| |

Received: 27 June 2018    Revised: 13 July 2018    Accepted: 3 August 2018

DOI: 10.1002/jcla.22661

RESEARCH ARTICLE

Differences in erythrocyte sedimentation rates using a

modified Westergren method and an alternate method

Elise Schapkaitz1  | Shilla RabuRabu2 | Marcel Engelbrecht1

1
Department of Molecular Medicine
and Hematology, Faculty of Health Introduction: Worldwide laboratories have adopted the use of modified or alternate
Sciences, University of Witwatersrand methods for measurement of the erythrocyte sedimentation rate (ESR). The iSED
Medical School, Johannesburg, South Africa
2
from Alcor Scientific is a novel, alternate ESR method based on photometric ag-
Department of Molecular Medicine and
Hematology, National Health Laboratory gregometry which offers improved operator safety and reduced analysis time. This
Service, Johannesburg, South Africa study evaluated the diagnostic utility of the iSED in a South African patient popula-
Correspondence tion with a range of inflammatory disorders.
Elise Schapkaitz, Department of Molecular Methods: We compared the iSED with the predicate modified Westergren method
Medicine and Hematology, University of
Witwatersrand Medical School, Faculty (StaRRsed, Mechatronics, Zwaag, the Netherlands) measured at 60 minutes. Analysis
of Health Sciences, Johannesburg, South was performed on K 2EDTA samples at three ESR measurement ranges (<20, 20-­80
Africa.
Email: elise.schapkaitz@nhls.ac.za and >80 mm/h) in 120 pediatric and adult inpatients and outpatients over a 2-­week
period. Precision, stability, and carryover were performed in accordance with the
revised International Council for Standardisation in Haematology guidelines.
Results: The iSED demonstrated acceptable imprecision with minimal carryover
(2.86%). The correlation coefficients at the 3 ESR measurement ranges were r = 0.58,
r = 0.71, and r = 0.56, respectively. The y-­intercepts were −10.74 (CI −29.17 to 7.69),
−5.95 (CI −18.60 to 6.69) and 246.05 (CI 591.42-­99.31). This indicated a difference of
a constant nature with an overall mean difference of 7.99 mm/h (CI 5.87-­10.13)
(P < 0.001). iSED ESR measurements were stable up to 24 hours when stored at
room temperature or at 4-­8°C.
Conclusion: This study demonstrated differences in ESR results, predominantly at
extremes of the analytical range, using an alternate method. Careful consideration
and performance monitoring of these novel methods are advised.

KEYWORDS
alternate erythrocyte sedimentation rate measurement, EDTA tube, erythrocyte
sedimentation rate, iSED, verification

1 |  I NTRO D U C TI O N monitor patients with Hodgkin Lymphoma post-­

c hemotherapy
for early disease relapse.4 The ESR is usually performed in con-
The erythrocyte sedimentation rate (ESR) or length of sedimen- junction with more sensitive and specific tests for evaluating dis-
tation reaction of blood, first described over 120 years ago, is ease activity such as the C-­reactive protein (CRP). The CRP, which
a widely used, non-­s pecific screening test.1,2 Although the test increases within 4-­6 hours of onset of inflammation or injury,
lacks specificity, the ESR is informative in a variety of inflamma- is more useful for disease diagnosis. In contrast, the ESR which
tory conditions, including rheumatoid arthritis, giant cell arteritis, rises within 24-­4 8 hours and declines slowly thereafter is more
polymyalgia rheumatica, and other connective tissue disorders. 3 valuable for predicting a response to treatment and monitoring
In addition, the ESR is frequently requested by oncologists to disease activity. 5,6

J Clin Lab Anal. 2018;e22661. wileyonlinelibrary.com/journal/jcla © 2018 Wiley Periodicals, Inc.  |  1 of 7
https://doi.org/10.1002/jcla.22661
 |
2 of 7       SCHAPKAITZ et al.

