjunct to other methods of assessment of nutritional
status.
However doubts have been raised whether Kt/V is a good index of dialysis dose
since survival rates on HD are higher in patients with larger body size and better
nutrition even though this patient group tends to have lower Kt/V values (13, 14).
Non-normalised dialysis dose (Kt) has been proposed as an alternative and better
index of dialysis dose adequacy to Kt/V since the former index obviates the trend
for smaller patients with poorer nutritional status to be accorded a higher dialysis
dose (15,16). In a large cross-sectional analysis using Kt as the index of dialysis
dose mortality risk was observed to fall if the delivered dialysis dose was a
minimum Kt of 42 litres in women and 48 litres in men (13). A further difficulty
with the use of the Kt/V index for other than thrice weekly HD is that the
significance of any weekly Kt/V value depends on the frequency of dialysis since
more frequent dialysis therapies, such as daily HD, will deliver greater small
solute removal at the same weekly Kt/V.
Most UK haemodialysis units only collect pre- and post-dialysis urea

concentration, and only a very few perform UKM. For comparative audit, the
choice therefore currently lies between calculation of URR and estimation of Kt/V
urea from such data.
The optimal dialysis dose has not been well defined but minimum targets of
delivered dose measured by URR and Kt/V have been established. A retrospective
analysis of the National Co-operative Dialysis Study suggested that a Kt/V of 1.0
was the watershed between ‘good’ dialysis (Kt/V >1.0) and inadequate dialysis
(Kt/V <1.0). Thereafter Kt/V survived as an index of dialysis adequacy (17).
More recent studies (7-9,18-20) have shown a reduction in mortality rates with
increases in dialysis dose measured in various ways with some of the studies
adjusting for co-morbidity (8,20). One study has shown no further reduction in
mortality above Kt/V of 1.3 or URR of 70% (7). Many commentators, however,
believed that some further improvement in mortality risk could be achieved with
Kt/Vs of up to 1.6 or even higher (21-23). The HEMO trial was a prospective
randomised controlled trial in which 1846 patients were randomised to achieve a
standard-dose goal of an equilibrated Kt/V of 1.05 (URR circa 65%) or a high-
dose goal of an eKt/V of 1.45 (URR circa 75%) and to synthetic or semi-synthetic
membranes of high or low flux in a 2 x 2 factorial design. (24). The HEMO study
showed no difference in patient survival or secondary end-points between the two
groups after a mean follow-up period of 2.8 years. No difference in patient
outcomes was observed in the two groups even although dialysis doses were well
separated with achieved eKt/V of 1.16 in the standard-dose group (spKt/V 1.3 +
0.1 ; URR 66.3 + 2.5%) and eKt/V of 1.53 in the high-dose group (spKt/V 1.7 +
0.1 ; URR 75.2 + 2.5%). Subgroup analysis of the HEMO study showed that
survival rates in women randomized to the higher dose group were higher than
women in the lower dose group (relative risk 0.81 ; p = 0.02) and this association
persisted after adjusting for different indices of body size (25). An association
between higher dose and lower mortality rates in women but not in men was
confirmed using the average URR of incident HD patients in the USA and eKt/V
of HD patients in the DOPPS data from 7 countries (26). Further analyses of the
HEMO study showed that differences in dialysis dose and membrane flux had no
effect on the proportion of infection-related deaths (27).
Based upon the above evidence we have retained the standard dose as a URR of
65% or an eKt/V of 1.2, which should be regarded as the minimum dialysis dose
delivered thrice weekly. To ensure as many patients as possible achieve this
standard consistently the target dose should be a URR of 70% or eKt/V of 1.4. As
with all standards, achievement is dependent on patients’ concordance with
treatment. This includes the agreement of the patient to increase treatment
duration if the delivered dialysis dose is inadequate after the dialyser blood flow
rate, dialysate flow rate and dialyser performance have been increased to the
maximum that can be achieved. Increased understanding amongst patients of the
benefits of an adequate dialysis dose should help to improve outcomes. The
proportion of dialysis sessions that are missed or shortened by the patient should
be audited in each unit.
Time-dependent Cox regression analysis of the HEMO study has shown that
mean pre-dialysis serum beta-2-microglobulin levels but not dialyser beta-2-
microglobulin clearances were associated with all cause mortality with a relative
risk of 1.11 per 10mg/L rise in the beta-2-microglobulin concentration above a
reference value of 27mg/L (CI 1.05-1.19 ; p = 0.001) after adjusting for residual
renal function and pre-study years on dialysis (28). This evidence provides
support for the use of beta-2-microglobulin to assess adequacy of HD in future
both as an indicator of patient outcome and a surrogate marker of middle
molecule removal (29). The apparent disparity between the prognostic effects of
serum beta-2-microglobulin levels and dialyser beta-2-microglobulin clearances
(28) is most likely due to the limited mass removal of beta-2-microglobulin in
high-efficiency dialysis due to intercompartmental transfer resistance within the
patient which results in rebound of serum beta-2-microglobulin levels at the end
of therapy (30). This observation on beta-2-microglobulin intradialytic kinetics
provides further support for the use of longer duration and/or more frequent
dialytic therapies (29).
References
1 Depner T. Hemodialysis kinetic modeling. In: Henrich WL (ed). The principles and
practice of dialysis, 2nd edn. Philadelphia: Lippincott, Williams and Wilkins,1999:
Chapter 7.
2 Gotch F. Urea kinetic modelling to guide haemodialysis therapy in adults. In:
Nissenson AR, Fine RN (eds). Dialysis therapy, 2nd edn. Philadelphia: Hanley and
Belfus, 2002; Section 6:117–121
3 Sherman RA, Hootkins R. Simplified formulas for monitoring haemodialysis
adequacy. In: Nissenson AR, Fine RN (eds). Dialysis therapy, 3rd edn. Philadelphia:
Hanley and Belfus, 2002; Section 6:122–126
4 Lowrie EG, Lew NL. The urea reduction ratio (URR): A simple method for
evaluating haemodialysis treatment. Contemp Dial Nephrol 1991; 12(2): 11–20
5 Depner TA. Assessing adequacy of haemodialysis: urea modeling. Kidney Int 1994;
45:1522–35
6 Sherman RA, Cody RP, Rogers ME et al. Accuracy of the urea reduction ratio in
predicting dialysis delivery. Kidney Int 1995; 47:319-321
7 Held PJ, Port FK, Wolfe RA et al. The dose of haemodialysis and patient mortality.
Kidney Int 1996; 50:550–556
8 McClellan WM, Soucie JM, Flanders WD. Mortality in end-stage renal disease is
associated with facility-to-facility differences in adequacy of haemodialysis. J Am
Soc Nephrol 1998; 9:1940–1947
9 Parker T, Husni L, Huang W et al. Survival of haemodialysis patients in the United
States is improved with a greater quantitiy of dialysis. Am J Kidney Dis 1994; 23:
670-680
10 Owen WF, Lew NL, Lowrie EG et al. The urea reduction ratio and serum albumin
concentration as predictors of mortality in patients undergoing haemodialysis. N Eng
J Med 1993; 329: 1001-1006
11 Movilli E. Simplified approaches to calculate Kt/V. Its time for agreement. Nephrol
Dial Transplant 1996; 11:24–27
12 Daugirdas JT. Second generation logarithmic estimates of single-pool
variable3volume Kt/V; an analysis of error. J Am Soc Nephrol 1993; 4:1205-1213
13 Lowrie EG, Chertow G, Lazarus JM. Owen WF. The mortality effect of the clearance
x time product. Kidney Int 1999; 56: 729-737
14 Lowrie EG, Chertow GM, Lew NL et al. The urea clearance x dialysis time product
(Kt) as an outcome-based measure of haemodialysis dose. Kidney Int 1999; 56: 729-
737
15 Li Z, Lew NL, Lazarus JM, Lowrie EG. Comparing the urea reduction ratio and urea
product as outcome-based measures of haemodialysis dose. Am J Kidney Dis 2000;
35: 598- 605
16 Lowrie EG, Li Z, Ofsthun N, Lazarus JM. Measurement of dialyser clearance,
dialysis time, and body size: death risk relationships among patients. Kidney Int
2004; 66: 2077- 2084
17 Gotch FA, Sargent JA. A mechanistic analysis of the National Cooperative Dialysis
Study (NCDS). Kidney Int 1985; 28:526–534
18 Hakim RM, Breyer J, Ismail N, Schulman G. Effects of dose of dialysis on morbidity
and mortality. Am J Kidney Dis 1994; 23:661–669
19 Collins AJ, Ma JZ, Umen A et al. Urea index and other predictors of haemodialysis
patient survival. Am J Kidney Dis 1994; 23:272–282
20 Bloembergen WE, Hakim RM, Stannard DC et al. Relationship of dialysis membrane
and cause-specific mortality. Am J Kidney Dis 1999; 33:1–10
21 Hornberger JC. The haemodialysis prescription and quality-adjusted life expectancy.
Renal Physicians Association Working Committee on Clinical Guidelines. J Am Soc
Nephrol 1993; 4:1004–1020
22 Hornberger JC. The hemodialysis prescription and cost effectiveness. Renal
Physicians Association Working Committee on Clinical Guidelines. J Am Soc
Nephrol 1993; 4:1021–1027
23 Wolfe RA, Ashby VB, Daugirdas JT et al. Body size, dose of haemodialysis, and
mortality. Am J Kidney Dis 2000; 35:80–88
24 Eknoyan G, Beck GJ, Cheung AK et al. Effect of dialysis dose and flux on mortality
and morbidity in maintenance hemodialysis patients: Primary results of the HEMO
study. N Engl J Med 2002; 347:2010-2019
25 Depner T, Daugirdas J, Greene T et al. Dialysis dose and the effect of gender and
body size on outcome in the HEMO study. Kidney Int 2004; 65: 1386-1394
26 Port FK, Wolfe RA, Hulbert-Shearon TE et al. High dialysis dose is associated with
lower mortality among women but not among men. Am J Kidney Dis 2004; 43:
1014-1023
27 Allon M. Depner TA, Radeva M, et al. Impact of dialysis dose and membrane on
infection-related hospitalisation and death: Results of the HEMO study. J Am Soc
Nephrol 2003; 14: 1863- 1870
28 Cheung AK, Rocco MV, Yan G et al. Serum beta-2-microglobulin levels predict
mortality in dialysis patients: results of the HEMO study. J Am Soc Nephrol 2006;
17: 546-555
29 Canaud B, Morena M, Cristal JP, Krieter D. Beta-2-microglobulin, a uraemic toxin
with a double meaning. Kidney Int 2006; 69; 1297-1299
30 Ward RA, Greene T, Hartmann B, Samtleben W. Resistance to intercompartmental
mass transfer limits beta-2-microglobulin removal by post-dilution
haemodiafiltration. Kidney Int 2006; 69:1431-1437

