Blood cell parameters for screening and diagnosis of hereditary spherocytosis

Background There is currently no single index for the diagnostic screening of hereditary spherocytosis (HS). However, hematology analyzers are widely used in hospital laboratories because of their highly automated performance and quality control procedure, and detection of some blood cell parameters may be useful for the early screening of HS. Methods We investigated the values of blood cell parameters for the screening and differential diagnosis of HS. We performed a descriptive study of 482 samples (67 cases of HS, 59 cases of G6PD deficiency, 57 cases of AIHA, 199 cases of thalassemia, and 100 cases of healthy controls) that were run on Beckman Coulter LH780 Hematology Analyzer. Results HS was characterized by increased MCHC, decreased MRV, MSCV‐MCV < 0, and increased Ret with no concomitant increase in IRF. The areas under the ROC curves were MSCV‐MCV (0.97; 95% CI 0.95‐1.0) > MRV (0.94; 95% CI 0.91‐0.97) > MCHC (0.92; 95% CI 0.88‐0.97) > Ret/IRF (0.77; 95% CI 0.7‐0.84). MSCV‐MCV ≤ 0.6 fl was valuable parameter for the diagnostic screening of HS, with a sensitivity of 95.5% and specificity of 94.9%. Conclusion These indices have high reference values for differentiating HS from thalassemia, AIHA, and G6PD deficiency.

while Wang et al 7 found that 18% of 140 pediatric patients with hemolytic anemia had HS, a proportion only exceeded by autoimmune hemolytic anemia. Diagnosis of HS depends on family medical history, clinical manifestations, and laboratory examinations.
Laboratory tests used to diagnose HS include red cell morphology examination, blood cell parameters, red cell osmotic fragility test, red cell membrane protein measurement, and membrane protein mutation detection. HS is characterized by an increased number of microspherocytes in the peripheral blood, with hyperchromatic small red cells being of particularly high diagnostic value. The presence of microcytic erythrocytes > 7.8% can be used for HS diagnosis with a sensitivity of 56.7% and specificity of 84.8%. 8 However, red cells shrink and are misshapen in patients with autoimmune hemolytic anemia (AIHA), and different numbers of microspherocytes may be present, thus reducing the specificity of red cell morphology for HS diagnosis. The current laboratory tests used to screen HS include the NaCl osmotic fragility test, acidified glycerol lysis test (AGLT), and sucrose hemolysis test. However, the NaCl osmotic fragility test has low sensitivity of 48%-95% because of different types of membrane protein deficiencies, making it unsuitable for screening of mild HS. 9 The sensitivity of the AGLT50 is higher than the NaCl osmotic fragility test; however, although AGLT50 is relatively short in patients with HS with a diagnostic sensitivity of 95%, 10 it is also shortened in patients with AIHA. 11 Regarding the sucrose hemolysis test, hemolysis > 2.8% has a diagnostic sensitivity of 78.7% and specificity of 95.3% for HS, 12 but is a complex procedure. A next-generation osmotic gradient ektacytometry (NG-OGE) assay was recently introduced for the screening of HS. The NG-OGE is useful for distinguishing HS from other hereditary red cell membrane disorders; however, it does not differentiate between HS and AIHA. 13,14 Unfortunately, the new generation of ektacytometer is difficult to be applied in the clinical laboratory, mainly used in some research centers or specialized laboratories. Red blood cell membrane protein detection tests include the eosin-5'-maleimide binding test and sodium dodecyl sulfate-polyacrylamide gel electrophoresis test. The former has a sensitivity of 93% and specificity of 98% for the diagnosis of HS, 11 and the latter is more specific for protein 4.2-and ankyrin-deficient HS, but less specific for asymptomatic and mild HS, and is thus unsuitable for determining the type of membrane protein deficiency in approximately 10% of HS patients. 15 Gene mutations in HS are scattered, and no specific mutation sites suitable for screening have yet been identified.
Hematology analyzers are widely used in hospital laboratories because of their highly automated performance and quality control procedures, and some blood cell parameters may be suitable for early screening of HS. Mean corpuscular hemoglobin concentration (MCHC) was included as one of the diagnostic criteria for HS in guideline for the diagnosis and management of HS in 2004, 16 and mean sphered corpuscular volume (MSCV) has also been reported to have diagnostic value in HS. [17][18][19] Furthermore, we recently found that mean reticulocyte volume (MRV) was a valuable index for the diagnosis of HS. 20 In this study, we investigated the values of blood cell parameters including MRV, MSCV, MCHC, and absolute reticulocyte count (Ret)/immature reticulocyte fraction (IRF) for the screening and differential diagnosis of HS.

| Equipment and methods
No therapeutic measures were given to patients in the HS group or the disease control group. Two milliliters of fasting venous blood was taken from each subject and transferred to an anticoagulation tube containing 10% EDTA-K2 (1.5 mg/mL) for analysis of red blood cell and reticulocyte parameters using a Coulter LH780 Hematology Analyzer (Beckman Coulter Inc, Fullerton, CA, USA).
During analysis, methylene-blue-stained red blood cells formed deproteinated spherocytes after treatment with an acidic hypoosmotic solution. The spherocytes could then be further divided into mature red blood cells and reticulocytes using the volume, conductivity, and light scatter technology. The mean volume of the spherocytes was measured as MSCV, and MCV, MCHC, MRV, and Ret of red blood cells and reticulocytes were detected simultaneously. The IRF was calculated by measuring the fluorescence intensity of RNA left in the reticulocytes using a Coulter LH780 Hematology Analyzer.
Measurement data were expressed as mean ± standard deviation, and means of each parameter were compared between samples by one-way analysis of variance. Normally distributed data were compared by least significant difference and non-normally distributed data by Tamhane's T2 tests. A level of P < 0.05 was considered statistically significant. The area under the receiver operating characteristic (ROC) curve was calculated, and the diagnostic values of MCHC, MRV, MSCV-MCV, and Ret/IRF for the diagnosis of HS were compared. The optimal diagnostic cutoff point for HS and the corresponding sensitivity and specificity were determined by analyzing the sensitivity and misdiagnosis rate of the ROC curve.

| RE SULTS
Blood cell parameters in the HS, disease control, and normal control groups are shown in Table 1   ces with wide application prospects for the differential diagnosis of HS from G6PD deficiency, THAL, and AIHA. Given the widespread use of fully automated hematology analyzers, these blood cell parameters represent potentially valuable tools for the preliminary screening and differential diagnosis of HS.

ACK N OWLED G M ENTS
This study was supported by the National Natural Science