Westergren defined the reference method for ESR mea-

2.2 | Study protocol
surement, which uses a dilution of four volumes of blood to one
volume of sodium citrate.7 This measures the distance in milli-
2.2.1 | Blood sampling
meters (mm) that red blood cells (RBC) fall in a vertical column
after 1 hour thereby measuring all three stages of RBC sedimen- Adult and pediatric samples with adequate volume (>4 mL) from in-
tation. In the first stage, the RBC form rouleaux, in the second patients and outpatients which covered ESR values in the analyti-
stage, settling of the aggregates occurs at a constant rate, and in cal range of 2-­120 mm/h were randomly selected in May 2018.15
the final stage, the rate of sedimentation slows as the aggregated Samples were collected in K 2EDTA tubes or BD Microtainer MAP mi-
RBC pack at the bottom of the tube. The end result is a compact crotubes (Vacutainer, Becton Dickinson, UK). Samples were stored
column of sedimented RBC with a clear plasma portion. The tradi- at room temperature (RT) and analyzed by a dedicated technologist
tional Westergren method has several well-­d escribed limitations within 4 hours of collection. Results which indicated analysis failure
for routine laboratory practice. 8 In order to address these, novel were excluded from the final analysis. Samples were collected anon-
alternate and modified methods have been developed for ESR ymously, and no identifying information was recorded.
measurement. Alternate ESR methods employ different princi-
ples such as centrifugation or photometric aggregometry. These
2.2.2 | Laboratory methods
methods measure the initial stage of the ESR, namely rouleaux
formation, which results in a significantly shorter analysis time. The manufacturer was responsible for installation and training of the
Furthermore, these methods offer technical innovations such as laboratory staff members on the instrument operation, quality con-
closed tube analysis with improved operator safety; use of eth- trol, and safety. Seditrol commercial controls (Alcor Scientific, Rhode
ylenediamine tetra-­a cetic acid (EDTA) tubes; reduced blood vol- Island, USA) used for the evaluation were supplied by the manufac-
ume requirements and longer sample stability. Several studies of turer. Maintenance procedures were performed daily as per the
modified ESR methods have demonstrated improved precision and manufacturer’s recommendations. According to the manufacturer’s
excellent correlation with the reference Westergren method.9-11 operator manual, K 2EDTA samples were inserted manually into the
However, variable results have been reported using alternative sample entry port. Twenty samples were loaded consecutively into
methods of ESR measurement in patients with a range of clinical the sample wheel. In the microflow cell, RBC aggregation is directly
11-14
disorders. This raises the concern over introduction of a test measured by an optical detector in 20 seconds.
principle that differs from the reference Westergren method. On the StaRRsed, K 2EDTA samples are diluted with citrate in a
The iSED (Alcor Scientific Inc, Rhode Island, USA) is a newly ratio of 1:4. Samples are subsequently aspirated into vertical glass
developed automated alternate ESR analyzer. Similar to other pipettes. ESR measurement is performed after 60 minutes by an op-
alternate ESR tests on the market, the iSED is a closed tube tical sensor which measures the difference in absorbance between
random-­a ccess analyzer which has been designed to improve the the RBC and plasma layers. Temperature correction is applied. The
laboratory workflow by producing results within 20 seconds from laboratory’s analyzer is compliant with international proficiency
EDTA tubes. This method is based on photometric aggregome- testing program. Daily internal quality control was performed
try and requires a minimum sample volume of 100ul. This study with normal and abnormal StaRRsed ESR controls (Mechatronics,
aimed to verify the diagnostic and clinical utility of the iSED in Richmond, USA).
accordance with the International Council for Standardisation in
Haematology (ICSH) recommendations before accepting it into
2.3 | Comparison study
routine practice.15
For the method comparison study, samples were analyzed in paral-
lel. The iSED analyzer’s measurement range (1-­130 mm/h) was es-
2 |  M ATE R I A L S A N D M E TH O DS tablished from the samples analyzed. The Bland-­Altman difference
plot was used to assess the absolute differences, and a Deming re-
gression analysis was used to determine the degree of correlation
2.1 | Study design
between the iSED and StaRRsed analyzers. Correlation coefficients
The iSED was verified with regards to precision, stability, carryo- and biases for samples in the low (<20 mm/h), middle (20-­8 0 mm/h),
ver, and comparability to the reference StaRRsed (Mechatronics, and upper third (>80 mm/h) of the analytical range were determined.
Zwaag, the Netherlands) analyzer. The StaRRsed is a modi- Statistical significance was assumed to be P < 0.05.
fied Westergren method which is currently used for measuring
ESR at the Haematology Laboratory at the Charlotte Maxeke
2.4 | Precision
Johannesburg Academic Hospital (CMJAH). This verification was
performed in accordance with the revised ICSH2011 and 2017 Between-­run precision was performed with normal and abnormal
and the Clinical and Laboratory Standards Institute (CLSI) (2011) controls which were analyzed three times a day for five consecu-
recommendations.15-17 tive days. Within-­run precision was performed using normal and
SCHAPKAITZ et al. |
      3 of 7

abnormal controls and three patient samples (in the low, middle, and 120

upper third of the analytical range) analyzed ten times each during
the same 8-­hour period. The coefficient of variation (CV) was com- 100
pared to the manufacturer’s precision limits.