The duration of thrice weekly HD in adult patients with minimal residual
renal function should not be reduced below 4 hours without careful
consideration (good practice).
Audit measure 7
The proportion of patient non-attendances for dialysis and the proportion of
dialysis sessions shortened at the patient’s request.
Rationale
It is difficult to separate the influence of dialysis time and dose on patient
outcomes (1). Early studies showed that the risk of death is associated with short
dialysis duration (2). Dialysers with higher mass transfer area coefficients in
combination with higher blood and dialysis fluid flow rates have been used to
provide higher efficiency HD than in the past. The urea clearance will depend on
whichever is the lowest of the blood flow rate, dialyser urea mass transfer
coefficient and dialysate flow rate. Since small solute urea removal can be
formally quantified by validated techniques, dialysis times have been reduced
while maintaining ‘equivalence’ in the degree of blood urea purification. A
crossover study of standard and higher efficiency HD prescriptions delivering
equal dialysis dose (urea removal) measured by direct dialysate quantification has
shown lower phosphate and beta-2-microglobulin removal and less bicarbonate
absorption during the shorter duration, higher efficiency prescription (3).
Improved clearance of iohexol was also observed on longer duration HD with
similar Kt/V. Thus, when short and standard duration HD provide equal urea
clearances, delivered dialysis therapy should not be regarded as equivalent.
Alternatively changing to treatment modalities that provide both convective and
diffusive removal of solutes such as haemodiafiltration have been used to lower
treatment times although shortening the duration of haemodiafiltration will tend to
negate its benefits of providing higher middle molecule clearances.
Retrospective data from a large Japanese population have shown that dialysis
duration up to 5.5 hours was associated with improved patient survival after
adjusting for dialysis dose (4). Very low mortality rates were observed in Tassin
in patients treated with long duration thrice weekly HD with mean spKt/V of 1.67
+ 0.41 (5). However, a Cox analysis showed that patient survival was linked to
improved blood pressure control and lower cardiovascular mortality related to the
achievement of better long-term control of dry body weight (5). Conversely high
efficiency HD has been associated with poor blood pressure control. In the USA
patients who received dialysis for less than 3.5 hours per session three times per
week had approximately twice the risk of death of patients on HD for more than 4
hours three times per week (2). Cox regression analyses of data from the Dialysis
Outcomes and Practice Patterns Study (DOPPS) and the Australian and New
Zealand Dialysis and Transplant Registry have shown that patient survival was
greater in patients if treatment times were above 4 hours and 4.5 hours,
respectively (6,7) and both of these observational studies have concluded that a
randomised controlled study of longer dialysis sessions in thrice weekly HD is
needed. These observations suggest that the duration of thrice weekly HD should
be reduced below 4 hours with caution unless the patient has significant residual
renal function or low body weight.
Delivered treatment times and hence weekly dialysis dose are reduced if either the
patient requests to discontinue the dialysis session early or if the patient attends
for dialysis irregularly. Non-adherence to the prescribed dialysis schedule should
be kept to a minimum and monitored.
References
1 Kjellstrand CM. Duration and adequacy of dialysis Overview: the science is easy, the
ethic is difficult. Asaio J 1997; 43:220–224
2 Held PJ, Levin NW, Bovbjerg RR et al. Mortality and duration of haemodialysis
treatment Jama 1991; 265:871–875
3 Mactier RA, Madi AM, Allam BF. Comparison of high-efficiency and standard
haemodialysis providing equal urea clearances by partial and total dialysate
quantification Nephrol Dial Transplant 1997; 12: 1182-1186
4 Shinzato T, Nikai S, Akiba T et al. Survival in long-term haemodialysis patients
results from the annual survey of the Japanese Society of Dialysis Therapy. Nephrol
Dial Transplant 1997; 12: 884-888
5 Charra B, Calemard E, Ruffet M et al. Survival as an index of adequacy of dialysis.
Kidney Int 1992; 41:1286-1291
6 Saran R, Bragg-Gresham JL, Levin NW et al. Longer treatment time and slower
ultrafiltration in haemodialysis: Associations with lower mortality in the DOPPS.
Kidney Int 2006; 69: 1222-1228
7 Marshall MR, Byrne BG, Kerr PG, McDonald SP. Associations of haemodialysis
dose and session length with mortality risk in Australian and New Zealand patients.
Kidney Int 2006; 69:1229-1236

Patients receiving dialysis twice weekly for reasons of geography should
receive a higher sessional dose of dialysis (good practice).
Rationale
A wider distribution of small satellite HD units would help reduce the need to
accept twice weekly HD for lifestyle reasons. Twice weekly HD effectively means
that the patient will require longer duration HD, usually at least 6 hours twice per
week. It should be acknowledged if this cannot be achieved and patients who are
receiving twice weekly HD without an increase in treatment time should be
informed explicitly that this is a compromise between the practicalities of dialysis
and their long-term health.

Measurement of the ‘dose’ or ‘adequacy’ of HD should be performed
monthly in all hospital HD patients and may be performed less frequently in
home HD patients (good practice).
Rationale
Monthly measurement of dialysis dose in hospital HD patients should be used to
optimize the HD prescription and may facilitate early detection of poorly
functioning vascular access. Monitoring of dialysis dose in home HD patients on a
monthly basis is impractical and may be performed on a less frequent basis such
as every 3 months. All dialysis units should collect and report data on dialysis
adequacy to their regional network and the UK Renal Registry. Meaningful
comparative audit within a renal unit or regional network requires the use of the
same methodology of measurement of dialysis dose and blood sampling during a
mid-week HD session in the census week.

Post-dialysis blood samples should be collected either by the stop-dialysate
flow method, the slow-flow method or the simplified stop-flow method. The
method used should remain consistent within renal units and should be
reported to the Registry. (Evidence)
Rationale
All methods of measuring dialysis dose require accurate measurement of pre-
dialysis and post-dialysis urea concentrations on a mid-week dialysis session. Full
urea kinetic modeling also requires:
• measurement of dialyser clearance
• measurement of weight loss during dialysis
• collection of an inter-dialytic urine
• pre-dialysis blood urea concentration from the subsequent dialysis session.
Contamination of the post-dialysis sample with blood returning from the dialyser
or heparin, or sampling from a fistula or other access device in which there is
recirculation of dialysed blood will lead to falsely low measurements, and thus to
over-estimation of dialysis dose. True venous blood urea concentration rises
rapidly in the first few minutes after dialysis as the effects of access and
cardiopulmonary recirculation dissipate (1). It continues to rise at a rate higher
than that expected from urea generation for up to 30 minutes, as a consequence of
continued transfer of urea from peripheral body compartments into the
bloodstream (2-6); the earlier the sample is drawn, the higher the apparent
delivered dialysis dose. Small variations in the timing and technique with which
post-dialysis blood samples are drawn can, therefore, result in clinically important
errors in the estimated dose of dialysis. Such variation has been shown to be
common in the USA (7) and in the UK (8,9). Techniques of post-dialysis blood
sampling that involve taking the sample immediately at the end of the HD session
were used commonly in the USA in the past (7), generate a higher apparent URR
and may have contributed to the rise in the URR deemed necessary for optimum
survival in observational studies.
Several methods of post-dialysis blood sampling are in use in the UK:
a) Stop dialysate flow method (validated by Drs. Geddes, Traynor and

Mactier, NHS Glasgow, and used by all of the HD units in Scotland since
1999; ref 8-10).
At the end of the dialysis time stop dialysate flow but keep the blood pump
running.
After 5 minutes with no dialysate flow take a blood sample from anywhere
in the blood circuit.
b) Slow-flow method (developed by F Gotch and M Keen, Davis Medical

Centre, San Francisco and used by Lister Renal Unit, East & North Herts NHS
Trust since 1990; ref 3).
At the end of the dialysis time turn the blood pump speed down to 100 ml
per min.
Override alarms to keep blood pump operating.
Wait 15–30 seconds and take samples from the “arterial” line sampling
port.
If more than one blood sample is required, the sample for urea should be
the first one taken.
c) Simplified stop-flow method (developed by EJ Lindley, V Osborne, S

Sanasy, D Swales and M Wright. The Leeds Teaching Hospitals NHS Trust).
When you are ready to take the sample turn the blood pump speed slowly
down to 50 ml per min.
Start counting to five; if the venous pressure alarm has not already stopped
the blood pump when you get to five stop the pump manually.
Disconnect the arterial line and take a sample from the needle tubing (or
the arterial connector of the catheter) within 20 seconds of slowing the
blood pump speed to 50 ml per min.
If more than one blood sample is required, the urea sample should be the
first one taken.
The stop dialysate flow method avoids the dilutional effects of access and
cardiopulmonary recirculation and is a 2 step process involving switching off the
dialysate flow for 5 minutes at the end of the HD session and then taking a blood
sample from the arterial or venous port (8). The stop dialysate flow method is
simple, easily reproducible, suitable for all forms of vascular access, validated in
haemodiafiltration as well as HD (8,9) and is currently the most widely used
method in the UK. The slow-flow method and the stop-flow method were devised
to give early post-dialysis measurements which avoid the effects of access re-
circulation but do not allow for cardiopulmonary recirculation which continues for
the first 2 minutes after the end of HD using a fistula or graft as vascular access
(1). The stop and slow flow methods will underestimate post-dialysis
“equilibrated” blood urea concentrations more than the stop dialysate flow method
and consequently overestimate urea removal by HD.
Post-dialysis rebound in venous blood urea concentration results from continued

return of blood from poorly dialysed body ‘compartments’, and is particularly
marked after high efficiency dialysis. Accurate comparison of delivered dialysis
dose therefore requires estimation of the equilibrated blood urea concentration,
allowing calculation of ‘equilibrated’ Kt/V. Full re-equilibration takes about 30
minutes, but it is impractical to ask patients to wait this long for post-dialysis
blood sampling on a routine basis. The amount of rebound is determined by
several factors including the efficiency of dialysis and the size of the patient.
Formulae have been validated for predicting 30 minute post-dialysis or
“equilibrated” blood urea from blood samples using either the stop dialysate flow
method (10) or similar sampling methods to the slow flow and stop flow methods
(3,6,11). A standardized approach to post-dialysis blood sampling is preferable for
comparative audit (12) and the stop dialysate flow method was adopted by all of
the adult renal units in Scotland since it is simple, practical, well validated and the
least likely method to overestimate the URR or Kt/V. The stop dialysate flow and
slow-flow methods are the two methods included in Guideline 3.4 of the latest
update of the KDOQI Clinical Practice Guidelines on Haemodialysis Adequacy
(13).
References
1 Schneditz D, Kaufman AM, Polaschegg HD et al. Cardiopulmonary recirculation
during haemodialysis. Kidney Int 1992; 42: 1450-1456
2 Lai YH, Guh JY, Chen HC, Tsai JH. Effects of different sampling methods for
measurement of post dialysis blood urea nitrogen on urea kinetic modeling derived
parameters in patients undergoing long-term haemodialysis. Asaio J 1995; 41:211–
215
3 Tattersall JE, De Takats D, Chammey P et al. The post-haemodialysis rebound:
predicting and quantifying its effect on Kt/V. Kidney Int 1996; 50:2094–2102102
4 Tattersall JE, Chammey P, Aldridge C et al. Recirculation and the post-dialysis
rebound. Nephrol Dial Transplant 1996; 11(Suppl 2):75–80
5 Daugirdas JT, Smye SW. Effect of a two compartment distribution on apparent urea
distribution volume. Kidney Int 1997; 51:1270–1273
6 Smye SW, Tattersall JE, Will EJ. Modeling the postdialysis rebound: the
reconciliation of current formulas. Asaio J 1999; 45:562–567
7 Beto JA, Bansal VK, Ing TS et al. Variation in blood sample collection for
determination of haemodialysis adequacy. Am J Kidney Dis 1998; 31:135-141
8 Geddes C C, Traynor J P, Walbaum D et al. A new method of post-dialysis blood
urea sampling: the “stop dialysate flow” method. Nephrol Dial Transplant 2000;
15:517-523
9 Traynor JP, Oun HA, McKenzie P et al. Assessing the utility of the stop dialysate
flow method in patients receiving haemodiafiltration. Nephrol Dial Transplant 2005;
20: 2479-2484
10 Traynor JP, Geddes CC, Ferguson C, Mactier RA. Predicting 30-minute postdialysis
rebound blood urea concentrations using the stop dialysate flow method. Am J
Kidney Dis 2002; 39:308-314
11 Daugirdas JT. Second generation logarithmic estimates of single-pool
variable3volume Kt/V; an analysis of error. J Am Soc Nephrol 1993; 4:1205-1213
2004.
13 (www.renalreg.com Renal Association Standards & Audit Subcommittee)
14 National Kidney Foundation-K/DOQI Clinical Practice Guidelines and Clinical
Practice Recommendations for Haemodialysis Adequacy, 2005 (in press)

