Cytometry Part B (Clinical Cytometry) 94B:189–195 (2018)

Original Article
Flow Cytometric Osmotic Fragility Test:
Increased Assay Sensitivity for Clinical
Application in Pediatric Hematology
Olga Ciepiela,1* Anna Adamowicz-Salach,2 Agata Zgodzin
 ska,3
3 1
Magdalena Łazowska, and Iwona Kotuła
1
Department of Laboratory Diagnostics and Clinical Immunology of Developmental Age, Medical University
of Warsaw, Warsaw, Poland
2
Department of Pediatric Hematology and Oncology, Medical University of Warsaw, Warsaw, Poland
3
Students’ Scientific Group at Department of Laboratory Diagnostics and Clinical Immunology of
Developmental Age, Medical University of Warsaw, Warsaw, Poland

Background: Osmotic fragility test (OFT) is widely considered as a sensitive indicator of red blood
cells’ sensitivity to the hypotonic solution. It is often used as a screening test for the diagnosis of hered-
itary spherocytosis (HS). Nowadays, the osmotic fragility test based on flow cytometric analysis (FCM
OF) is widely used in laboratory practice. The purpose of this study was to optimize the assay sensitivity
and to validate its clinical application in the diagnostic screening of childhood anemias.
Methods: The study was conducted on 175 children suffering from various types of anemia (including 30 chil-
dren with proven hereditary spherocytosis, HS) and 16 healthy subjects. All children were aged between 3 months
and 17 years, including 94 boys and 97 girls. FCM OF was performed on every subject according to two different
analysis time patterns (hemolysis was analyzed for 214 or 300 s) using Cytomics FC500 flow cytometer.
Results: Significant higher sensitivity was demonstrated by the tests carried out according to the longer
analysis time pattern (90.0 vs. 83.33%). The level of specificity of both the analysis patterns was similar.
When an extended analysis time was used, the percentage of red cell survival levels in HS patients were
significantly lowered compared to the same cases analyzed with shorter incubation times and all other
non-HS anemic cases (9.31 6 4.69 vs. 35.59 6 15.30%, P < 0.05). During the shorter analysis time,
the values obtained were 13.76 6 7.92% for HS and 48.18 6 19.04% for non-HS, P < 0.0001. The
300-s test is very useful in distinguishing thalassemia patients from patients with other types of anemias
(94.74% sensitivity and 90.12% specificity) and provided the values of remaining red blood cells as
70.46 6 12.29% for thalassemia and 27.16 6 13.01% for nonthalassemia subjects, P < 0.0001.
Conclusion: Flow cytometric osmotic fragility test with a longer (300-s) analysis time demonstrated an
increased sensitivity in detecting HS in anemic children. V C 2017 International Clinical Cytometry Society

Key terms: Key terms: flow cytometric osmotic fragility test; flow cytometry; hereditary spherocytosis;
osmotic fragility; thalassemia

How to cite this article: Ciepiela, O., Adamowicz-Salach, A., Zgodzi nska, A., Łazowska, M. and Kotuła, I. Flow
Cytometric Osmotic Fragility Test: Increased Assay Sensitivity for Clinical Application in Pediatric Hematology.
Cytometry Part B 2018; 94B: 189–195.

*Correspondence to: Olga Ciepiela, Department of Laboratory Received 5 June 2016; Revised 11 January 2017; Accepted 13
Diagnostics and Clinical Immunology of Developmental Age, Medical January 2017
University of Warsaw, Zwirki i Wigury 63a; 02-091 Warsaw, Poland Published online 19 January 2017 in Wiley Online Library (wileyon-
Email: olga.ciepiela@wum.edu.pl linelibrary.com).
Conflict of Interest: Authors declare that they do not have any con- DOI: 10.1002/cyto.b.21511
flict of interests.