80
2.5 | Carryover

ISED ESR
Using the Broughton formula, three consecutive analyses of a pa- 60

tient sample with high viscosity (H1, H2, and H3) were analyzed fol-
lowed by three consecutive analyses of a patient sample with a low 40
viscosity (L1, L2, and L3). Carryover was calculated from the formula:
Carryover (%) = (L1-­L3)/(H3-­L3) × 100. Parametric paired t test was
20
used to assess the observed differences.

0
2.6 | Sample stability 0 20 40 60 80 100
Reference ESR
Stability analysis was performed on ten samples: five normal and five
abnormal. Analysis was performed at time zero (or as close to zero as F I G U R E   2   Correlation of the iSED and the reference method
possible) on the iSED. The samples were then divided into aliquots
and stored at RT and at 4-­8°C. Subsequent testing was performed
3 | R E S U LT S
after 4, 8, 12 and 24 hours of storage. The results were compared
using a parametric paired t test.
3.1 | Comparison study
One hundred and twenty samples (Female: Male of 5.4:1) were
2.7 | Reference ranges
included in the final analysis with an age range of 6-­85 years. The
Reference intervals were verified using existing samples in accord- ESR requests were from the rheumatology/orthopedics outpa-
ance with the CLSI guidelines.16 tient department (51.4%), the emergency medicine department
(37.1%), the infectious disease department (7.1%), and the oncol-
ogy department (4.3%). The StaRRsed produced no results in 10
2.8 | Ethics
samples which were excluded from the final analysis. Of these,
This study was approved by the Human Research Ethics Committee five were turbid samples (flagged as “hazy > 25 mm”) and a further
of the University of the Witwatersrand (M-­180236). five were low-­s ample volumes. In addition, the StaRRsed flagged
an additional 13 samples as unreliable owing to the presence of
bubbles on sampling. The iSED produced no error readings or anal-
60
ysis failures during the study.
The analytical range of the iSED was 2-­
130 mm/h. The me-
50
Difference (ISED ESR - Reference ESR)

dian (interquartile range) iSED ESR was 42.00 (29.00) mm/h which

40 was significantly higher than the reference ESR value of 30.90
(20.69) mm/h (P < 0.009). ESR values on the StaRRsed were falsely
30
increased in samples with low hematocrit levels (<0.36 L/L) and
were adjusted for the patients’ hematocrit results (P = 0.316).18
20
The mean difference between the two methods was 7.99 mm/h (CI
10 5.87-­10.13) (Figure 1). Deming regression analysis yielded a r = 0.88,
y-­intercept = 3.98 (CI 0.01-­7.95) and slope = 1.11 (CI 0.99-­1.22)
0 (Figure 2). The majority of the data points were within the 95% lim-
0 20 40 60 80 100 120
its of agreement with few outliers (n = 6). The outliers have been
-10
verified as true measurements performed within 4 hours of collec-
-20 tion. The y-­intercept of 3.98 (CI 0.01-­7.95), however, indicated a sig-
Average (Reference ESR + ISED ESR)/2 nificant difference of a constant nature between the two methods
(P < 0.001).
Bias CI Bias (95%) CI (95%)
We divided the results into subgroups according to the
F I G U R E   1   Bland–Altman difference plot comparing the iSED Westergren ESR levels. The results of the Deming regression and
and the reference method Bland-­
Altman analysis are shown in Table 1. Correlations in the
 |
4 of 7       SCHAPKAITZ et al.

