ISSN: 2320-5407 Int. J. Adv. Res.

11(01), 1017-1022

Journal Homepage: -www.journalijar.com

Article DOI:10.21474/IJAR01/16113
DOI URL: http://dx.doi.org/10.21474/IJAR01/16113

RESEARCH ARTICLE
QUALITY CONTROL MANAGEMENT ISSUES OF ROUTINE COAGULATION TEST IN A LOW-
INCOME COUNTRY (MADAGASCAR)

S. Niry Manantsoa1, J. Andriamampihonona2, F.H Rakotonjafiniarivo1, A. Rasamoelina3, T.M Raheritiana1,

A.L Rakotovao1, JDLC Rasolonjatovo2 and A.O Rakoto Alson1
1. Department of Medical biology, Faculty of Medicine, University of Antananarivo, Madagascar.
2. Faculty of Medicine, University of Toamasina, Madagascar.
3. Faculty of Sciences, University of Antananarivo, Madagascar.
……………………………………………………………………………………………………....
Manuscript Info Abstract
……………………. ………………………………………………………………
Manuscript History Introduction: Reliability of biological assays’ results depends closely
Received: 30 November 2022 on the quality of the 3 biological stages: pre-analytical, analytical, and
Final Accepted: 31 December 2022 post-analytical. In each stage, a quality program including internal
Published: January 2023
quality control (IQC) and external quality assurance (EQA) has to be
assessed to ensure test result accuracy and precision. The coagulation
Key words:-
Quality Control, Hemostasis, Internal laboratory plays an important role in the diagnosis and treatment of
Quality Control, External Quality people with blood disorders, so though several lack of facilities in
Assurance, Madagascar Malagasy labs, QC has to be implemented. Thus, this study aimed to
analyze IQC and EQA in hemostasis tests in the Hematology laboratory
of the CHU-HJRA in Antananarivo Madagascar in order to improve its
analytical phase’s quality.
Methods: This retrospective descriptive and analytical study was
conducted from January 1st, to October 31st, 2021 in the Hematology
laboratory of CHU-JRA within the routine coagulation tests
(Prothrombin time PT and Activated partial prothrombin time aPTT).
Regarding the EQA, percentage of deviation from the peer group’s
results were calculated. Then in order to analyze IQC’s results,
coefficient of variation and biases of IQC’s results, corresponding
respectively to precision and accuracy, were calculated and Levey-
Jennings graphs were analyzed.
Results: Regarding global results of EQA, 50% were "within
consensus" compared to peer-group. And regarding IQC, for precision,
one coefficient of variation for both PT and aPTT exceeded 5%, and for
accuracy, no bias exceeds 10% deviation from the expected result.
Levey-Jennings’s plot showed more than 90% of the results to be within
M±2 SD range. A violation 10x Westgard rules, corresponding to a
systematic error was observed.
Conclusion: Each out-of-control results has to be analyzed as part of
strengthening quality program guaranteeing coagulation tests’
reliability.

Copy Right, IJAR, 2023,. All rights reserved.

……………………………………………………………………………………………………....

Corresponding Author:- S. Niry Manantsoa

Address:- Department of Medical biology, Faculty of Medicine, University of 1017
Antananarivo, Madagascar.
ISSN: 2320-5407 Int. J. Adv. Res. 11(01), 1017-1022

Introduction:-
Coagulation laboratory plays an important role in the diagnosis and treatment of patients with or without blood
disorders such as anti-thrombotic therapies monitoring, or biological preoperative assessment. According to the
World Health Organization, "hundreds of thousands of deaths or serious illnesses worldwide are attributable each
year to inaccuracies or errors in medical laboratories [1]. It is therefore essential to perform a laboratory quality
program in order to get coagulation tests’ results reliability [2].

« Quality » can be simply represented by 2 statements « Do the thing right » and « Do the right thing » [3].
Regarding laboratory QC program, it can be defined as the accuracy, reliability and relevance of test results and it
includes Internal Quality Control (IQC) and External Quality Evaluation (EQA) [3]. In case of out-control results,
corrective actions should be implemented [3].

Thus, this study aims to analyze IQC and EQA in hemostasis tests in the Hematology laboratory of the CHU-HJRA
in Antananarivo Madagascar in order to improve its analytical phase’s quality.

Methods:-
This is a 10-months retrospective cross-sectional and analytical study, carried out from January 1 st to October 31st,
2021 within the QC program in the Hematology Laboratory CHU-JRA in Antananarivo Madagascar.