C 2017 International Clinical Cytometry Society

V
190 CIEPIELA ET AL.
Diagnosis and Types of Tests Used for the Disease Confirmation of the Enrolled Children
Diagnosis
Hereditary spherocytosis
(n 5 30, 11 boys, 19 girl) age 7.1 6 6.1 years
Autoimmune hemolytic anemia
(n 5 13, 7 boys, 6 girls) age 7.1 6 6.1 years
Thalassemia
(n 5 19, 13 boys, 6 girls) age 9.2 6 6.4 years
Iron deficiency anemia
(n 5 32, 16 boys, 16 girls) age 6.1 6 5.3 years
Anemia associated with chronic renal failure
(n 5 6, 5 boys, 1 girl)12.7 6 4.3 years
Anemia of chronic disease
(n 5 17, 8 boys, 9 girls) 9.3 6 5.4 years
Anemia of undefined etiology
(n 5 58, 24 boys, 34 girls) 8.8 6 5.3 years
Healthy
(n 5 16, 8 boys, 8 girls) 10.3 6 4.7 years
CBC, complete blood count; EMA, eosin-50 -maleimide binding test; DAT, direct antiglobulin test; Hb, hemoglobin; TIBC, total
iron-binding capacity; CRP, C-reactive protein.
INTRODUCTION
Hereditary spherocytosis (HS) is an inherited deficien-
cy of the cytoskeleton proteins in erythrocytes plasma
membrane (1,2). The main diagnostic tests are complete
blood count (CBC) with mean corpuscular hemoglobin
concentration (MCHC) and red blood cell distribution
width (RDW), osmotic fragility test (OF), and eosin-50 -
maleimide test (EMA) (3). Most often patients suffering
from HS have high MCHC and RDW, normal mean cor-
puscular volume (NCV) of red blood cells, high osmotic
fragility, and significant fluorescence decrease in EMA
test. Recently, a flow cytometric method for OF assess-
ment has been introduced (4,5). This assay is easy to
perform and needs usage of saline, deionized water, and
flow cytometry (FCM), which helps to identify and
select the red blood cells electronically, thanks to their
physical properties. For this test, no fluorescent dyes
and detectors of fluorescence were required.
Osmotic fragility of red blood cells describes the cells’
ability to incorporate hypotonic solution. Due to their
surface-area-to-volume ratio, regular RBCs can incorpo-
rate hypotonic solution in limited quantities without
damaging the cells. Spherocytes, which have low
surface-area-to-volume ratio, easily hemolyze in the solu-
tions with <0.9% NaCl. This property forms the basis
for the development of a method to measure osmotic
fragility. The flow-based method is based on counting
the red blood cells that remain intact after incubation
with hypotonic saline for a predefined time and ratioing
this value with a baseline count made in normal saline.
Despite its ease of performance, it is necessary to assess
the reference ranges for the studied population and for
the laboratory equipment used.
There are few reports on the use of flow cytometric
osmotic fragility test showing impressive values for spe-
cificity and sensitivity. However, studies that describe
the flow cytometric osmotic fragility test (FC OF) do not
indicate its performance accurately enough (1,4). Thus,
Table 1
Confirmation tests
CBC, EMA <81.5%, clinical evaluation
CBC, DAT positive
CBC, Hb electrophoresis
CBC, ferritin <61.4 pmol/L, TIBC > 77 mmol/L
CBC, urinalysis, serum creatinine and urea, GFR <60 mL/min/1.73m2
CBC, transferrin <1.88 g/L, ferritin >561 pmol/L, CRP >3 mg/L
CBC
CBC, CRP <3 mg/L, physical examination
our aim was to validate the flow cytometric osmotic fra-
gility test and to propose an extension of analysis time
to improve the diagnostic value of the test.
MATERIALS AND METHODS
A group of 175 children with different types of ane-
mia (decrease in Hb concentration below the reference
value) and 16 healthy children after physical examina-
tion was enrolled to the study. Their diagnosis and types
of tests used for the disease confirmation are presented
in Table 1. This study was approved by the Ethical Com-
mittee at Medical University of Warsaw, no. KB/71/2015.
For the flow cytometric osmotic fragility test, appro-
priate volume of peripheral blood collected in tube con-
taining K3EDTA was suspended in 1100 mL 0.9% NaCl.
The volume of blood was calculated according to the
equation that allowed collecting similar number of cells
from every specimen. The equation was as follows:
130
V ðlLÞ5 RBC =106 , where V is the volume of blood col-
lected for the analysis and RBC the number of red blood
cells in 1 mL of the analyzed sample.
Next 10 mL of primary suspension was then sus-
pended in 1100 mL of saline in a cytometric tube. The
secondary suspension was analyzed with the flow
cytometer Cytomics FC500 (Beckman Coulter, Brea, CA,
USA). The red blood cells were gated in a forward scat-
ter/side scatter cytogram in logarithmic scale. The
events acquisition was performed using a cytogram with
time versus forward scatter in logarithmic scale. After
30 s of analysis, the cytometer was paused, 900 mL of
deionized water was added to the tube and the analysis
was continued as shown in Figure 1 (1,6).
Due to some variations in the literature regarding the
proposed sample incubation time, two different patterns
of events acquisition were performed. According to the
first pattern of analysis, the test took 300 s equally divid-
ed into nine gates of 30 s each (Fig. 1A). In the second
pattern of analysis, the whole test took 214 s divided into
Cytometry Part B: Clinical Cytometry
 FLOW CYTOMETRIC OSMOTIC FRAGILITY TEST IN CHILDREN 191