Patients with acute renal failure should initially receive daily renal
replacement therapy. (Evidence)
Rationale
At present there is no evidence to show whether continuous or intermittent renal
replacement therapies or whether haemofiltration or HD provide better survival in
patients with acute renal failure. In a randomised, risk stratified, dose equivalent
prospective comparison of continuous veno-venous HD (CVVHD) versus
intermittent HD in 80 intensive care unit patients with acute renal failure the
CVVHD group had greater daily fluid volume removal but no improvement in
patient survival, preservation of urinary output or recovery of renal function (1).
A randomized study of extended daily HD and continuous HD in intensive care
patients with acute renal failure showed no difference in haemodynamic stability
(2). However there is evidence that survival in patients with acute renal failure is
better with daily than alternate day renal replacement therapy (3). In this
randomized prospective study of 160 critically ill patients with acute renal failure
the mortality rate using an intention-to-treat analysis was 28% with daily HD and
46% with alternate day HD (p<0.01). The frequency of renal replacement therapy
may be reduced once the metabolic syndrome and fluid status of patients with
acute renal failure is stable. Initial randomized studies showed that the use high
flux biocompatible membranes were associated with improved patient survival
rates in acute renal failure but this has not been confirmed in follow-up studies
(4). 58% of the 90 patients randomly assigned to bioincompatible Cuprophan
dialysers survived compared with 60% of the 90 patients assigned to
polymethylmethacrylate membranes (4). A randomized study of continuous veno-
venous haemofiltration in acute renal failure has shown improved patient survival
in patients prescribed at least 35ml/hour/kg body weight (5). Extended daily HD
and post-dilutional continuous veno-venous haemofiltration are widely utilized in
the management of acute renal failure in the UK and both provide long duration
therapy to help maintain adequate fluid balance with minimal adverse
haemodynamic effects in this critically ill patient group.
References
4 Augustine JJ, Sandy D, Seifert TH, Paganini EP. A randomized controlled trial
comparing intermittent with continuous dialysis in patients with acute renal failure.
Am J Kidney Dis 2004; 44:1000-1007
5 Kumar VA, Yeun JY, Depner TA, Don BR. Extended daily dialysis versus
continuous haemodialysis for ICU patients with acute renal failure: a two-year single
centre report. Int J Artif Organs 2004; 27: 371-379
6 Schiffl H, Lang SM, Fischer R. Daily haemodialysis and the outcome of acute renal
failure. N Eng J Med 2002; 346: 305-310
7 Jorres A, Gahl, GM, Dobis C et al. Haemodialysis-membrane biocompatibility and
mortality of patients with dialysis-dependent acute renal failure: a prospective
randomized multicentre trial. International Multicentre Study Group. Lancet 1999;
354: 1337-1341
8 Ronco C, Bellomo R, Homel P et al. Effects of different doses in continuous veno-
venous haemofiltration on outcomes of acute renal failure: a prospective randomized
trial. Lancet 2000; 356: 26-30
6 Laboratory and clinical indices of dialysis

adequacy other than dialysis dose

Blood sampling for biochemical and haematological measurements should be
performed before a mid-week HD session using a dry needle or syringe (good
practice).
Rationale
Too much emphasis may have been placed in the past on achieving a given
standard of small solute clearance at the expense of addressing a wide range of
other important clinical and laboratory parameters of dialysis adequacy. A global
assessment of dialysis adequacy includes achievement of good control of:
• hyperkalaemia and metabolic acidosis

• bone metabolism
• anaemia
• hypertension and fluid balance
• traditional and non-traditional cardiovascular risk factors
• nutritional status
Variability in interdialysis fluid weight gains after the 1 or 2 day intervals between
HD sessions may be expected to cause differing degrees of haemodilution and so
influence pre-dialysis haemoglobin and albumin concentrations. A recent study
has shown higher pre-dialysis serum calcium and phosphate concentrations after
the longer interdialysis interval in the absence of evidence of different levels of
haemodilution between short and long interdialysis intervals (1). These findings
indicate that time-interval related interdialytic and non-dialytic factors may
influence pre-dialysis biochemical and haematological results and reinforce the
need for standardization of timing of pre-dialysis blood sampling in HD patients.
The UK Renal Registry and Scottish Renal Registry have employed audit
measures using measurement of laboratory values from samples that were
collected before commencing HD after a one day interdialysis interval. To avoid
blood sampling at weekends blood sampling is effectively limited to either a
Wednesday or Thursday dialysis session. This restricted timing of blood sampling
allows standardization of interpatient and intrapatient interdialysis fluid weight
gains and it is important that all samples are taken using a dry needle or syringe to
ensure dilutional sampling errors are avoided.
References
1 Sigrist MK, Devlin L, Taal MW, Fluck RJ, McIntyre CW. Length of interdialysis
interval influences serum calcium and phosphorus concentrations. Nephrol Dial
Transplant 2005; 20:1643-1646

Monitoring of pre-dialysis biochemical and haematological parameters
should be performed monthly in hospital HD patients and at least 3 monthly
in home HD patients (good practice).
Rationale
Standardised analytical methods of measuring laboratory indices are required if
comparative audit against target standards is to be meaningful. Difficulties still
arise since laboratories across the UK use different methods to measure serum
albumin and different correction factors for adjusting serum calcium levels (1).
References
2004. (www.renalreg.com)

Pre-dialysis serum bicarbonate concentrations measured with minimum
delay after venepuncture should be between 20 and 26mmol/l. (good
practice).
Rationale
The main causal factors of metabolic acidosis in stable HD patients are inadequate
dialysis delivery, excessive animal protein (sulphur containing amino acid) intake
and high interdialysis weight gains. Whole-body base balance studies in 18 anuric
HD patients have highlighted the importance of interdialysis dilution in the
aetiology of predialysis acidosis (1). In ill patients metabolic acidosis may also be
due to increased protein catabolism, hypotension or hypoxia induced lactate
production or bicarbonate losses associated with co-morbid illness. Pre-dialysis
metabolic acidosis has a range of adverse consequences: an increase in protein
catabolism and anti-anabolic effects, negative inotropic effect, loss of bone
mineral, insulin resistance, growth retardation in children, reduced thyroxine
levels, altered triglyceride metabolism, hyperkalaemia, lower serum leptin levels
and greater accumulation of beta-2-microglobulin.
Pre-dialysis venous bicarbonate levels between 17.5 and 20 mmol/l were

associated with the lowest risk of death in a large cohort study of 13535
hemodialysis patients whilst the relative risk of death was increased threefold if
the pre-dialysis venous bicarbonate was < 15 mmol/l (2). In a DOPPS study of
more than 7000 unselected HD patients the corrected mid-week serum bicarbonate
concentration averaged 21.9 mmol/l and correlated inversely with the nPCR and
serum albumin (3). The adjusted risk of death, hospitalization or malnutrition was
higher in patients with serum bicarbonate levels less than 16 or above 24 when
compared with patients in the reference group with moderate pre-dialysis acidosis
(3). Short-term benefits of correcting pre-dialysis acidosis from below 19mmol/l
to 24mmol/l, by either increasing the dialysate bicarbonate concentration (4-7) or
the addition of oral bicarbonate supplements (8), have been shown in several
small crossover studies. Correction of acidosis reduced whole body protein
degradation in a study of 6 patients (4), increased the sensitivity of the parathyroid
glands to serum calcium in studies of 21 and 8 patients (5,6), improved triceps

skin thickness as an index of nutritional status in 46 patients (7) and increased
serum albumin after 3 months in 12 patients without any change in body weight,
Kt/V, and nPCR (8). Other studies have shown no increase in serum albumin after
correction of acidosis. Complete correction of pre-dialysis metabolic acidosis in
HD patients may lead to post-dialysis metabolic alkalosis and consequently
hypoventilation, phosphate transfer into cells and a higher risk of soft tissue and
vascular calcification. The prerequisite additional oral or dialysate bicarbonate
(and sodium) load may contribute to higher sodium (and fluid) retention and
hypertension. Review of the target pre-dialysis serum bicarbonate levels set by
international clinical practice guidelines indicates that a mild degree of pre-
dialysis acidosis is recommended to minimize the risk of adverse events.
References
1 Mioni R, Gropuzzo M, Messa M et al. Acid production and base balance in patients
in chronic haemodialysis. Cli Sci 201;101:329-37
2 Lowrie EG, Lew NL. Death risk in haemodialysis patients: The predictive value of
commonly measured variables and an evaluation of death rate differences between
facilities. Am J Kidney Dis 1990; 15: 458-482
3 Bommer J, Locatelli F, Satayathum S et al. Association of predialysis serum
bicarbonate levels with risk of mortality and hospitalisation in the Dialysis Outcomes
and Practice Patterns Study (DOPPS). Am J Kidney Dis 2004;44:661-71
4 Graham KA, Reaich D, Channon SM et al. Correction of acidosis in haemodialysis
decreases whole-body protein degradation. J Am Soc Nephrol 1997;8: 632-7
5 Lefebvre A, de Verneioul MC, Gueris J et al. Optimal correction of acidosis changes
progression of dialysis osteodystrophy. Kidney Int 1989;36:1112-8
6 Graham KA, Hoenich NA, Tarbit M et al. Correction of acidosis in haemodialysis
patients increases the sensitivity of the parathyroid glands to calcium. J Am Soc
Nephrol;1997;8:627-31
7 Williams AJ, Dittmer ID, McArley A, Clarke J. High bicarbonate dialysate in
haemodialysis patients: effects on acidosis and nutritional status. Nephrol Dial
Transplant 1997;12:2633-7
8 Movilli E, Zani R, Carli O et al. Correction of metabolic acidosis increases serum
albumin concentrations and decreases kinetically evaluated protein intake in
haemodialysis patients: a prospective study. Nephrol Dial Transplant 1998;13:1719-
22

Pre-dialysis serum potassium should be between 3.5 and 6.5 mmol/l in HD
patients (good practice).
Audit measure 8
Cumulative frequency curves of pre-dialysis serum potassium concentration.
Rationale
The risk of developing hyperkalaemia is inversely related to renal function.
Hyperkalaemia is a common indication for emergency dialysis in patients already
on HD and 3-5% of deaths in dialysis patients have been attributed to
hyperkalaemia (1). Non-compliance with the HD prescription and/or diet is the
main cause of hyperkalaemia in dialysis patients but drug therapy, such as ACE
inhibitors, angiotensin receptor blockers, non-steroidal anti-inflammatory drugs,

beta-blockers and potassium supplements, may be implicated.
HD is the most reliable and immediate treatment of hyperkalaemia in dialysis

patients and the serum potassium will usually fall by 1 mmol/l after the first hour
of HD and a further 1mmol/l after the next 2 hours of HD (2). The rate of
potassium removal may be enhanced by an increase in the dialyser blood flow
rate, a rise in the dialysate bicarbonate concentration or a decrease in the dialysate
potassium concentration (3). A recent review paper (4) has highlighted the benefit
in performing an urgent electrocardiogram to guide management in patients with
serum potassium above 6mmol/l and help decide which patients need emergency
administration of intravenous 10ml 10% calcium chloride over 5 minutes. A
Cochrane meta-analysis of non-dialytic emergency interventions for
hyperkalaemia concluded that intravenous glucose with insulin and nebulised or
inhaled salbutamol were effective in reducing serum potassium levels but the
studies were limited by the absence of data on cardiac arrhythmia or mortality
rates (3). Whilst the combination of salbutamol and intravenous glucose with
insulin was probably more effective than either therapy alone the evidence for
efficacy of intravenous bicarbonate or potassium exchange resins in this Cochrane
review of randomized or quasi-randomised trials was equivocal and neither should
be used as monotherapy for severe hyperkalaemia .
Hypokalaemia towards the end or immediately after HD is not uncommon and

may be corrected by relaxing dietary potassium restriction or, if necessary, by
increasing the dialysate potassium concentration (5, 6).
References
1 Morduchowicz G, Winkler J, Drazne E et al. Causes of death in patients with end-
stage renal disease treated by dialysis in a centre in Israel. Isr J Med Sci 1992;28:776-
9
1 Ahmed J, Weisberg LS. Hyperkalaemia in dialysis patients. Semin Dial 2001;14:348-
56
2 Mahoney BA, Smith WAD, Lo DS, Tsoi K, Tonelli M, Clase CM. Emergency
interventions for hyperkalaemia (Cochrane Review). In: The Renal health Library,
2005. Oxford: Update Software Ltd (www.update-software.com)
3 Alfonzo AVM, Isles C, Geddes C, Deighan C. Potassium disorders - clinical
spectrum and emergency management. Resuscitation 2006;70:10-25
4 Wiegand CF, Davin TD, Raij L, Kjellstard CM. Severe hypokalemia induced by
hemodialysis. Arch Int Med 1981;141: 167-70
5 Bleyer AJ, Hartman J, Brannon PC et al. Characteristics of sudden death in
hemodialysis patients. Kidney Int 2006; 69: 2268-73

Pre-dialysis serum phosphate should be between 1.1 and 1.8mmol/l. (good
practice).