Regarding EQA, the QC program is provided by the UK-based NEQUAS (National External Quality Assurance
Scheme). QC materials are received 5 times a year, and consisted on lyophilized plasma from real patients with
known values so results. In, this study, 4 surveys were included. Analytical performance of the laboratory about PT
and aPTT’s results in each EQA survey were analyzed by calculating the percentage deviation of each test from the
average of peer-group by formula seen in Table I. The result of the test was "within consensus" if the percent
deviation was less than 15%, and "out of consensus" if the percent deviation was greater than 15%.

Regarding IQC, materials consisted on a pool of daily fresh plasma from 10 different samples received in the
laboratory which tests’ results are normal for PT, aPTT and Fibrinogen assays. Precision and accuracy of the PT an
aPTT results were analyzed. IQC is performed daily before patient’s samples’ run. When a value is out of control,
corresponding corrective action is set up. So IQC values represented in this study are from the first run, before any
corrective measure. Precision’s results were determined by monthly calculation of mean, standard deviation and
coefficient of variation of each test’s results. Accuracy’s results were determined by calculating the bias or
inaccuracy degree comparing to the expected result of each test (Table I).

Precision and accuracy are considered acceptable when PT and aPTT’s results don’t exceed 5% for coefficient of
variation, and no bias exceeds 10% deviation from the expected result. Levey-Jenning’s graph were analyzed using
the six main Westgard rules, stating that the assay series is accepted if the results of both batches are within the
acceptable limits of M ± 2SD; and rejected (or placed on watch) if the following rules are violated: 1 2s; 1₃ ѕ, 2₂ ѕ,
R4s, 4₁ ѕ, 10x (Table II).

Table I:- Formulas for calculating EQA and IQC parameters.

Parameters Formula
EQA
Percentage of deviation (lab result – average of peer −group )
%Dev= ∗ 100
average of peer −group
IQC
Mean (𝑥) ∑ 𝑥𝑖
𝑥= 𝑥 = 𝑚𝑜𝑛𝑡ℎ𝑙𝑦 𝑚𝑒𝑎𝑛 𝑜𝑓 𝐼𝑄𝐶
𝑛
𝑥𝑖 = 𝑑𝑎𝑖𝑙𝑦 𝑟𝑒𝑠𝑢𝑙𝑡𝑠 𝑜𝑓 𝐼𝑄𝐶
𝑛 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐼𝑄𝐶 𝑟𝑒𝑠𝑢𝑙𝑡𝑠
Standard deviation (𝑆𝐷) √(𝑥𝑖 − 𝑥)2
𝑆𝐷 =
(𝑛 − 1)
Coefficient of variation (𝐶𝑉) 𝐶𝑉 = 100 ∗ 𝑆𝐷/𝑥
Bias (%) 𝑥−𝑣
𝐵𝑖𝑎𝑠 = ∗ 100 𝑣 = 𝑒𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑣𝑎𝑙𝑢𝑒
𝑣

1018
ISSN: 2320-5407 Int. J. Adv. Res. 11(01), 1017-1022

Coagulation tests are performed with a STar-T4 semi-automat from StagoDiagnostica® which measures the clotting
time by a chronometric method. Reagents used for PT and aPTT are both derived from rabbit cerebral tissue and
stored at +2 to +8°C. These are BIO-TP from BIOLABO® (Lot Number: 112028A, Expiry date: 11/2023) for PT;
BIO-CK from BIOLABO® (Lot Number: 032042A, Expiry date: 04/2023) and Calcium chloride (Lot Number:
052157A1, Expiry date: 06/2024) for aPTT.

Table II:- Westgard rules.

Westgard rules Details Action
12s One control measurement exceeding 2 standard Warning rule to trigger careful
deviations of control limits either above or below inspection of the control data
the mean
1₃ ѕ Control limits are set as the mean +3 SD of control Run rejected
limits
2₂ ѕ Control measurements 2 SD of control limits on Run rejected
the same side of mean
R4s 2 control measurements in a group exceed the Run rejected
mean with a 4 SD difference between the 2
controls
4₁ ѕ 4th consecutive control measurement exceeding 1 Run rejected
SD on the same side of the mean
10x 10 consecutive controls on the same side of the Run rejected
mean
Source: Hanes H. Westgard rules guidelines. John Hopkins University; April 2020

Results:-
Results of the 4 surveys of EQA are resumed in Table III, 50% were "within consensus" compared to peer-group.

Table III:- EQA results.