FIG. 1. Flow cytograms presenting pattern of analysis. Cytograms forward scatter in logarithmic scale versus time were divided into eight gates, 30
(A) or 22 (B) s each. Analysis presented in (A) last for 300 s and in (B) for 214 s. Arrow points to a break during acquisition when water was added
to the suspension of erythrocytes. [Color figure can be viewed at wileyonlinelibrary.com]

eight gates of 22 s each. After the first gate, where acqui-
sition of nonwater-treated cells was performed, an
ungated region that lasts for 30 s is placed. This break in
acquisition was necessary for water addition (Fig. 1B).
The results of the test are expressed as a percentage
of residual red blood cells measured with the primarily
proposed Eq. (4), which is as follows:
ðY 1ZÞ=2
% RRBC5 3100%;
A3 1:1
2 sitivity of the test was 90.0%, and the specificity 84.6%.
where RRBC is the remaining red blood cells, A the
number of cells in the first gate, Y the number of cells
in the next to last gate, and Z the number of cells in the
last gate.
Statistical analysis was performed using the GraphPad
Prism 6 software. The receiver–operator curve (ROC)
was used to assess the specificity, sensitivity, and area
under curve (AUC) values. A Mann–Whitney U test was
used to compare the values of RRBC for patients and
control groups. A P values of <0.05 was considered as
being significant for all statistical analyses.
RESULTS
The mean values of the remaining red blood cells in
the 300-s pattern of analysis for HS was 9.31 6 4.69 vs.
35.59 6 15.30% for all non-HS samples included in the
study (Table 2). At the cutoff value of 17.3%, RRBC sen-
Area under curve from ROC analysis was 0.9444,
P < 0.0001 (Fig. 2A).
For shorter analysis time, obtained values of remaining
red blood cell percentage for HS patients were
13.76 6 7.92 vs. 48.18 6 19.04% for other (non-HS)
samples, P < 0.0001 (Table 2). The ROC curve analysis
showed a sensitivity of 83.33% and specificity of 87.65%
at a cutoff value of 23.34%. The AUC was 0.9222,
P < 0.0001 (Fig. 2B).

Cytometry Part B: Clinical Cytometry

192 CIEPIELA ET AL.

FIG. 2. Results of analysis of HS subjects. The ROC curve and range of values of %RRBC for flow cytometric osmotic fragility test performed
according to 300-s pattern of analysis (A). The ROC curve and range of values of %RRBC for flow cytometric osmotic fragility test performed accord-
ing to 214-s pattern of analysis (B). The bar “non-HS” includes all non-HS subjects (anemic and healthy) enrolled to the study. [Color figure can be
viewed at wileyonlinelibrary.com]

Moreover, we made a comparison of the FC–OF
results and EMA test results for children with HS. For
the study group of 30 HS individuals, EMA test result
was 75.02 6 10.72% for control red blood cells fluo-
rescence. The sensitivity of the EMA test was 86.7%,
four children, despite clinically confirmed disease,
had negative (EMA fluorescence >81.5%) result. Inter-
estingly, those four children had a positive result for
FC–OF (12.38 6 4.69% RRBC), and children with
negative FC–OF test had a positive EMA test result
(73.05 6 6.77% of control red blood cells
fluorescence).

Table 2
Values of Remaining Red Blood Cells Percentage (%RRBC) for Different Studied Groups According to 300-S and 214-S Pattern
of Analysis
300 s of analysis 214 s of analysis
% RRBC SD P1 P2 % RRBC SD P1 P2
HS 9.31 4.69 <0.0001 <0.0001 14.64 8.78 <0.0001 <0.0001
Thal 70.45 12.29 <0.0001 <0.0001 74.60 12.45 <0.0001 <0.0001
AIHA 19.93 7.16 0.0354 0.0016 37.80 15.61 0.4424 0.7558
IDA 34.89 14.25 0.3056 0.7083 43.16 18.39 0.9588 0.5894
CKD 19.76 7.29 0.1562 0.0193 46.83 20.71 0.7012 0.4463
ACD 30.10 12.90 0.7747 0.1962 45.99 18.51 0.6017 0.3713
Anemia of undefined 30.98 9.98 n/a 0.1201 47.85 17.75 n/a 0.1945
etiology
Healthy 36.82 10.05 n/a n/a 39.84 10.42 n/a n/a