Pre-dialysis serum calcium, adjusted for serum albumin, should be within the
normal range (good practice).

Pre-dialysis serum albumin corrected calcium x phosphate product should be
less than 4.8 mmol2/l2. (good practice).

Serum PTH levels should be more than twice and less than 4 times the upper
limit of normal for the intact PTH assay used.
Serum PTH levels do not need to be performed routinely more often than every 3
months (good practice).
Audit measure 9
Cumulative frequency curves of pre-dialysis serum calcium, phosphate,
calcium x phosphate product and PTH concentrations

Serum aluminium concentration should be measured every three months in
all patients receiving oral aluminium containing phosphate binders (good
practice).

Pre-dialysis haemoglobin concentration should be 10.5-12.5g/dl. (evidence).
The target haemoglobin concentration should be at least 11g/dl to allow for the
normal distribution around the mean haemoglobin value of the patient population
and intraindividual variation of laboratory measurements and hydration status.
Audit measure 10
Cumulative frequency curves of pre-dialysis haemoglobin concentration.
Rationale
The detailed rationale for these guidelines (Guidelines 6.5 - 6.10) is available in
the “Complications of Chronic Kidney Disease” module of the updated RA
guidelines, 2007. Defined ranges of several biochemical variables (Guidelines 6.5
- 6.8) have been associated with better survival rates of HD patients in large
observational studies (1-11). These laboratory indices, which have been associated
with improved patient outcomes in large datasets of hospital HD patients, were
used to develop the audit measures and clinical practice guidelines for thrice
weekly HD within this update. The laboratory based guidelines that are
recommended for thrice weekly HD in this update are consistent with previous
versions of the Renal Association HD guidelines, the UK Renal Registry, Scottish
Renal Registry and Quality Improvement Scotland (QIS) and also with the clinical
practice guidelines for HD that have been generated in Europe, Australasia and
North America. There are no evidence-based guidelines for these laboratory
parameters in patients with end-stage chronic renal failure on other than thrice
weekly HD or in patients with dialysis dependent acute renal failure. The
standards set in this module apply equally to home and hospital HD patients.
Similar audit measures have been used in the preparation of previous UK Renal
Registry Annual Reports (12).
References
6 Lowrie EG, Lew NL. Death risk in haemodialysis patients: The predictive value of
commonly measured variables and an evaluation of death rate differences between
facilities. Am J Kidney Dis 1990; 15: 458-482
7 Iseki K, Uehara H, Nishime K et al. Impact of the initial levels of laboratory variables
on survival in chronic dialysis patients. Am J Kidney Dis 1996; 28:541–548
8 Iseki K, Miyasato F, Tokuyama K et al. Low diastolic blood pressure,
hypoalbuminemia, and risk of death in a cohort of chronic haemodialysis patients.
Kidney Int 1997; 51:1212–1217
9 Owen WF, Lew NL, Lowrie EG et al. The urea reduction ratio and serum albumin
concentration as predictors of mortality in patients undergoing hemodialysis. N Engl
J Med 1993; 329:1001–1006
10 Block GA, Hulbert-Shearon TE, Levin NW et al. Association of serum phosphorus
and calcium x phosphorus product with mortality risk in chronic haemodialysis
patients: a national study. Am J Kidney Dis 1998; 31:607-617
11 Young EW, Albert JM, Satayathum S et al. Predictors and consequences of altered
mineral metabolism: The Dialysis Outcomes and Practice Study. Kidney Int 2005;
67: 1179-1187
12 Block GA, Klassen PS, Lazarus JM et al. Mineral metabolism, mortality and
morbidity in maintenance hemodialysis. J Am Soc Nephrol 2004; 15: 2208-2218
13 Kestenbaum B, Sampson JN, Rudser KD et al. Serum phosphate levels and mortality
risk among people with chronic kidney disease. J Am Soc Nephrol 2005; 16:520-528
14 Noordzij M, Korevaar JC, Boeschoten EW et al. The Kidney Disease Outcomes
Quality Initiative (K/DOQI) guideline for bone metabolism and disease in CKD:
association with mortality in dialysis patients. Am J Kidney Dis 2005; 46:925-932
15 Robinson BM, Joffe MM, Berns JS et al. Anaemia and mortality in haemodialysis
patients: Accounting for morbidity and treatment variables updated over time. Kidney
Int 2005; 68:2323-2330
16 Agarawal R. Hypertension and survival in chronic haemodialysis patients - past
lessons and future opportunities. Kidney Int 2005; 67:1-13
2004. (www.renalreg.com)

Data on the frequency of dialysis-related hypotension, defined as an acute
symptomatic fall in blood pressure during dialysis requiring immediate
intervention to prevent syncope, should be collected and audited (good
practice).
Audit measure 11
The incidence of symptomatic hypotensive episodes during dialysis sessions.
Rationale
Dialysis-related hypotension is the most frequent symptomatic complication of
HD and historically in some studies occurred in more than 15% of HD sessions
(1). As well as being extremely unpleasant hypotensive episodes can shorten the
time on dialysis and reduce the efficiency of delivered dialysis (1). Dialysis-
related hypotension also has been shown to be an independent predictor of poor
patient survival (2). The frequency of this event is, therefore, an important
indicator of the quality of dialysis from the patient’s perspective. It is caused by a
reflex withdrawal of sympathetic tone resulting from decreased left ventricular
filling, and is therefore dependent on the rate of fluid removal from the vascular
space, the rate of re-filling from the interstitial space, venous tone, and many other
variables (3). Patients experiencing frequent dialysis-related hypotension are at
higher risk of death (4) and this may be because dialysis-related hypotension is a
marker for severe cardiac disease (5). Adjustment of the rate of fluid removal,
dialysate sodium concentration and dialysate temperature during dialysis, or
combinations thereof, can reduce the incidence of this complication (6-9).
Interdialysis weight gains can be reduced if dietary sodium intake is kept below
100 mmol/day and thereby reduce thirst and subsequent fluid intake. Dialysate
sodium modeling or ramping can reduce intradialysis cramps and hypotension but
incurs the risk of greater problems with interdialysis thirst, weight gain and
hypertension (7). A recent randomized trial of intradialytic blood volume
monitoring and conventional monitoring showed no difference in weight, blood
pressure or frequency of dialysis-related complications whilst hospitalization and
mortality rates were lower in the group assigned to conventional monitoring (10).
However the conventional monitoring group had atypically low hospitalisation
and mortality rates in comparison with local prevalent HD patients (10). There is
also the question of increased cost if on-line monitoring of changes in relative
blood volume (by measurement of changes in optical density of blood) is used to
assess dry body weight in an attempt to reduce the incidence of intradialytic
hypotension (11). A recent systematic review of 22 studies has concluded that a
reduction in dialysate temperature is effective in decreasing the incidence of
intradialytic hypotension without affecting dialysis adequacy (12). Increasing the
dialysis treatment time to reduce the fluid ultrafiltration rate or decreasing the
dialysate fluid temperature are the most reliable and practical methods of reducing
the incidence of intradialytic hypotension without causing adverse sequelae.
References
1 Ronco C, Brendolan A, Milan M et al. Impact of biofeedback-induced cardiovascular
stability on haemodialysis tolerance and efficiency. Kidney Int 2000; 58:800–808
2 Shoji T, Tsubakihara Y, Fujii M et al. Haemodialysis-associated hypotension as an
independent risk factor for two-year mortality in haemodialysis patients. Kidney Int
2004; 66:1212-1220
3 Daugirdas JT. Dialysis hypotension: a hemodynamic analysis. Kidney Int 1991;
39:233–46
4 Koch M, Thomas B, Tschope W et al. Survival and predictors of death in dialysed
diabetic patients. Diabetologia 1993; 36:1113–1117
5 Poldermans D, Man in ‘t Veld AJ, Rambaldi R. Cardiac evaluation in hypotension-

prone and hypotension-resistant haemodialysis patients. Kidney Int 1999; 56:1905–
1911
6 Dheenan S, Henrich WL. Preventing dialysis hypotension: A comparison of usual
protective maneuvres. Kidney Int 2001; 59:1175–1181
7 Sang GLS, Kovithavongs C, Ulan R, Kjellstrand K. Sodium ramping in hemodialysis:
A study of beneficial and adverse effects. Am J Kidney Dis 1997; 29: 669-677
8 Yu AW, Ing TS, Zabaneh RJ, Daugirdas JT. Effect of dialysate temperature on
central hemodynamics and urea kinetics. Kidney Int 1995; 48: 237-243
9 Maggiore Q, Pizzarelli F, Santoro A et al. The effects of control of thermal balance
on vascular stability in hemodialysis patients: results of the European randomised
clinical trial. Am J Kidney Dis 2002; 40: 280-290
10 Reddan DN, Szczech LA, Hasseblad V et al. Intradialytic blood volume monitoring
in ambulatory haemodialysis patients: a randomised trial. J Am Soc Nephrol 2005;
16:2162-2169
11 Rodriguez HJ, Domenici R, Diroll A, Goykhman I. Assessment of dry weight by
monitoring changes in blood volume during haemodialysis using Crit-Line. Kidney
Int 2005; 68:854-861
12 Selby NM, McIntyre CW. A systematic review of the clinical effects of reducing
dialysate fluid temperature. Nephrol Dial Transplant 2006; 21: 1883-98
7 Vascular access (Guidelines 7.1 - 7.16)

To preserve veins for creation of vascular access venepuncture or insertion of
peripheral venous cannulae should be avoided in the forearm or arm of all
patients with advanced renal failure whenever possible (good practice).
Rationale
To preserve veins for vascular access all healthcare staff and patients with
progressive renal failure should be aware of the need to avoid venepunctures and
insertion of peripheral intravenous catheters in the forearm and elbow, especially
the cephalic veins of the non-dominant arm (1).
References

The preferred mode of vascular access for haemodialysis patients is a native
arteriovenous fistula (good practice).
Audit measure 12
The proportion of prevalent patients on long-term haemodialysis who use an
arterio-venous fistula, arterio-venous graft and tunneled or non-tunneled
central venous catheters as the mode of vascular access.
Rationale
A native arteriovenous fistula (AVF) is the preferred access in the great majority
of HD patients as it produces the highest flows, minimises sepsis and has the
greatest longevity (1-4). The rate of vascular access related infection was 2.5 per
1000 dialysis sessions for patients with native fistulae or grafts, 13.6 per 1000
dialysis sessions for tunneled central venous catheters and 18.4 per 1000 dialysis
sessions with temporary central venous catheters (5). The CHOICE study of the
effect of the type of vascular access on survival among 616 incident patients
showed that the adjusted relative risk of death compared with AVF was 1.2 for an
arteriovenous graft and 1.5 for a central venous catheter (6). Clinical assessment
of the upper limbs prior to access placement has been used successfully to indicate
if Doppler ultrasound is required to select the most appropriate site for access
creation (7). A radio-cephalic and then brachial-cephalic fistula is the preferred
order of access placement whenever possible. Thereafter a transposed brachial-
basilic vein fistula or arterio-venous synthetic graft should be considered before
relying on a central venous catheter for long-term vascular access. In a small
number of patients with severe cardiac dysfunction fistula construction may be
contra-indicated since a high flow AVF can contribute to high output cardiac
failure. An active program of AVF creation in an USA centre using vascular
mapping increased the prevalence of functional AVF from 24% to 44% which was
associated with a significant reduction in hospitalization rates (p< 0.001) (8).
Data from the Dialysis Outcomes and Practice Patterns Study (DOPPS) show that
67% of prevalent patients in the UK have functioning AVFs, compared with the
European average of 80% (1). This study reports on data from the UK, France,
Germany, Italy and Spain (1). Quality Improvement Scotland has recommended
that at least 70% of prevalent chronic HD patients should have a functioning AVF
whilst the National Service Framework Part 1 has indicated that 90% of prevalent
chronic HD patients should have an AVF for vascular access. The National
Dialysis Access Survey showed that only 69.9% of prevalent HD patients in the
UK in 2005 had definitive access for dialysis (65.6% AVF and 4.3%
arteriovenous graft) (9). Rather than specify a minimum proportion of HD patients
who should have a functioning AVF this guideline emphasizes that as many
patients as possible should have a functioning AVF in preference to other forms of
vascular access and the prevalent form of vascular access should be audited in
each unit at least annually.
References
1 Pisoni RL, Young EW, Dykstra DM et al. Vascular access use in Europe and the
United States; Results from the DOPPS. Kidney Int 2002; 61:305–316
2 Woods JD, Port FK. The impact of vascular access for haemodialysis on patient
morbidity and mortality. Nephrol Dial Transplant 1997; 12: 657-659
3 Nassar GM, Ayus JC. Infectious complications of haemodialysis access. Kidney Int
2001; 60:1–13
4 Hirth RA, Turenne MN, Woods JD et al. Predictors of type of vascular access in
haemodialysis patients. JAMA 1996; 276: 1303-1308
5 Stevenson KB, Adcox MJ, Mallea MC et al. Standardized surveillance of
haemodialysis vascular access infections: 18-month experience at an outpatient,
multifacility, haemodialysis centre. Infect Control Hosp Epidemiol 2000; 21: 200-203
6 Astor BC, Eustace JA, Powe NR et al. Type of vascular access and survival among
incident haemodialysis patients: the Choices for Healthy Outcomes in Caring for
ESRD (CHOICE) Study. J Am Soc Nephrol 2005; 16:1449-1455
7 Wells AC, Fernando B, Butler A et al. Selective use of ultrasonographic vascular
mapping in the assessment of patients before haemodialysis access surgery. Br J Surg
2005; 92:1439-1443
8 Ackad A, Simonian GT, Steel K et al. A journey in reversing practice patterns: a
multidisciplinary experience in implementing DOQI guidelines for vascular access.
Nephrol Dial Transplant 2005; 20:1450-1455
2005, Chapter 6. (www.renalreg.com Renal Association Standards & Audit
Subcommittee)