TA B L E   1   Comparison statistics according to lower, middle, and upper third of the analytical range

Correlation coefficient
Analytical range N Bias (95% CI) (r) Intercept Slope

ESR (<20 mm/h) 26 1.79 (−1.76-5.35) 0.58 −10.74 (−29.17-7.69) 2.76 (1.22-­4.29)

ESR (20-­8 0 mm/h) 86 7.69 (4.95-­10.44) 0.71 −5.95 (−18.60-­6.69) 1.34 (0.99-­1.69)
ESR (>80 mm/h) 8 8.66 (−0.42-­17.73) 0.56 246.05 (591.42-­99.31) 3.85 (0.09-­7.60)

lower and upper third of the analytical range were weak to moder- time.15 This study evaluated the diagnostic utility of the iSED alternate
ate with a difference of a constant nature (r = 0.58, P < 0.001 and ESR method compared to the StaRRsed in an adult and pediatric pa-
r = 0.56, P = 0.059, respectively). The Bland–Altman difference plot tient population with a range of inflammatory disorders. The iSED is
showed a significant increase in the differences between the two a small bench-­top device (36cmx27cmx34 cm) which has been devel-
tests at ESR values >80 mm/h with an observed mean difference of oped for small to medium-­sized laboratories. The pre-­analytical steps
8.66 mm/h (CI −0.42-­17.73). of mixing are fully automated. In this study, the laboratory staff re-
Using the StaRRsed as the reference, 103 (85.83%) ESR results ported that the instrument was easy to operate following a short train-
from the iSED provided the correct clinical interpretation. However, ing period. Also, instrument maintenance was minimal and the iSED
there were 17 (14.17%) ESR measurements which would have re- produced no error readings or analysis failures during the study. In
sulted in a different clinical interpretation. Of these, 14 (11.67%) contrast, the StaRRsed is a large analyzer (182 cm × 153 cm × 95 cm)
were false-­positive results and 3 (2.50%) were false-­negative re- which produces results in 60 minutes and thus places practical limi-
sults. Additional parameters were retrospectively retrieved from the tations on laboratory workflow. The StaRRsed reported 10 sampling
laboratory information system in order to identify possible interfer- errors and flagged 13 unreliable results in this study which may have
ing variables, namely CRP, mean cell volume (MCV) and hematocrit. resulted in incorrect clinical interpretation and management.
The mean hematocrit and MCV levels in the group which would have The ESR measurement on the iSED is a calculated ESR value in
resulted in a different clinical interpretation were 0.39 ± 0.75 L/L mm/h derived from the initial stage of the ESR, namely RBC aggre-
and 90.69 ± 5.48, respectively. There was no statistically significant gation during rouleaux formation. Complete sedimentation of the
difference between the clinical groups (P = 0.511 and P = 0.565, re- RBC aggregates is not measured. However, a significant mean dif-
spectively). In addition, the corresponding CRP levels did not dif- ference of 7.99 mm/h (CI 5.87-­10.13) (P < 0.001) between the iSED
fer significantly between the iSED and StaRRsed clinical groups ESR and the StaRRsed ESR was observed in this study. On subgroup
(P = 0.187). analysis, a poor correlation between the iSED and StaRRsed meth-
ods at ESR levels at the lower and upper third of the analytical range
(2-­130 mm/h) with a large mean bias of 1.79 mm/h (CI −1.76 to 5.35)
3.2 | Precision
and 8.66 mm/h (−0.42 to 17.73), respectively, were found.
The results of the within-­run and between-­run precision analysis This significant difference we observed in this study has also been
on the iSED and StaRRsed with commercial controls are presented observed in previous studies of the iSED.3,12 These studies compared
in Table 2. The within-­run precision analysis for patient samples at to methods using citrated whole-­blood samples also reported a large
low, middle, and high ESR levels was within the manufacturer’s limits positive bias, in particular at higher ESR levels. Westergren ESR test-
(Table 3). ing methods are influenced by intrinsic factors such as hematocrit
and viscosity as well as extrinsic factors such as sample length and
temperature,13 whereas these factors do not interfere significantly
3.3 | Stability
with ESR measurement on the iSED. Bogdaycioglu et al12 found cor-
The undiluted ESR was stable for 24 hours on the iSED analyzer relation of 0.76 with the manual Westergren method across all ranges
when stored at RT and at 4-­8°C (Figure 3). with a mean bias of 13 mm/h (CI −35.7 to 61.6). Biederman et al19
reported a mean bias of 6 ± 36 mm/h at ESR levels > 30 mm/h when
compared to the Streck ESR Autoplus (Streck, Omaha, NE). However,
3.4 | Carryover
across all ranges, the mean bias was small at −0.3 ± 12 mm/h.
The percentage carryover for the iSED ESR measurement was 2.86%. Similarly other novel methods, have also been reported to overes-
This was not clinically significant and within the manufacturer’s limit. timate the ESR compared to the reference Westergren method.10,14,20
Results of proficiency testing indicate differences of ~40% between
Westergren and non-­Westergren-­based methods in particular at upper
4 |  D I S CU S S I O N and lower ends of the analytical range.15 Shelat et al14 studied a pediatric
population and found a correlation 0.63 with a mean bias of 3.30 mm/h
Worldwide most laboratories have adopted modified or alternate ESR (CI 1.52-­5.08) in the normal range 0-­20 mm/h with the ESR STAT PLUS
methods which offer improved operator safety and reduced analysis (HemaTechnologies, Lebanon, NJ). The results of this centrifugation
SCHAPKAITZ et al. |
      5 of 7