The P1 indicates statistical significance of difference between %RRBC in patients from analyzed group and all others. And P2
indicates statistical significance of difference between %RRBC in patients from analyzed group and healthy controls. HS, hereditary
spherocytosis; Thal, thalassemia; AIHA, autoimmune hemolytic anemia; IDA, iron deficiency anemia; CKD, chronic kidney disease;
ACD, anemia of chronic disorders. All P values <0.05 were bolded.

Cytometry Part B: Clinical Cytometry

 FLOW CYTOMETRIC OSMOTIC FRAGILITY TEST IN CHILDREN 193

FIG. 3. Results of analysis of thalassemia subjects. The ROC curve and range of values of %RRBC for the flow cytometric osmotic fragility test per-
formed according to 300-s pattern of analysis (A) and the ROC curve and range of values of %RRBC for the flow cytometric osmotic fragility test per-
formed according to 214-s pattern of analysis (B). The bar “non-TH” includes all nonthalassemic subjects (anemic and healthy) enrolled to the
study. [Color figure can be viewed at wileyonlinelibrary.com]

When comparing the thalassemia subjects with other
(nonthalassemic) children included in the study (Table
2), FC OF reached 94.74% sensitivity and 90.12% specif-
icity at a cutoff value of 48.01% RRBC and with AUC of
0.9731 in the longer time pattern of analysis,
P < 0.0001. The second pattern of analysis at a cut-off
value of 60.54% RRBC gave 84.21% sensitivity and
80.23% specificity, with AUC of 0.8934, P < 0.0001
(Fig. 3).
Interestingly, there was a significant difference
between %RRBC in shorter and longer time analysis pat-
tern only in patients with chronic kidney disease
(P 5 0.047) and anemia of undefined etiology
(P < 0.0001). The smallest difference in %RRBC
between 300-s and 214-s pattern of analysis was found
in thalassemic and healthy subjects.
DISCUSSION
A flow cytometry-based diagnosis of HS is still under
research and development. Recently, Agarwal et al. (7)
have published a paper that presents a new application
of eosin-50 -maleimide test based on the analysis of EMA
footprint, which indicates the position of EMA-stained
red blood cells in the cytogram with respect to EMA
fluorescence and cells size (forward scatter). Similarly, a
new test proposed by Won and Shu (4) describes the
flow cytometry-based evaluation of osmotic fragility. In
this study, we analyze different patterns of test interpre-
tation and indicate that longer time of analysis allows to
obtain more accurate results with higher sensitivity and
specificity.
The sensitivity of the test in detecting HS was compa-
rable with values obtained by other studies which were
100% (1), 93.9% (5), 91.3% (8), and 85.7% (9), respec-
tively. However, it is worth mentioning that those stud-
ies were performed mostly in adult patients suffering
from HS. Very similar sensitivity of FC OF test was
reached in another study performed from the whole
blood collected in a capillary tube (2). The main differ-
ence between the present study and the other studies is
the choice of reference group. In this study, we com-
pare the flow cytometric osmotic fragility test results
with different groups of anemic children, believing that
screening for HS is more frequently performed among
children with anemia rather than among the healthy
children with no symptoms. We found that FC OF has
high level of significance in distinguishing HS patients
from other anemias. The results stay in line with the
observation made by Park et al. (5).
Due to the discrepancies in the described time of
analysis in papers, where the studied test was used, we
decided to analyze two patterns of tests that were