There should be enough dedicated theatre sessions for access surgery to
provide one session per week for every 120 patients on dialysis. With this
level of access surgery provision no patient on dialysis, including those who
present late, should wait more than four weeks for fistula construction (good
practice).
Audit measure 13
The number of dedicated renal failure access surgery sessions per 120
dialysis patients.
Rationale
Vascular access surgery for haemodialysis needs to improve in the UK. The UK
has been lagging behind most countries in Europe with regard to the proportion of
HD patients using natural AVFs. Fewer patients have AVFs for two main reasons.
Firstly, up to 45% of patients starting HD do so as uraemic emergencies where
there has been no time to create permanent access. Secondly, most renal units in
the UK have insufficient access to surgical support including theatre sessions
dedicated to renal failure surgery. One dedicated access surgery session per 120
dialysis patients has been identified as best practice but only 4 of the 10 adult
renal units in Scotland achieved this audit standard of access surgery provision in
2003 (1). This level of access surgery provision should decrease the waiting time
for access surgery, which is particularly important for patients who are either
already on dialysis or predicted to start dialysis within a few months.
References

Patients should undergo fistula creation between 6 and 12 months before
haemodialysis is expected to start to allow time for adequate maturation of
the fistula or time for a revision procedure if the fistula fails or is inadequate
for use (good practice).
Audit measure 14
The dates of first referral to nephrology, referral for creation of vascular
access and creation of first vascular access and the date and mode of vascular
access at the initiation of dialysis should be recorded and audited in all
incident chronic haemodialysis patients.
Rationale
Ideally patients should undergo AVF creation between 6 and 12 months before
HD is expected to start to allow time for adequate maturation of the AVF or time
for a revision procedure if the AVF fails or is inadequate for use (1). The frequent
need for revision surgery is emphasized by a meta-analysis of 8 prospective and
30 retrospective studies of the outcome of radiocephalic AVF creation that
showed an initial failure rate of 15% and primary and secondary patency rates of
radio-cephalic AVF at 1 year of 62% and 66% respectively (2). Patients more than
65 years old were shown to have a relative risk of 1.7 of an AVF failing to mature
compared to patients less than 65 years old (3). The recommendation that an AVF
should be created early in the year before dialysis is anticipated to start is
supported further by a retrospective study of 5924 Canadian HD patients which
showed that the subgroup (n=1240) with fistula creation more than 4 months
before beginning dialysis had a lower risk of sepsis and death, primarily related to
their reduced use of central venous catheters for vascular access (4). Patients with
timely creation of an AVF ready for use at the start of HD were shown to have
better survival than patients who started HD with a catheter and converted to an
AVF who in turn had improved survival rates in comparison to patients who
continued to use a catheter for vascular access (5).
The creation of vascular access ready for use at the initiation of HD in the
majority of patients would be a major step in achieving improved patient
outcomes on dialysis (6). The Dialysis Outcomes and Practice Pattern Study
(DOPPS) has identified that there is wide variation in the time delay among
European countries between referral to a nephrologist and creation of an arterio-
venous fistula (7) and improvements in planning the creation of vascular access
should be a high priority in the UK. The National Dialysis Access Survey
indicated that only 31% of incident HD patients started dialysis with definitive
access and of those known to a renal unit for at least a year only half started HD
with definitive access (8). This survey also indicated that avoidable delay in
referral for vascular access in patients known to a renal unit for at least 6 months
was common; only 33% of this patient group were referred for access creation
more than 6 months before starting dialysis only 48% were referred for vascular
access surgery more than 3 months before starting dialysis (8). The temporal
relationships between first nephrology referral, referral for and creation of
vascular access and mode of functioning access at the time of starting HD should
be audited to promote quality improvements in the planning of vascular access.
References
2 Rooijens PP, Tordoir JH, Stijnen T et al. Radiocephalic wrist arteriovenous fistula for
haemodialysis: meta-analysis indicates a high primary failure rate. Eur J Endovasc
Surg 2004; 28:583-589
3 Lok CE, Oliver MJ, Su J et al. Arteriovenous fistula outcomes in the era of the elderly
dialysis population. Kidney Int 2005; 67:2462-2469
4 Oliver MJ, Rothwell DM, Fung K et al. Late creation of vascular access for
hemodialysis and increased risk of sepsis. J Am Soc Nephrol 2004; 15: 1936-1942
5 Ortega T, Ortega F, Diaz-Corte C et al. The timely construction of arteriovenous
fistulae: a key to reducing morbidity and mortality and to improving cost
management. Nephrol Dial Transplant 2005; 20:598-603
6 Pereira B. Optimisation of pre-ESRD care: The key to improved dialysis outcomes.
Kidney Int 2000; 57:351-365
7 Rayner HC, Pisoni RL, Gillespie BM et al. Creation, cannulation and survival of
arterio-venous fistulae - data from the Dialysis Outcomes and Practice Patterns Study
(DOPPS). Kidney Int 2003; 63:323-330.
Subcommittee)

The time to first cannulation of an AVF should be a minimum of 1 month
and preferably at least 2 months after creation (good practice).
Rationale
First cannulation may be considered between 2 and 4 weeks after creation if this is
the alternative to insertion of a central venous catheter and a nephrologist or
experienced haemodialysis nurse has assessed that the fistula has matured
adequately for use for dialysis.
The time required before an AVF is considered to have adequate blood flow rates
and vessel wall “maturity” to allow safe, repeated cannulation for HD varies in
different countries from less than 2 weeks to more than 3 months (1). A shorter
maturation time would reduce dependence on temporary vascular access but may
reduce AVF survival. The use of a central venous catheter for HD and
nephrology referral within a month before starting dialysis were both predisposing
factors to first cannulation within 28 days after creation of an AVF in a DOPPS
study of first AVF in 642 incident HD patients (2). Both of these DOPPS studies
also showed that maturation times of less than 14 days before first cannulation of
an AVF were associated with lower AVF survival (1,2). An Italian multi-centre
study of first AVF in incident patients confirmed the strong association between
late referral, use of central venous catheters for vascular access and earlier first
cannulation time and showed that shorter maturation times of an AVF before first
cannulation were associated with lower unassisted and assisted patency rates (3).
This contrasts with the findings of the DOPPS report (1) that showed no
difference in fistula failure rates whether the fistula was first cannulated between
15 and 28 days or between 43 and 84 days. In an Italian multicentre study first
cannulation within 1 month of creation was associated with a 94% higher risk of
primary failure (p < 0.001) and within 2 weeks was associated with 111%
increased risk of failure (p < 0.009) (3). These observational studies lend support
to the current recommendation in the K/DOQI guidelines that there should be a
minimum of 1 month and preferably at least 2 months before first utilization of an
AVF (4) except when the fistula has matured adequately before 4 weeks and using
the fistula is the alternative to insertion of a central venous catheter. A recent
commentary on the available evidence has concluded that cannulation of newly
created fistulae within 2 weeks should be avoided, first cannulation between 2 and
4 weeks may be possible in individual cases if the fistula is deemed mature by the
nephrologist or surgeon and it is probably safe to cannulate a fistula after 4 weeks
of creation (5). Conversely a mature access that has required surgical or
radiological intervention to restore adequate blood flow rates may be cannulated
as soon as clinically indicated.
References
1 Rayner HC, Pisoni RL, Gillespie BM et al. Creation, cannulation and survival of
arterio-venous fistulae - data from the Dialysis Outcomes and Practice Patterns Study
(DOPPS). Kidney Int 2003; 63:323-330
United States; Results from the DOPPS. Kidney Int 2002; 61:305–316
3 Ravani P, Brunori G, Mandolfo S et al. Cardiovascular comorbidity and late referral

impact arterio-venous fistula survival: a prospective multicentre study. J Am Soc
Nephrol 2004; 15:204-209
5 Saran R, Pisoni RL, Young EW. Timing of first cannulation of arteriovenous fistula:
are we waiting too long? Nephrol Dial Transplant 2005; 20:688-690

At least 65% of patients presenting more than three months before initiation
of dialysis should start HD with a usable native arteriovenous fistula (good
practice).
Audit measure 14
The dates of first referral to nephrology, referral for creation of vascular
access and creation of first vascular access and the date and mode of vascular
access at the initiation of dialysis should be recorded and audited in all
incident chronic haemodialysis patients.
Rationale
With this audit measure the proportion of incident patients on long-term
haemodialysis who present more than three months before initiation of dialysis
and have permanent vascular access at the start of dialysis can continue to be
assessed.
The time period between fistula creation and first use of a fistula for HD varies
considerably amongst European countries from less than 1 month up to 4 months.
Only 47% of UK patients started dialysis with a functioning AVF, compared with
the European average of 66% in the DOPPS (1). The National Dialysis Access
Survey indicated that only 31% of incident HD patients in the UK in April 2005
started dialysis with definitive access (2). If European practices for vascular
access were extended to the UK at least 65% of patients presenting more than
three months before initiation of dialysis should be able to start HD with a usable
native AVF. Four of the 10 adult renal units in Scotland achieved this standard
when audited in 2003 (3).
References
United States; Results from the DOPPS. Kidney Int 2002; 61:305–316.
Subcommittee)