TA B L E   2   Within-­run and between-­run

Expected
precision analysis with commercial
Mean ± SD (mm/h) CV (%) range (mm/h)
controls
Within-­run precision
Normal control
iSed method 10.30 ± 1.34 12.99 3-­17
StaRRsed method 4.70 ± 0.48 10.28 0-­10
Abnormal control
iSed method 62.40 ± 2.41 3.87 38-­90
StaRRsed method 39.80 ± 0.42 1.06 33-­53
Between-­run precision
Normal control
iSed method 10.20 ± 1.32 12.9 3-­17
StaRRsed method 4.60 ± 0.52 11.23 0-­10
Abnormal control
iSed method 62.67 ± 1.99 3.17 38-­90
StaRRsed method 41.30 ± 1.42 3.43 33-­53

TA B L E   3   Within-­run precision analysis

Normal Abnormal-­Middle range Abnormal-­High range
on the iSED ESR analyzer with patient
samples Measured mean 13.50 21.00 85.00
(mm/h)
SD 2.07 2.67 7.44
CV (%) 15.33 12.72 8.75
Manufacturer’s <20 <15 <10
limits (%)

method, however, are strongly influenced by extrinsic factors such as significantly from measurements reported by the Westergren
mixing. Mahlangu et al and Vennapusa et al showed a correlation of 0.96 method.10 The imprecision of the abnormal ESR control, which en-
and 0.83 with a mean bias of 6.6 mm/h (CI 5.0-­8.1) and 7.13 mm/h, re- compasses the clinically significant ESR range, was low. However,
10,20
spectively, across all ranges on the Streck ESR Autoplus. In contrast, the imprecision of the normal ESR control was high. This, however,
other novel methods have been reported to underestimate the ESR, in does not affect the clinical reliability of the measurements as it
particular at lower ESR levels.11,13 Curvers et al13 showed a correlation encompasses the ESR of the normal population. 22 Accordingly, the
of 0.83 with a mean bias of −5.7 mm/h (CI −50.8 to 39.4) on the Ves-­ College of American Pathologists’ proficiency testing scheme for
Matic Cube 200 (Diesse Diagnostica Senese, Siena, Italy), as compared 2017 reported a %CV of 17.4% for ESR results in the normal refer-
to the Westergren method. Variable results, however, have been re- ence range on 310 iSED users. In this study, normal and abnormal
ported with this method, which measures all three stages of ESR sedi- patient samples were stable when analyzed on the iSED beyond
mentation with an analogue sensor.12,21 Further, this method has several 24 hours when stored at RT and 4-­8 °C, thus excluding instability
limitations. These include a minimum sample volume of 4 mL, analysis as a potential factor. A major limitation of the Westergren ESR
time of 20 minutes and no automated dilution. As such hematocrit and is the need to analyze the test within 4 hours from the time of
viscosity will interfere significantly with results of this sedimentation collection when stored at RT.15 The result is that many referred
method. The Test-­1(Alifax S.p.A., Polverara, Italy), in contrast, performs ESR samples in everyday practice are rejected. Alternate and mod-
the ESR from a small volume of undiluted blood in three minutes. A pre- ified ESR’s performed in EDTA, however, have demonstrated im-
vious report by Van der Maas et al11 however, showed a correlation of proved sample stability.12,21 Both test methods use EDTA as an
0.67 with a mean bias of −4.4 mm/h (CI −7.8.4 to 1.0) as compared to anticoagulant; therefore, higher hematocrit values may have been
the StaRRsed at 60 minutes. These studies confirm that systemic biases a contributing factor. Our analysis showed that the StaRRsed had
exist among automated testing devices based on aggregation or centrif- a tendency to overestimate the ESR in low hematocrit samples,
ugation principles as compared to sedimentation methods. whereas intrinsic factors did not interfere with ESR measurement
We considered additional potential factors that might be re- on the iSED. This effect was reduced by applying the Fabry correc-
sponsible for the observed systematic bias between the iSED and tion to the StaRRsed results.18
StaRRsed. The differences were not related to imprecision be- In order to determine the suitability of the iSED for this specific
cause the precision of the iSED was acceptable and did not differ patient population, the clinical performance of this method was
 |
6 of 7       SCHAPKAITZ et al.