Cytometry Part B: Clinical Cytometry

194 CIEPIELA ET AL.
carried on for 300 and 214 s. We also studied the pat-
tern of analysis lasting for 110 s, but the results were
not reliable at all and did not allow distinguishing
between studied types of anemia. The discrepancies
that we found concerning the different times of analysis
that were mentioned in the description of the test are
shown in the figure. Won and Suh described that the
acquisition lasts for 204.8 s, but in the same paragraph
they add that it lasted for approximately 2 min. These
two time values are totally different, and readers cannot
get clear information regarding the actual time of sample
analysis. Additionally, a picture illustrating the test per-
formance contains eight gates for, according to the
authors’ instruction, 11 s each. Based on the figure with
eight gates and a time-break in acquisition for water
addition, the whole analysis should last for approximate-
ly 110 s, but in the picture the time of analysis is given
as 204.8 s (4). In the present study, we decided to ana-
lyze the time of analysis similar to the one performed by
the other study, 214 s and longer with 300 s, as we
observed a difference in results between both patterns
in preliminary studies. Park et al. also introduced a mod-
ification of the primarily proposed scheme of FC–OF.
They modify the equation used for calculation of the
result by removing a dilution factor from the denomina-
tor. The authors state that it allowed to shorten the time
of obtaining the final result of the test and did not influ-
ence the diagnostic value of the analysis (5).
The cutoff value of the percentage of residual red
blood cell for HS screening mostly depends on the stud-
ied group and the analyzer used. In our study, the cutoff
values were different for both the times of analysis. Due
to the higher specificity and sensitivity of 300-s analysis
pattern, we chose a threshold of 17.3% for distinguish-
ing between the HS and non-HS individuals. It is less
than what Tao et al. (23.6%) (9) or Crisp et al. (22.8%)
(2) indicated. The cutoff value in different experiments
might depend on the severity of the disease of enrolled
subjects. Shim et al. found the correlation between
residual red blood cells percentage and severity of the
disease. The more severe the disease is, the lower is the
number of residual red blood cells that remain in
the suspension. The authors found higher sensitivity of
the test when the result is expressed as a ratio between
the healthy individual and patient’s %RRBC, compared
to only using the %RRBC of the studied subject (8).
We found that the sensitivity of the FC–OF test for HS
diagnosis is comparative with the sensitivity of EMA test
in our studied group. Interestingly, patients with nega-
tive EMA test had a positive FC–OF and those with nega-
tive FC–OF had a positive EMA. Both tests need small
volume of blood (5–15 mL of EDTA anticoagulated sam-
ple), and they could be performed from one sample col-
lected for complete blood count analysis. The advantage
of EMA test is the possibility to perform the test up to
7 days after blood collection (10), but FC-OF should be
performed on the day of sample collection. The specific-
ity of EMA tests, which vary in different papers between
90 and 100%, is higher than FC–OF; however, the costs
of FC–OF which are significantly lower compared with
the costs of an EMA test account for performing FC–OF
primarily to EMA as a screening test for HS diagnosis.
Moreover, the turnaround time of FC–OF is <30 min,
and the procedure of EMA staining including selection
of proper control samples takes >2 h. Since the specific-
ity of both tests is similar and they could complement
each other, we suggest to perform both tests for every
patient suspected of HS.
Some studies indicate that flow cytometric osmotic
fragility test could be useful in thalassemia diagnostics
(6). We found that FC OF test presents even better labo-
ratory parameters (AUC) in screening for thalassemia
than HS. The high utility of the test in thalassemia diag-
nosis was also found by Farias and Freitas (11). In our
study, sensitivity of FC OF for thalassemia diagnostics
was 94.74%, while Mohapatra et al. (6) reported 98.33%
for b- and 100% for a-thalassemia traits. The high utility
of FC OF for thalassemia screening makes this test very
promising, especially in communities with high frequen-
cy of a or b chains of hemoglobin mutations. It is easy
to perform and is much cheaper than electrophoresis of
hemoglobins or molecular diagnostics. The flow cyto-
metric osmotic fragility test could be referred when a
primary screening test—a complete blood count—
presents typical results for thalassemia. Interestingly,