Investigation of the AVF or graft to assess for evidence of arterial or venous
stenoses or access recirculation is required if there is a significant fall in the
blood flow rate that can be achieved, a reduction in delivered dialysis dose or
a persistent rise in venous pressure in sequential dialysis sessions (good
practice).
Rationale
There is no consensus on the effectiveness of interventions to prolong the use-life
of AVFs and grafts for vascular access (1,2). The rate of AVF thrombosis is 0.2-
0.4 per patient year compared with 0.8-1.2 per patient year for synthetic
arteriovenous grafts (3). Clinical monitoring of the fistula or graft function can
help detect the vascular access at risk. A significant fall in the blood flow rate that
can be achieved, a reduction in delivered dialysis dose or a persistent rise in
venous pressure at the same blood flow rate and using the same gauge of needle in
sequential dialysis sessions should prompt further investigation of the AVF or
graft to assess for evidence of arterial or venous stenoses or access recirculation
(3-5). If there is evidence of greater than 50% stenosis of the fistula percutaneous
angioplasty or surgical revision should be considered to prolong the use-life of the
fistula. Interventions once thrombosis of a fistula has occurred have not shown
good results unless there is no history of fistula dysfunction and the fistula has
become occluded recently. Routine surveillance of fistula blood flow rates has not
yet been shown to enhance the use-life of fistulae and regular access monitoring
may or may not extend the use-life of arteriovenous grafts (6-10). The K/DOQI
guidelines recommend that all patients undergo a program of regular access
monitoring preferably by assessing access flow rates combined with prompt
imaging and elective correction of stenosis in low flow accesses (11). However
recent randomized studies suggest that radiological and/or surgical intervention is
more likely to be clinically effective and cost-effective if assessment of the AVF
or graft for intervention is restricted to patients with the aforementioned clinical
indicators of poor vascular access function (6, 10).
References
1 Sands SS. Vascular access monitoring improves outcomes. Blood Purif 2005; 23:45-
49
2 Paulson WD. Access monitoring does not really improve outcomes. Blood Purif
2005; 23:50-56
3 Woods JD, Turenne MN, Strawderman RL et al. Vascular access survival among
incident haemodialysis patients in the United States. Am J Kidney Dis 1997; 30: 50-
57
4 Sherman RA. The measurement of dialysis access recirculation. Am J Kidney Dis
1993; 22: 616-621
5 Tattersall JE, Farrington K, Raniga PD et al. Haemodialysis recirculation detected by
the three-sample method is an artefact. Nephrol Dial Transplant 1993; 8: 60-63
6 Shahin H, Reddy G, Sharafuddin M et al. Monthly access flow monitoring with
increased prophylactic angioplasty did not improve fistula patency. Kidney Int 2005;
68:2352-2361
7 Malik J, Slavikowa M, Svobodova J, Tuka V. Regular ultrasonagraphic screening
significantly prolongs patency of PTFE grafts. Kidney Int 2005; 67:1554-1558
8 Besarab A, Sullivan KL, Ross RP et al. Utility of intra-access pressure monitoring in

detecting and correcting venous outlet stenoses prior to thrombosis. Kidney Int 1995;
47: 1364-1373
9 Schwab SJ, Raymond JR, Saeed M et al. Prevention of hemodialysis fistula
thrombosis: Early detection of venous stenoses. Kidney Int 1989; 36: 707-711
10 Moist LM, Churchill DN, House AA et al. Regular monitoring of access flow
compared with monitoring of venous pressure fails to improve graft survival. J Am
Soc Nephrol 2003; 14:2645-2653

All patients should be evaluated for a secondary arteriovenous access after
each episode of access failure (good practice).
Rationale
To maximize the use of arteriovenous access for HD the NKF-KDOQI guidelines
on vascular access have recommended that the patient should be assessed fully for
secondary arteriovenous access creation after every episode of AVF or graft loss
(1). This approach has been shown to be successful in subsequent reports (2).
References
2 Asif A, Unger SW, Briones P et al. Creation of secondary arteriovenous fistulas:
maximising fistulas in prevalent hemodialysis patients. Semin Dial 2005;
18:420-424

As few HD patients as possible should rely on central venous catheters for
vascular access.
As an audit measure less than 20% of patients on long-term HD should use

tunneled or non-tunneled central venous catheters as the form of vascular access
(good practice).
Audit measure (2.12 DIAL-HD)

The proportion of prevalent patients on long-term haemodialysis who use an
arterio-venous fistula, arterio-venous graft and tunneled or non-tunneled
central venous catheters as the mode of vascular access.
Rationale
Insertion of a non-cuffed (temporary) or cuffed tunneled (semi-permanent) central
venous catheter is an unfortunate necessity for many patients who need to start
HD before there has been time for creation or maturation of an AVF. Once
established on HD via a catheter some patients may, despite counseling, refuse to
have an AVF constructed. Use of dialysis catheters and PTFE grafts for dialysis is
associated with a greatly increased risk of hospitalisation and sepsis than use of
AVFs (1-4). In a large prospective cohort of incident HD patients the relative risk
of bacteraemia was 1.95 for HD with tunneled catheters and 1.05 for HD with
grafts when compared to patients with an AVF (4). Infection-related
hospitalization in the HEMO study was also shown to be more frequent in patients
relying on central venous catheters for vascular access but was not reduced by the
use of high flux dialysers or a higher dialysis dose (5). Patients with central
venous dialysis catheters and consequent risk of catheter-related infection have
been shown to require higher doses of erythropoietin stimulating agents (ESAs) to
maintain similar or slightly lower mean haemoglobin values (6). Vascular access
using dialysis catheters is also associated with a higher risk of central venous
stenoses and lower blood flow rates. Loss of patency of central venous catheters is
common (7). Each unit should have standardized protocols to attempt
thrombolysis of tunneled central venous catheters using either urokinase or
thromboplastin activator (7). Using a program of vascular access counseling,
vascular mapping, a full range of surgical techniques and salvage procedures the
majority of patients using tunneled dialysis catheters were provided with a
functional arteriovenous access, mainly AVFs (8). The National Dialysis Access
Survey of prevalent HD patients in the UK on 31st March 2005 showed that
27.5% of the 13,260 patients were using non-tunneled central venous catheters for
vascular access and a further 2.0% were using non-tunneled venous catheters (9).
References
1 Chesser AM, Baker LR. Temporary vascular access for first dialysis is common,
undesirable and usually avoidable. Clin Nephrol 1999; 51:228–232
2 Schwab SJ, Beathard G. The haemodialysis catheter conundrum: hate living with
them, but can’t live without them. Kidney Int 1999; 56:1–17
3 Schwab SJ, Harrington JT, Singh A et al. Vascular access for haemodialysis. Kidney
Int 1999; 55:2078–90
4 Ishani A, Collins AJ, Herzog CA, Foley RN. Septicaemia, access and cardiovascular
disease in dialysis patients: the USRDS Wave 2 study. Kidney Int 2005; 68:311-318
5 Allon M. Depner TA, Radeva M, et al. Impact of dialysis dose and membrane on
infection-related hospitalisation and death: Results of the HEMO study. J Am Soc
Nephrol 2003; 14: 1863-1870
6 Roberts TL, Obrador GT, St Peter WL et al. Relationship among catheter insertions,
vascular access infections and anaemia management in haemodialysis patients.
Kidney Int 2004; 66:2429-2436
8 Asif A, Cherla G, Merrill D et al. Conversion of tunneled haemodialysis catheter-
consigned patients to arteriovenous fistula. 2005; 67:2399-2406
Subcommittee)

Cuffed, tunneled double-lumen central venous catheters are preferred if
temporary vascular access is likely to be needed for more than 3 weeks. Non-
cuffed double-lumen catheters should be used if temporary vascular access
for haemodialysis is predicted to be required for less than 3 weeks (good
practice).
Rationale
The incidence of bacteraemia in a prospective study of non-tunneled HD catheters
was 5% after 3 weeks of placement in the internal jugular vein (1). Cuffed,
tunneled rather than non-tunneled central venous catheters are preferred if
vascular access is likely to be required for more than 3 weeks since tunneled
catheters are associated with a lower rate of infections and can provide higher
blood flow rates (2-5). This approach permits immediate vascular access for a
period of months with multiple options of site of catheter insertion.
References
1 Oliver MJ, Callery SM, Thorpe KE et al. Risk of bacteraemia from temporary
hemodialysis catheters by site of insertion and duration of use: A prospective
study. Kidney Int 2000; 58:2543-2545
2 Moss AH, Vasilakis C, Holley JL et al. Use of a silicone dual-lumen catheter
with a Dacron cuff as a long-term vascular access for hemodialysis patients.
Am J Kidney Dis 1990; 16: 211-215
3 Bander SJ, Schab SJ. Central venous angioaccess for hemodialysis and its
complications. Semin Dial 1992; 5:121-128
4 Lee T, Barker J, Allon M. Tunnelled catheters in hemodialysis patients:
reasons and subsequent outcomes. Am J Kidney Dis 2005; 46: 501-508
5 National Kidney Foundation-K/DOQI Clinical Practice Guidelines for
vascular access 2000. Am J Kidney Dis 2000; 37: 1 Supplement 1 S137-S180

The preferred insertion site for central venous catheters is the internal
jugular vein and the catheter should not be placed on the same side as a
planned or maturing upper limb arterio-venous access, whenever possible
(good practice).
Rationale
Central venous catheters should be inserted in the internal jugular vein, preferably
the right internal jugular vein, since this site provides a more direct route to the
superior vena caval-atrial junction than the left side and is more likely to be
contralateral to the non-dominant arm, which is more frequently used for first
attempts at fistula and graft placement (1). The subclavian veins should be
avoided as sites of catheter placement to reduce the risk of compromising the
successful use of an AVF or graft in the ipsilateral arm (2). Nevertheless central
venous stenosis remains relatively common in the era of minimal use of
subclavian venous catheters and was observed on venography in 55 of 133
patients with poor vascular access, 52 of whom had had previous dialysis
catheters (3). Femoral venous catheters may be used for emergency HD without
need for radiological confirmation of catheter position or exclusion of
complications. However femoral dialysis catheters are prone to problems with
patency and dislodgement if used in patients who are not bed bound and non-
cuffed femoral catheters should be removed within 1 week to reduce the risk of
infection (1). The incidence of bacteraemia in a prospective study of non-tunneled
femoral HD catheters was 11% after 1 week of placement (4).
References
2 Barrett N, Spences S, McIvor J, Brown EA. Subclavian stenosis: A major
complication of subclavian dialysis catheters. Nephrol Dial Transplant 1988; 3:
423-425
3 MacRae JM, Ahmed A, Johnson N et al. Central vein stenosis: a common problem in
patients on hemodialysis. ASAIO J 2005; 51:77-81
4 Oliver MJ, Callery SM, Thorpe KE et al. Risk of bacteraemia from temporary
hemodialysis catheters by site of insertion and duration of use: A prospective study.
Kidney Int 2000; 58:2543-2545

All renal units should use real-time ultrasound to guide insertion of central
venous catheters (good practice).
Rationale
Real-time ultrasound is recommended to guide insertion of central venous
catheters to improve the success rate of placement and reduce insertion-related
complications (1). With the use of ultrasound guidance relatively inexperienced
operators can insert internal jugular dialysis catheters reliably and safely (2).
Fluoroscopy screening is mandatory for optimum localization of the catheter tip of
tunneled central venous catheters (1).
References
2 Geddes CC, Walbaum D, Fox JG, Mactier RA. Insertion of internal jugular
temporary haemodialysis cannulae by direct ultrasound guidance - a prospective
comparison of experienced and inexperienced operators. Clin Nephrol 1998; 50:320-
325

All renal units should have protocols to ensure that full barrier precautions
are followed during insertion of temporary and tunneled central venous
dialysis catheters. (Evidence)
Rationale
The risk of infection with central venous catheters can be reduced by using full
barrier precautions during catheter insertion and ensuring that all catheter
connections and disconnections are performed under aseptic conditions by trained
staff (1). Catheter removal is usually indicated in all episodes of bacteraemia
related to temporary central venous dialysis catheters and in episodes of
bacteraemia related to tunneled catheters associated with exit-site or tunnel
infection, persistent fever after commencing antibiotics or metastatic infection (2).
References
1 Raad II, Hohn DC, Gilbreath BJ et al. Prevention of central venous catheter-related
infections using maximal sterile barrier precautions during insertion.
2 Infect Control Hosp Epidemiol 1994; 15: 231-238
3 Allon M. Dialysis catheter-related bacteraemia: treatment and prophylaxis. Am
J Kidney Dis 2004; 45: 779-791 (review)