350
(A)
300

250

200

150

100

50

-50

-100

-150

-200
4h 8h 12h 24h

Storage temperature

250
(B)
200

150

100

50

-50

-100

-150 F I G U R E   3   Boxplot of the mean

4h 8h 12h 24h
percentage difference during storage at
room temperature (A) and 4-­8°C (B) for
Storage temperature
the iSED ESR measurement

assessed. There were 14 samples (11.67%) which would have resulted that alternate methods estimate the ESR from light reflected from or
in a different clinical interpretation. On analysis of the corresponding transmitted through a blood sample during initial rapid erythrocyte
CRP, hematocrit, and MCV levels, a significant difference was not ob- aggregation.23 It follows that careful consideration and performance
served. This raises concern that alternate and Westergren methods monitoring of these novel methods are advised.
are not interchangeable. Fundamentally, the alternate and Westergren In this study, the majority of the patients were rheumatology out-
methods measure different analytical endpoints. Westergren methods patients with rheumatoid arthritis. Van der Maas et al11 previously
measure erythrocyte sedimentation, whereas alternate methods mea- demonstrated that use of an alternate ESR method affected the valid-
sure erythrocyte aggregation. Real-­time studies have demonstrated ity of the Disease Activity Score 28 (DAS 28). In their study of patients
SCHAPKAITZ et al. |
      7 of 7