S50 S51 S53 S54
Survey
PT aPTT PT aPTT PT aPTT PT aPTT
ID W20:08 W20:09 W20:13 W20:14 W20:08 W21:09 W21:13 W21:14
Lab results 32,5% 1,9 35% 1,13 84% 1,4 93,5% 0,81
%Dev 11% 0,5% 2% 21% 12,69% 26,7% 20,1% 39,1%
Perf. WIC WIC WIC OC WIC OC OC OC
Legends:ID: Sample identity; Perf: Performance; WIC: within consensus; OC: out of consensus

In term of IQC, precision’s analysis showed one coefficient of variation for both PT and aPTT exceeding 5% in
April 2021, and accuracies showed no bias exceeding 10% deviation from the expected result (Table IV).

Table IV:- Precision and accuracy analysis.

Precision
J F M A M J J A S O
CV (%) PT 2.75 3.06 2.51 5.22 4.10 2.69 2.63 2.69 2.52 2.45
CV (%) aPTT 3.82 4.23 4.43 5.30 4.94 4.18 4.69 4.04 4.63 4.45
Accuracy
J F M A M J J A S O
Bias (%) PT 2.30 2.53 2.00 1.00 1.30 1.23 1.53 2.61 1.53 1.07
Bias (%) aPTT 3.52 3.09 0.90 0.58 0.65 1.38 1.67 2.54 2.90 2.18
PT Levey-Jennings graphs’s analysis showed the violation of the following Westgard rules: 1 3s in January, 12s in
February and March; 10x in August and October and aPTT Levey-Jennings graphs’s analysis showed the violation
22s in March; 10x in May; 7x in August; 12s in September (Figure 2).

1019
ISSN: 2320-5407 Int. J. Adv. Res. 11(01), 1017-1022

Figure 1:- Levey-Jennings graphs analysis.

Legends :A : PT Mai 2021 : No Westgard rule violated ; B : PT January 2021 :13s Westgard rule violated ; C :aPTT
March 2021 : 22s Westgard rule violated ; D :aPTT May 2021 : 10x Westgard rule violated

Discussion:-
This study aimed to analyze routine tests QC including EQA and IQC about PT and aPTT results. Although many
coagulation QC programs have been in place for decades, most of these programs begin with the evaluation of
routine coagulation tests such as PT, aPTT, and fibrinogen [4]. In Madagascar, few health care system are available,
so many patients have to pay for their own health care cost including biological tests that should be well ranked.
Therefore, PT and aPTT are among the common tests ordered by Malagasy physicians to screen coagulation issues
in preoperative patients or in suspected blood disorder. That was why these routine coagulation tests were first
subjected to a QC analysis.

EQA was provided by UK-based NEQUAS which is based in Sheffield, UK, and has been inspected by the UK
Accreditation Service Ltd and has received certification to ISO 17043 [5]. This program was chosen for hematology
laboratory in Antananarivo Madagascar in this study thanks to the humanitarian aid program of the World
Federation of Hemophilia which sustain the latter for hemophilia and blood disorder’s diagnosis [5]. The QC
materials are sent by mail 5 times a year. In this study, post-office issues made that only 4 surveys were received
among the 5 that have been sent, and all of them were received late. In this study 4 out of 8 tests were out of

1020
ISSN: 2320-5407 Int. J. Adv. Res. 11(01), 1017-1022

consensus for PT and aPTT that can be due to both random and systematic errors. These errors can be due to
multiple factors [6] such as variations in environmental conditions (temperature, humidity) as local temperature may
vary to extreme values; degradation of reagents or kits or QC materials during shipment; a manual error by the
operator or a defect of the equipment and the small material (accuracy of the volumes distributed by the pipettes) or
calibration error. These errors can be avoided by careful reading and following each step of the standardized
operating procedure, identifying the critical points of the technical assay [7]. Advocacy to reduce the administrative
procedures for customs clearance for the mail sent by EQA can be useful to reduce the issues from quality. Other
studies have shown that education and training of laboratory personnel can improve the quality of test results [8].

IQC is implemented in a laboratory to detect in real time any malfunction that may occur in order to take the
necessary corrective actions [9]. IQC materials must be a commercialized control plasma of known values (high,
normal or low) or prepared by the laboratory and are to the closest possible from a patient’s sample. In this study,
the laboratory used as QC material a pool of normal plasmas. Its advantages are limitation for cost allocated for
commercialized QC material, no reconstitution errors. But the disadvantage of this control is the stability and the
expiration period which are not reliable. The storage condition may be affected by the laboratory environment.
These labs manufactured QC are stable one month at -20°C for one month, according to a study conducted in 2021
[10].