prolonged incubation time on FC–OF did not influence
the %RRBC in samples from thalassemic children. The
decreased osmotic fragility of red blood cells in thalasse-
mia was reported previously several times. This phe-
nomenon is explained by high surface area-to-volume
ratio in target cells widespread in peripheral blood of
thalassemic subjects (12,13). The difference in sensitivi-
ty and specificity of the FC–OF in thalassemia diagnosis
results mostly from the increased osmotic fragility of red
blood cells from the reference group (children with all
other anemias) after longer incubation in a hypoosmotic
environment and not the osmotic fragility of the studied
(thalassemic) group.
The flow cytometry-based tests for the screening of
red blood cells disorders, since eosin-50 -maleimide bind-
ing assay was introduced by King et al. (14), are still
under development. The osmotic fragility assessment
was easily transferred from spectrophotometry to flow
cytometry (4). Modifications of the analysis are continu-
ously under development (15). The present study
focused on the choice of the most suitable time of analy-
sis to obtain the most reliable results. What we found is
that the elongation of analysis time increases the sensi-
tivity of the test. It is also more reliable according to the
area under the curve obtained in the ROC analysis.
LITERATURE CITED
1. Warang P, Gupta M, Kedar P, Ghosh K, Colah R. Flow cytometric
osmotic fragility—An effective screening approach for red cell mem-
branopathies. Cytom B Clin Cytom 2011;80:186–190.
2. Crisp RL, Solari L, Vota D, Garcıa E, Miguez G, Chamorro ME,
Schvartzman GA, Alfonso G, Gammella D, Caldarola S, Riccheri C,
Vittori D, Venegas B, Nesse A, Donato H. A prospective study to assess
the predictive value for hereditary spherocytosis using five laboratory
Cytometry Part B: Clinical Cytometry
 FLOW CYTOMETRIC OSMOTIC FRAGILITY TEST IN CHILDREN
tests (cryohemolysis test, eosin-5’-maleimide flow cytometry, osmotic
fragility test, autohemolysis test, and SDS-PAGE) on 50 hereditary
spherocytosis families in Argentina. Ann Hematol 2011;90:625–634.
3. Bolton-Maggs PH, Langer JC, Iolascon A, Tittensor P, King MJ, Gen-
eral Haematology Task Force of the British Committee for Standards
in H. Guidelines for the diagnosis and management of hereditary
spherocytosis—2011 update. Br J Haematol 2012;156:37–49
4. Won DI, Suh JS. Flow cytometric detection of erythrocyte osmotic
fragility. Cytom B Clin Cytom 2009;76:135–141
5. Park SH, Park CJ, Lee BR, Cho YU, Jang S, Kim N, Koh KN, Im HJ,
Seo JJ, Park ES, Lee JW, Yoo KH, Jung HL. Comparison study of the
eosin-5’-maleimide binding test, flow cytometric osmotic fragility
test, and cryohemolysis test in the diagnosis of hereditary spherocy-
tosis. Am J Clin Pathol 2014;142:474–484
6. Mohapatra R, Warang P, Ghosh K, Colah R. Hemoglobinopathy
screening by osmotic fragility test based on flow cytometer or
naked eye. Cytom B Clin Cytom 2016;90:279–284
7. Agarwal AM, Liew MA, Nussenzveig RH, Sangle N, Heikal N, Yaish
H, Christensen R. Improved harmonization of eosin-5-maleimide
binding test across different instruments and age groups. Cytom B:
Clin Cytom 2016;90:512–516
8. Shim YJ, Won DI. Flow cytometric osmotic fragility testing does
reflect the clinical severity of hereditary spherocytosis. Cytom B
Clin Cytom 2014;86436–443.
Cytometry Part B: Clinical Cytometry
195
9. Tao YF, Deng ZF, Liao L, Qiu YL, Deng XL, Chen WQ, Lin FQ.
Evaluation of a flow-cytometric osmotic fragility test for heredi-
tary spherocytosis in Chinese patients. Acta Haematol 2016;135:
88–93.
10. Szmydki-Baran AA-SA, Gołe R biewska-Staroszczyk S, Stelmaszczyk-
Emmel A, G orska E, Potapinska O. Evaluation of long-term blood
sample storage and its influence on result of EMA binding test anal-
ysis. Initial report. Pediatr Pol 2009;84:423–425.
11. Farias MG, Freitas PA. Percentage of hyperdense cells: Automated
parameter to hereditary spherocytosis screening. Clin Biochem
2015;48:1341–1343.
12. Tao YF, Deng ZF, Liao L, Qiu YL, Chen WQ, Lin FQ. Comparison
and evaluation of three screening tests of hereditary spherocytosis
in Chinese patients. Ann Hematol 2015;94:747–751.
13. Wiwanitkit V. Resistance to fragility test of red blood cell in thalas-
semia and reduction of smotic force at cell surface. Irn J Med
Hypotheses Ideas 2009;3:1–2.
14. King MJ, Behrens J, Rogers C, Flynn C, Greenwood D, Chambers K.
Rapid flow cytometric test for the diagnosis of membrane
cytoskeleton-associated haemolytic anaemia. Br J Haematol 2000;
111:924–933.
15. Yamamoto A, Saito N, Yamauchi Y, Takeda M, Ueki S, Itoga M,
Kojima K, Kayaba H. Flow cytometric analysis of red blood cell
osmotic fragility. J Lab Autom 2014;19:483–487