with rheumatoid arthritis, 26% of patients were misclassified when the 6. Litao MK, Kamat D. Erythrocyte sedimentation rate and C-­reactive
Westergren ESR was replaced by the Alifax Roller Test-­ITH ESR. As protein: how best to use them in clinical practice. Pediatr Ann.
2014;43:417‐420.
such, careful clinical and laboratory monitoring of these novel methods
7. International Committee for Standardization in Haematology.
is indicated. In conjunction with the ICSH recommendation, we advise Reference method for the erythrocyte sedimentation rate (ESR)
that the following interpretative comment be added to results from the test on human blood. Br J Haematol. 1973;24:671‐673.
iSED in order to assist clinicians with the correct interpretation: “This 8. Besson I. Evaluation of 3 automatic systems for measurement of the
erythrocyte sedimentation rate. Sangre. 1995;40:103‐107.
result was obtained with an ESR instrument that is not based on the
9. Horsti J, Rontu R, Collings A. A comparison between the StaRRsed
standard Westergren methodology. The sensitivity and specificity of auto-­compact erythrocyte sedimentation rate instrument and the
this alternate method for various disease states are 97.78% and 44.44%, Westergren method. J Clin Med Res. 2010;2:261‐265.
respectively.15” 10. Mahlangu JN, Davids M. Three-­way comparison of methods for the
measurement of the erythrocyte sedimentation rate. J Clin Lab Anal.
A limitation of this study is that samples with interfering sub-
2008;22:346‐352.
stances such as fibrinogen and paraprotein, which increase rouleaux 11. van der Maas A, van den Ende CH, van Eerd J, Fransen J, den
formation, were not collected during the study period. Additionally, Broeder AA. The use of different methods for rapid determination
the iSED was not compared with the manual Westergren method of the ESR induces DAS28 misclassification in clinical practice. Clin
which is the gold standard. This study, however, contributes to the Exp Rheumatol. 2010;28:477‐482.
12. Bogdaycioglu N, Yilmaz FM, Sezer S, Oguz E. Comparison of iSED
published literature as the manual Westergren method has largely
and Ves-­Matic Cube 200 erythrocyte sedimentation rate measure-
been replaced worldwide by automated methods owing to the in- ments with Westergren method. J Clin Lab Anal. 2015;29:397‐404.
herent limitations of the manual method. 13. Curvers J, Kooren J, Laan M, et al. Evaluation of the Ves-­Matic
Cube 200 erythrocyte sedimentation method: comparison with
Westergren-­based methods. Am J Clin Pathol. 2010;134:653‐660.
14. Shelat SG, Chacosky D, Shibutani S. Differences in erythrocyte
5 | CO N C LU S I O N sedimentation rates using the Westergren method and a centrifu-
gation method. Am J Clin Pathol. 2008;130:127‐130.
In conclusion, the iSED is a novel alternate method for ESR measure- 15. Kratz A, Plebani M, Peng M, et al. ICSH recommendations for mod-
ified and alternate methods measuring the erythrocyte sedimenta-
ment which provides rapid and precise determination of the ESR. The
tion rate. Int J Lab Hematol. 2017;39:448‐457.
analyzer offers major advantages including reliability, simplicity, use 16. Clinical Laboratory Standards Institute (CLSI). Procedure for the
of EDTA samples with reduced sample volume and prolonged sample Erythrocyte Sedimentation Rate (ESR) Test; Approved Standard
stability. This has the potential to reduce the sample rejection rate in (5th edn., H2-A5), Villanova, PA. CLSI. 2011.
17. Jou JM, Lewis SM, Briggs C, et al. ICSH review of the measure-
routine clinical practice. This study confirmed differences in ESR re-
ment of the erythocyte sedimentation rate. Int J Lab Hematol.
sults measured by the iSED as compared to the reference StaRRsed 2011;33:125‐132.
method. Difference was largely observed at the high and low ends of 18. Fabry TL. Mechanism of erythrocyte aggregation and sedimenta-
the analytical range. Careful consideration and performance moni- tion. Blood. 1987;70:1572‐1576.
19. Biederman L, Gulati G, Gong G. Comparison of Erythrocyte
toring of these novel methods are advised when used in routine labo-
Sedimentation Rate (ESR) between the alcor iSED ESR analyzer and
ratory practice. Further studies evaluating the impact of alternate the streck autoplus ESR analyzer. Am J Clin Pathol. 2014;142:A024.
ESR methods on clinical diagnosis and management are required. 20. Vennapusa B, De La Cruz L, Shah H, Michalski V, Zhang QY.
Erythrocyte sedimentation rate (ESR) measured by the Streck ESR-­
Auto Plus is higher than with the Sediplast Westergren method: a
ORCID validation study. Am J Clin Pathol. 2011;135:386‐390.
21. Perovic E, Bakovic L, Valcic A. Evaluation of Ves-­Matic Cube 200–
Elise Schapkaitz  http://orcid.org/0000-0002-1534-2930 an automated system for the measurement of the erythrocyte sed-
imentation rate. Int J Lab Hematol. 2010;32(1 Pt 2):88‐94.
22. Plebani M, De Toni S, Sanzari MC, Bernardi D, Stockreiter E. The
REFERENCES TEST 1 automated system: a new method for measuring the eryth-
rocyte sedimentation rate. Am J Clin Pathol. 1998;110:334‐340.
1. Grzybowski A, Sak J. A short history of the discovery of the eryth-
23. Ziya Isiksacan Z, Erelb O, Elbuken C. A portable microfluidic system
rocyte sedimentation rate. Int J Lab Hematol. 2012;34:442‐444.
for rapid measurement of the erythrocyte sedimentation rate. Lab
2. Piva E, Sanzari MC, Servidio G, Plebani M. Length of sedimenta-
Chip. 2016;24:4682‐4690.
tion reaction in undiluted blood (erythrocyte sedimentation rate):
Variations with sex and age and reference limits. Clin Chem Lab Med.
2001;39:451‐454.
3. Brigden ML. Clinical utility of the erythrocyte sedimentation rate. How to cite this article: Schapkaitz E, RabuRabu S,
Am Fam Physician. 1999;60:1443‐1450. Engelbrecht M. Differences in erythrocyte sedimentation
4. Haybittle JL, Hayhoe FGJ, Easterling MJ, Jeliffe AM. Review of rates using a modified Westergren method and an alternate
British National Lymphoma Investigation studies of Hodgkin’s dis-
method. J Clin Lab Anal. 2018;e22661. https://doi.
ease and development of prognostic index. Lancet. 1985;1:967‐972.
5. Costenbader KH, Chibnik LB, Schur PH. Discordance between org/10.1002/jcla.22661
erythrocyte sedimentation rate and C-­reactive protein measure-
ments: clinical significance. Clin Exp Rheumatol. 2007;25:746‐749.