Precision corresponds to the perfect agreement between repeated measurements on the same sample. It characterizes
the dispersion of the values obtained during repeated measurements on the same sample by taking into account
factors such as operator, time, reagent batches, calibrations. It can be quantified by the standard deviation (SD) or
the coefficient of variation (CV). In this study, CV was used for precision’s analysis and showed 2 out-of-control
values (CV > 5%). Lack of precision is mostly indicative of a random error [3]. They are most often detected by the
violation of 12s, 13s, R4s Westgard's rules. Random errors can be related to the operator (incorrect execution of the
measurement process or non-observance of instrument maintenance), to the reagents (batch change or deterioration
of the reagent during storage or use), to the instruments (dysfunction of the sampling system, of the reaction medium
mixing process, of the photometer, dirty cuvettes) [2]. Corrective actions have been conducted regarding the
pipetting qualities and instrument’s laboratory checking (temperature, reaction vessels, metal beads). This type of
error in general does not reflect a defect in the analysis system, and therefore is not expected to be repeated.

Accuracy indicates the closeness of agreement between the mean value obtained from a series of test results and a
value that is accepted as either a conventionally true value or an accepted reference value. It can be quantified by
calculating bias which is a percentage deviation from the expected value.

In this study, all values were in-control (Bias < 10%). Lack of accuracy of the analytical process is most indicative
of a systematic error [3] Systematic error is detected as soon as a change in the mean of the control value appears.
This change in the mean may be gradual and appear as a drift or it may be sudden and come out as an offset. A drift
indicates a progressive loss of reliability in the analytical system. Drifts are usually subtle [11]. A shift occurs when
there is an abrupt change in the control mean. Shifts in quality control data represent a sudden, large, positive or
negative change in the performance of the analytical system [11]. These errors are most commonly evidenced by
violation of the 22S, 41S, and 10x Westgard rules. Inaccuracy of IQC results in a laboratory can also be related to
reagents, equipment, personnel, calibration and internal control procedures.

Conclusion:-
Quality control program in hematology laboratory at CHU-JRA Antananarivo Madagascar is well-established
despite several issues in EQA materials’ transportation, lab’s technical facilities resulting in random or systematic
errors. Each out-of-control results have to be analyzed as part of strengthening quality program guaranteeing
coagulation tests’ reliability.

Acknowledgements:-
Special thanks to all the lab’s staff and to the Humanitarian Aid Program of WFH for providing ability to be part of
an EQA program

Disclosure of interest:
Authors declare no conflict of interest in this manuscript’s publication.

1021
ISSN: 2320-5407 Int. J. Adv. Res. 11(01), 1017-1022

References:-
1. Drillaud N, Siaka C, Khalil A, Yin L, Prot JM, Caliste JP, et al. Qualité Operational quality for laboratories
with « 5S ». IRBM News. Febr 2017 ;38(1):82-6
2. Chouaieb S, Maamouri W, Jouini H, Zili M. Internal quality control in haemostasis. Revue Francophone des
Laboratoires. March 2019 ; (510) :16-23
3. Bonar R, Favaloro EJ, Adcock DM. Quality in coagulation and haemostasis testing. Biochemia Medica 2010
;20(2) :184–99
4. International Organization for Standardization ISO 15189. Medical laboratories - Quality and competence
requirements. 3rd edition (reviewed version 2014-08-15) Geneva. 2012.
5. Smock KJ, Moser KA. What have we learned from coagulation laboratory participation in external quality
programs. Int J Lab Hematol. 2019 ;41(Suppl. 1) :49–55
6. IEQAS WFH Participants. Manual General Info 2022-2023.
7. Kiatsiri K, Kanka A, Lerdsukviboon K, Sanrattakit C, Tubseenik P, Suksa-ard V, et al. Evaluation of glucose
testing performance in clinical laboratory by single rule and westgard multi rule. Chula Med J 2009 Mar-Apr
;53 (2): 91-106.
8. Drzevieck-renard S, Genesse M. Quality in biochemistry: precision and accuracy. Annales de biologieclinique.
2006 ;1 :2-3.
9. Le Moël G, Piton A, Pontezière C, Claisse C, Laureaux C, Marie B, et al. Quality assurance : internal
quality control and external quality assessment.Annales de Biologie Clinique. 2000 ; 58(1) : 103-10.
10. NiryManantsoa S, Rabearisoa JA, Rakotomalala RT, Raheritiana TM, Rasamoelina A, et al. Effects of
conserving plasma pools on coagulation exploration tests. International Journal of Biomedical and Advance
Research 2020; 11(07): e5445
11. Kamleh MA, Anas M, Ebbels TMD, Spagou K, Masson P and Want EJ. Optimizing the Use of Quality
Control Samples for Signal Drift Correction in Large-Scale Urine Metabolic Profiling Studies.Anal.
Chem. 2012, 84, 6, 2670–2677.

1022