All central venous catheter connections and disconnections should be
performed under aseptic conditions by trained staff (good practice).
Rationale
Prevention of vascular access-related infection should be a high priority in the
renal unit. Each unit should have strict concordance with infection control
measures and protocols whenever cannulating AVFs and grafts or manipulating
central venous catheters to connect/disconnect to the patient’s bloodstream.
Because of the relatively high risk of catheter-related infection all connections and
disconnections should be performed under aseptic conditions by fully trained staff
wearing a face mask or visor and preferably with the patient wearing a surgical
mask to decrease the risk of infection from nasal carriage of Staphylococcus
aureus (1). Local protocols may incorporate infection control measures that
previous randomized studies have shown can lead to a reduction in catheter-
related infections. These procedures include the use of dry gauze instead of
transparent dressings (2), disinfection with chlorhexidine solutions instead of
povidone-iodine (3), topical mupirocin, Medihoney or antiseptic at the catheter
exit site (4-7) and citrate or antibiotics with heparin as a catheter locking solution
(8-10). The need for effective procedures to prevent catheter-related infections has
been reviewed recently (11).
References
1 Yu VL, Goetz A, Wagener M et al. Staphylococcus aureus carriage rate of patients
receiving long-term haemodialysis. New Engl J Med 1986; 315: 91- 96
2 Conly JM, Grieves K, Peters B. A prospective, randomised study comparing
transparent and dry gauze dressings for central venous catheters. J Infect Dis 1989;
159: 310-319
3 Chaiyakunapruk N, Veenstra DL, Lipsky BA, Saint S. Chlorhexidine compared with
povidone-iodine solution for vascular catheter-site care: a meta-analysis. Ann Intern
Med 2002; 136:792-801
4 Maki DG, Band JD. A comparative study of polyantibiotic and iodophor ointments in
prevention of vascular catheter-related infection. Am J Med 1981; 70: 739-744
5 Levin A, Mason AJ, Jindal KK et al. Prevention of haemodialysis subclavian vein
catheter infections by topical povidone-iodine. Kidney Int 1991; 40: 934-938
6 Maki DG, Ringer M, Alvarado CJ. Prospective randomised trial of povidone-iodine,
alcohol and chlorhexidine for prevention of infection associated with central venous
and arterial catheters. Lancet 1991; 338: 339-343
7 Sesso R, Barbosa D, Leme IL et al. Staphylococcus aureus prophylaxis in
haemodialysis patients using central venous catheter: effect of mupirocin ointment. J
Am Soc Nephrol 1998; 9:1085-1092
8 Weijmer MC, van den Dorpel MA, Van de Ven PJ et al. Randomised clinical trial
comparison of trisodium citrate 30% and heparin as catheter-locking solution in
hemodialysis patients. J Am Soc Nephrol 2005; 16: 2769-2777
9 Johnson DW, van Eps C, Mudge DW et al. Randomised, controlled trial of topical
exit-site application of honey (Medihoney) versus mupirocin for prevention of
catheter-associated infection in haemodialysis patients. J Am Soc Nephrol 2005; 16:
1456-1462
10 Dogra GK, Herson H, Hutchison B et al. Prevention of tunneled haemodialysis
catheter-related infections using catheter-restricted filling with gentamicin and citrate:
a randomised controlled study. J Am Soc Nephrol 2002; 13:2133-2139
11 McIntyre CW, Hulme IJ, Taal M, Fluck RJ. Locking of tunneled haemodialysis
catheters with gentamicin and heparin. Kidney Int 2004; 66:801-85
12 Jaber BL. Bacterial infections in haemodialysis patients: Pathogenesis and
prevention. Kidney Int 2005; 67:2508-2519

Peripheral and central line blood cultures should be taken prior to starting
antibiotics in all cases of suspected catheter-related infection (good practice).
Rationale
Multiple sets of blood cultures taken from both the central venous catheter and
peripheral veins increase the diagnostic yield in patients with catheter-related
bacteraemia. As well as helping treatment by identifying the causative
organism(s) and antimicrobial sensitivities the higher positive culture rate
facilitates microbiological surveillance within the renal unit and hospital,
especially the incidence of antibiotic resistant organisms such as MRSA (1,2).
The National Dialysis Access Survey indicated that the incidence of
Staphylococcus aureus and MRSA bacteraemia during 2004 averaged 13 and 4
episodes per 100 HD patients per year, respectively (3).
References
1 Allon M, Radeva M, Bailey J et al HEMO Study Group. The spectrum of infection-
related morbidity in hospitalised haemodialysis patients. Nephrol Dial Transpant
2005; 20: 1180-1186
Subcommittee)

All HD units should collect and audit data on the form of vascular access in
use in incident and prevalent haemodialysis patients and the rates of infection
per 1000 patient days using central venous catheters, arterio-venous grafts
and fistulae (good practice).
Audit measure 15
The rates of bacteraemia (and specifically the rates of MRSA bacteraemia)
observed per 1000 patient days using central venous catheters,
polytetrafluoroethylene (PTFE) grafts and arterio-venous fistulae.
Rationale
Comparative audit of the different forms of vascular access that are used for HD
and the incidence of access related infections should help identify to what extent
the above guidelines have been achieved and promote good clinical practice.
8 Access to and withdrawal from dialysis

(Guidelines DIAL-HD 8.1 - 8.5)

All patients with advanced renal failure (eGFR <30ml/min), who have a life
expectancy of more than 3 months, should be considered for renal
replacement therapy and should be referred to a nephrologist (good
practice).
Rationale
Estimated GFR (eGFR) is used to report measurements of renal function
whenever the GFR is below 90 ml/min (England & Wales) or below 60ml/min
(Scotland). The MDRD equation based on age, sex, race and serum creatinine is
the preferred and most practical method of estimating GFR in advanced renal
failure although the mean of 24 hour urinary urea and creatinine clearances is
utilized in assessing the adequacy of peritoneal dialysis and estimating residual
renal function in HD patients. The routine reporting of eGFR should promote
universal access for consideration of RRT. This approach helps to identify patients
with significant chronic kidney disease and promote timely referral to a
nephrologist. Avoiding late referral provides the opportunity for intervention to
prevent or reduce the complications of renal failure and time to plan for renal
replacement therapy. Patients who have been under nephrology care for more than
1 month are more likely to start HD using an AVF (1). A retrospective analysis of
109,321 incident HD patients in the USA found that the relative risk of death of
patients with no pre-dialysis nephrology care was 1.51 and the relative risk of
death of patients with one or two months pre-dialysis nephrology care was 1.23
when compared with patients with at least 3 months nephrology pre-dialysis care
(2).
Now that every patient with advanced chronic renal failure, regardless of age and
co-morbidity, is at least considered as a potential recipient of dialysis, questions of
whether or not to start dialysis have assumed increasing importance. Until
recently, acceptance or non-acceptance for dialysis in the UK was predetermined
by accidental and occasionally deliberate failure of referral so that the decision not
to initiate renal replacement therapy was taken by family members or referring
physicians alone, rather than in conjunction with a nephrologist (3). It is often
difficult to decide if patients with major comorbidity will or will not benefit from
starting dialysis, even if referred well in advance of needing renal replacement
therapy, and there have been few studies of the decision not to start dialysis (4).
Nevertheless the problems of late referral or non-referral can be avoided if all
patients who have advanced chronic renal failure (eGFR <30ml/min) and a
predicted life expectancy of at least 3 months are considered for RRT and are
referred to a nephrologist.
References
1 Rayner HC, Besarab A, Brown WW et al. Vascular access results from the Dialysis
Outomes and Practice Patterns Study (DOPPS): performance against Kidney Disease
Outcomes Quality Initiative (K/DOQI) Clinical Practice Guidelines. Am J Kidney
Dis 2004; 44 (Suppl 3): 22-26
2 Khan SS, Xue JL, Kazmi WH et al. Does pre-dialysis nephrology care influence
patient survival after initiation of dialysis? Kidney Int 2005; 67:1038-1046
3 Challah S, Wing AJ, Bauer R et al. Negative selection of patients for dialysis in the
United Kingdom. Br Med J 1984; 288:1119–1122
4 Hirsch DJ, West ML, Cohen AD, Jindal KK. Experience with not offering dialysis to
patients with a poor prognosis. Am J Kidney Dis 1994; 23:463–466

If there is no medical contraindication the choice of initial dialysis modality
should be based on patient choice (good practice).
Rationale
The provision of patient choice and equity of access to dialysis and transplantation
have been reinforced by the National Service Framework Part 1 Dialysis and
Transplantation (1). There has been only one small prospective randomized trial
comparing HD and peritoneal dialysis in incident patients and this showed no
differences in short-term patient outcomes in the small numbers of patients that
could be enrolled into the study but the study data were not powered adequately to
reach any other conclusion (2). Local access to hospital HD should not be an
influential factor in the patient reaching a decision about their preferred initial
mode of dialysis. In the absence of evidence that either HD or peritoneal dialysis
provide superior patient outcomes the selection of initial dialysis modality should
be based on the patient’s choice after full education about the different forms of
renal replacement therapy that are available, including home HD and live donor
and cadaveric transplantation (3).
References
2 Korevaar JC, Feith GW, Dekker FW et al. Effects of starting with haemodialysis
compared with peritoneal dialysis in patients new on dialysis treatment: A
randomised controlled trial. Kidney Int 2003; 64 :2222-2228
3 National Institute of Clinical Excellence. Full guidance on home compared with
hospital haemodialysis for patients with end-stage renal failure October 2002
(www.nice.org.uk)

After full education and counseling a small proportion of patients may opt
for active non-dialytic management of advanced chronic kidney disease,
including nutritional, medical and psychological support rather than plan to
initiate dialysis. The numbers of patients not taken on to dialysis and the
reasons for this decision should be subject to audit (good practice).
Audit measure 16
The proportion of patients with advanced renal failure (CKD stage 5) who
are treated with conservative medical therapy.
Rationale
The decision whether to start or not to start RRT may be difficult (1). It is
impossible to set quantitative standards in this difficult area of care, but principles
of action can be enunciated and agreed. All patients who are found to have
advanced renal failure should be considered for dialysis, and the patient’s age,
social circumstances or required level of community support should not be a
factor leading to exclusion. Nor should lack of facilities for dialysis be acceptable
on its own as grounds for exclusion, or fear of litigation a basis for a decision in
either direction. Careful medical assessment of any co-morbid conditions from
which the patient may suffer is needed, together with whatever medical measures
(short of dialysis) are required to correct them or minimise their effects (2).
Particular attention needs to be paid to potentially reversible mental states.
Similarly, patients who have deteriorated will need careful medical and
psychological assessment. If it appears that only a brief period of survival of
unacceptable quality is likely on dialysis (eg less than three months), then the
possibility of not starting or stopping dialysis needs to be considered. The interest
of the individual patient must remain paramount, and although the opinions of
relatives should be consulted, they should not be binding. The responsible
consultant nephrologist should solicit views of the patient’s family doctor, next of
kin, and all carers within the multidisciplinary caring team. The decision to start
or not to start RRT must be taken by both the consultant (who must assess the
patient personally), and the patient. The patient will need to be fully informed
throughout, and to be aware of their options, so far as their mental status permits.
The most realistic and accurate description of starting or not starting, continuing
or not continuing dialysis should be given. The substance of these discussions
must be recorded in the patient’s notes. If the decision is taken not to initiate, or to
stop dialysis, then a management plan of supportive care must be put in place.
This must then be carried through in a way that ensures continued support,
achieves what seems best from the patient’s point of view, and finally enables the
patient to die with dignity, when the time comes. Achieving this will often require
co-ordinated work with the palliative care team, who should be involved early in
the management plan (3). Some patients who are severely ill, often with
conditions affecting several organs, may have a concurrent acute deterioration of
their chronic renal failure. The referring physician (who may be in a different
hospital) and the nephrologist, may feel, after discussion, that dialysis is
inappropriate given the very poor prognosis from the underlying conditions.
Under these circumstances the referring physician would discuss matters with the
patient, if possible, and with the family. Guidelines on shared decision-making in
the initiation or withdrawal of dialysis have been developed (4).
Two approaches may be taken when a patient presents in uraemia whose ability to
cope with, and to enjoy and benefit from dialysis treatment is doubtful. The first
approach attempts to make a ‘clean’ decision on whether or not to start dialysis
after a process of consultation and discussion; the second, often called ‘trial of
dialysis’, involves starting a proportion of such patients on dialysis, but with a
pre-discussed plan to review whether this should continue beyond a specified
point in the near future – usually a few weeks or months. Clearly the expectation
is that the outcome in this case will be withdrawal of some patients from dialysis.
References
1 Tobe SW, Senn JS (for the End-Stage Renal Disease Group). Foregoing renal
dialysis: case study and review of ethical issues. Am J Kidney Dis 1996; 28:147–153
2 Campbell ML. Terminal care of ESRD patients forgoing dialysis. ANNA J 1991;
18:202–204
3 Cohen LM, Germian M, Poppel DM et al. Dialysis discontinuation and palliative
care. Am J Kidney Dis 2000; 36:140–144
4 Galla JH. Clinical practice guideline on shared decision-making in the appropriate
initiation of and withdrawal from dialysis. J Am Soc Nephrol 2000; 11:1340-1342

Renal replacement therapy should be commenced when a patient with an
eGFR < 15ml/min/1.73m2 has symptoms or signs of uraemia, fluid overload
or malnutrition in spite of medical therapy and be considered carefully when
an asymptomatic patient has an eGFR < 6ml/min/1.73m2 (good practice).
Audit measure 17
Record of the serum creatinine, the estimated GFR and co-morbidity at
initiation of chronic renal replacement therapy (dialysis or transplantation).
Rationale
There are no criteria based on definitive evidence to advise when to start dialysis.
In the absence of severe hyperkalaemia or pericarditis there is no definitive
evidence to indicate when an asymptomatic patient with advanced renal failure
should initiate dialysis. There is consensus that patients should start dialysis when
they develop symptoms or signs of fluid overload, hypertension, poor nutrition or

uraemia which cannot be controlled by medical therapy such as high dose
diuretics, even if their estimated residual renal function is relatively high.
Nutritional status and dietary protein intake decrease progressively as renal
function declines (1). The medical treatment of the complications of renal failure
such as anaemia has improved in the past 10 years and this may explain recent
reports of a lack of any relationship between the presence or absence of traditional
symptoms of uraemia and residual renal function in patients with stage 5 chronic
kidney disease (2). The patients with a higher haemoglobin concentration had
fewer symptoms (2) and so relying on the onset of symptoms may result in
patients starting dialysis too late. Conversely studies in the Netherlands and
Scotland comparing patients who started dialysis at two different levels of residual
renal function have shown no advantage to patient survival if adjustments are
made for lead time bias in the group of patients starting dialysis with higher
residual renal function (3-6). In the multicentre prospective Netherlands study 94
of the 253 incident patients began dialysis later than recommended in the US NKF
KDOQI guideline and the adjusted benefit in survival after 3 years on dialysis was
2.5 months in the timely starter group (4). However this benefit may be attributed
to lead-time bias since the average delay in initiation of dialysis in the late starter
group was 4.1 months. A randomized prospective study to compare 3 year
morbidity and mortality after initiating dialysis when patients have a Cockcroft
and Gault creatinine clearance of 10-14ml/min/1.73m2 or 5-7ml/min/1.73m2 is
underway (IDEAL study) (7).
With the evidence that nutritional status deteriorates progressively as renal

function declines (1) and symptoms of advanced renal failure are not closely
related to the degree of residual renal function in the modern era (2) it is
appropriate that international guidelines have attempted to identify the level of
residual renal function at which an asymptomatic patient should initiate dialysis.
The above considerations fit well with the European Best Practice Guidelines
which recommended that renal replacement therapy should commence when a
patient with an eGFR < 15ml/min/1.73m2 has symptoms or signs of uraemia,
fluid overload or malnutrition in spite of medical therapy or before an
asymptomatic patient has an eGFR < 6ml/min/1.73m2 (8).
References
1 Ikizler TA, Greene JH, Wingarde RL et al. Spontaneous dietary protein intake during
progression of chronic renal failure. J Am Soc Nephrol 1995; 6: 1386-1391
2 Curtis BM, Barrett BJ, Jindal K et al. Canadian survey of clinical status at dialysis
initiation 1998-1999: a multicentre prospective study. Clin Nephrol 2002; 58:282-288
3 Traynor JP, Simpson K, Geddes CC et al. Early initiation of dialysis fails to prolong
survival in patients in end-stage renal failure. J Am Soc Nephrol 2002; 13:2125-2132
4 Korevaar JC, Jansen MA, Dekker FW et al. When to initiate dialysis: effect of
proposed US guidelines on survival. Lancet 2001; 358:1046-1050
5 Korevaar JC, Dekker FW, Krediet RT. Initiation of dialysis: is the problem solved by
NECOSAD? Nephrol Dial Transplant 2003; 18: 1228-1229
6 Termorshuizen F, Korevaar JC, Dekker FW et al. Time trends in initiation and dose
of dialysis in end-stage renal disease patients in The Netherlands. Nephrol Dial
Transplant 2003; 18: 552-558
7 Cooper BA, Branley P, Bulfone L et al. The Initiating Dialysis Early and Late
(IDEAL) study: study rationale and design. Perit Dial Int 2004; 24: 176-181

Any decision to discontinue haemodialysis should be made jointly by the
patient and the responsible consultant nephrologist after consultation with
relatives, the family practitioner and members of the caring team (good
practice).
Rationale
In addition to patients who clearly present greater than average problems from the
outset, there are individuals who have had a period of worthwhile life on dialysis,
but whose quality of life worsens because of medical or psychological
deterioration, or both simultaneously. Additional difficulty arises when dementia,
often fluctuating, or irrecoverable neurological deficit after a cerebrovascular
event makes it difficult or impossible to ascertain what the patient’s own feelings
and wishes might be (1). In practice, the decision to withdraw dialysis has much in
common with decision not to start a patient on dialysis. This is because caring
staff, patients and relatives all face similar difficult judgements and decisions
about the likely quality and quantity of life on dialysis. A similar process to that
outlined in deciding whether or not to plan to start dialysis (see above) should be
followed when assessing if withdrawal of dialysis is appropriate. There is one
study from the UK that suggests that withdrawal from dialysis plays a major role
(17%) in overall death rates on dialysis (2), as it does in the USA and Canada
(3,4). Recent data from the Dialysis Outcomes and Practice Patterns Study have
shown that the rate for withdrawal from HD is 3.5 per 100 patient-years and that
not surprisingly “do not resuscitate” orders are associated with older age and
nursing home residence (5). In a recent UK study withdrawal of dialysis was the
commonest cause of death (38%) in the group of patients commencing dialysis
when more than 75 years old (6). Withdrawal of dialysis is an increasing cause of
death in dialysis patients and the date of the decision and the reasons for it should
be recorded in the patient’s casenotes (7). Renal units should develop guidelines
for withdrawal of dialysis that include liaison with palliative care and community
services.
References
1 Singer J, Thiel EC, Naylor D et al. Life-sustaining treatment preferences of
hemodialysis patients: implications for advance directives. J Am Soc Nephrol 1995;
6:1410–1417
2 Catalano C, Goodship THJ, Graham KA et al. Withdrawal of renal replacement
therapy in Newcastle upon Tyne: 1964-1993. Nephrol Dial Transplant 1996; 11:133–
139
3 Cohen LM, McCue JD, Germain M, Kjellstrand CJ. Dialysis discontinued: a good
death? Arch Intern Med 1995; 155:42–47
4 Friedman EA. The best and worst times for dialysis are now. ASAIO J 1994; 40:107–
108
5 Fissell RB, Bragg-Gresham JL, Lopez AA et al. Factors associated with "do not
resuscitate" orders and rates of withdrawal from haemodialysis in the international
DOPPS. Kidney Int; 2005 68: 1282-1288
6 Munshi SK, Vijayakumar N, Taub NA et al. Outcome of renal replacement therapy in
the very elderly. Nephrol Dial Transplant 2001; 16:1721-1722
7 McLean AM. Dialysis treatment withdrawal – Legal aspects (UK). Nephrol Dial
Transplant 1998; 13:1152-1153
Acknowledgements and declarations of

interest
Several members and affiliates of the Association of Renal Technologists
helped prepare the standards and rationale of sections 2 and 3 and their
assistance and expertise are acknowledged gratefully. In particular Mr. Gerard
Dean, Principal Clinical Engineer, Nottingham University Hospitals Trust, Mr.
Bill Fisken, senior renal technical officer, NHS Greater Glasgow & Clyde, Dr.
Nicholas Hoenich, School of Clinical Medical Sciences, University of
Newcastle, Dr. Elizabeth Lindley, Leeds Teaching Hospitals NHS Trust and
Mr. Andy Mosson, Chairman, Association of Renal Technologists, checked
that the standards for haemodialysis equipment, dialysate concentrates and
treated water stated in this module are correct and still current. Dr. Mark
MacGregor, consultant nephrologist, Crosshouse Hospital, Ayrshire provided
constructive and detailed comments on the first draft of the guidelines which
was circulated to the RA Clinical Practice Guidelines Committee, RA
Executive and Clinical Directors Forum.
Dr. Robert Mactier wishes to acknowledge and declare the following potential
conflicts of interest: study investigator for multicentre research studies
conducted by Roche and Baxter, member of the clinical advisory board for
Baxter in 2005, receipt of sponsorship to attend scientific meetings from
Roche and Baxter. To his knowledge, has had no other direct support from the
renal technology industry.

Module 3a - Haemodialysis - 4th Edition PDF

Uploaded by

Copyright:

Available Formats

Module 3a - Haemodialysis - 4th Edition PDF

Uploaded by

Document Information

Original Title

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

Module 3a - Haemodialysis - 4th Edition PDF

Uploaded by

Copyright:

Available Formats

CLINICAL PRACTICE GUIDELINES

Dr. Robert Mactier,

SUMMARY OF CLINICAL PRACTICE GUIDELINES FOR HAEMODIALYSIS

2 Haemodialysis equipment and disposables (Guidelines DIAL-HD 2.1 – 2.3) 6

3 Concentrates and water for haemodialysis (Guidelines DIAL-HD 3.1 – 3.4) 6

4 Haemodialysis membranes (Guidelines DIAL-HD 4.1 – 4.6) 7

5 Haemodialysis dose, frequency and duration (Guidelines DIAL-HD 5.1 – 5.7) 8

7 Vascular access (Guidelines DIAL-HD 7.1 – 7.16) 11

8 Access to and withdrawal from dialysis (Guidelines DIAL-HD 8.1 – 8.5) 13

SUMMARY OF AUDIT MEASURES FOR HAEMODIALYSIS 14

FULL CLINICAL PRACTICE GUIDELINES FOR HAEMODIALYSIS 16

2 Haemodialysis equipment and disposables (Guidelines DIAL-HD 2.1 – 2.3) 23

3 Concentrates and water for haemodialysis (Guidelines DIAL-HD 3.1 – 3.4) 25

4 Haemodialysis membranes (Guidelines DIAL-HD 4.1 - 4.6) 34

5 Haemodialysis dose, frequency and duration (Guidelines DIAL-HD 5.1 – 5.7) 42

7 Vascular access (Guidelines 7.1 - 7.16) 63

8 Access to and withdrawal from dialysis (Guidelines DIAL-HD 8.1 - 8.5) 76

ACKNOWLEDGEMENTS AND DECLARATIONS OF INTEREST 83

Innovations and changes in HD practice have seldom been underpinned by

This module provides an expansion on the 2002 guidelines in HD to

3 Renal Association Standards & Audit Subcommittee "Treatment of adults &

Summary of clinical practice guidelines

1 Haemodialysis facilities (Guidelines DIAL-

Guideline 1.1 DIAL-HD

Guideline 1.2 DIAL-HD

Guideline 1.3 DIAL-HD

Guideline 1.4 DIAL-HD

Guideline 1.5 DIAL-HD

Guideline 1.6 DIAL-HD

2 Haemodialysis equipment and disposables

Guideline 2.1 DIAL-HD

Guideline 2.2 DIAL-HD

Guideline 2.3 DIAL-HD

3 Concentrates and water for haemodialysis

Guideline 3.1 DIAL-HD

Guideline 3.2 DIAL-HD

Guideline 3.3 DIAL-HD

Guideline 3.4 DIAL-HD

4 Haemodialysis membranes (Guidelines

Guideline 4.1 DIAL-HD

Guideline 4.2 DIAL-HD

Guideline 4.4 DIAL-HD

Guideline 4.5 DIAL-HD

Guideline 4.6 DIAL-HD

5 Haemodialysis dose, frequency and

Guideline 5.1 DIAL-HD

Guideline 5.2 DIAL-HD

Guideline 5.3 DIAL-HD

Guideline 5.4 DIAL-HD

Guideline 5.5 DIAL-HD

Guideline 5.6 DIAL-HD

Guideline 5.7 DIAL-HD

6 Laboratory and clinical indices of dialysis

Guideline 6.1 DIAL-HD

Guideline 6.2 DIAL-HD

Guideline 6.3 DIAL-HD

Guideline 6.4 DIAL-HD

Guideline 6.5 DIAL-HD

Guideline 6.6 DIAL-HD