Ann Hematol DOI 10.1007/s00277-010-1112-0
ORIGINAL ARTICLE
A prospective study to assess the predictive value for hereditary spherocytosis using five laboratory tests (cryohemolysis test, eosin-5′-maleimide flow cytometry, osmotic fragility test, autohemolysis test, and SDS-PAGE) on 50 hereditary spherocytosis families in Argentina Renée L. Crisp & Liliana Solari & Daiana Vota & Eliana García & Gabriela Miguez & Maria E. Chamorro & Gabriel A. Schvartzman & Graciela Alfonso & Daniel Gammella & Sergio Caldarola & Cecilia Riccheri & Daniela Vittori & Belen Venegas & Alcira Nesse & Hugo Donato
Received: 8 July 2010 / Accepted: 21 October 2010 # Springer-Verlag 2010
Abstract This prospective study was carried out to assess the usefulness of five laboratory tests in the diagnosis of hereditary spherocytosis (HS), based on the correlation of erythrocyte membrane protein defects with clinical and laboratory features, and also to determine the membrane protein deficiencies detected in Argentina. Of 116 patients and their family members tested, 62 of them were R. L. Crisp : G. Alfonso División Hematología Clínica, Departamento de Medicina, Hospital Nacional Alejandro Posadas, Buenos Aires, Argentina G. Alfonso e-mail:
[email protected] R. L. Crisp : D. Vota : M. E. Chamorro : D. Vittori : A. Nesse Laboratorio de Análisis Biológicos, Departamento de Química Biológica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina D. Vota e-mail:
[email protected] M. E. Chamorro e-mail:
[email protected] D. Vittori e-mail:
[email protected] A. Nesse e-mail:
[email protected] R. L. Crisp : G. A. Schvartzman : H. Donato Consultorios de Hematología Infantil, Buenos Aires, Argentina
diagnosed to have HS. The specificity of cryohemolysis (CH) test was 95.2%, and its cut-off value to distinguish HS from normal was 2.8%. For flow cytometry, cut-off points of 17% for mean channel fluorescence (MCF) decrease and 14% coefficient of variation (CV) increase showed 95.9% and 92.2% specificity, respectively. Both tests showed the highest percentages of positive results for diagnosis. Either G. A. Schvartzman e-mail:
[email protected] H. Donato e-mail:
[email protected] L. Solari : D. Gammella : B. Venegas Laboratorio de Citometría, Departamento de Diagnóstico, Hospital Nacional Alejandro Posadas, Buenos Aires, Argentina L. Solari e-mail:
[email protected] D. Gammella e-mail:
[email protected] B. Venegas e-mail:
[email protected] E. García : C. Riccheri División Hematología y Oncología Pediátrica, Departamento de Pediatría, Hospital Nacional Alejandro Posadas, Buenos Aires, Argentina E. García e-mail:
[email protected]
Ann Hematol
CH or flow cytometry was positive in 93.5% of patients. In eight patients, flow cytometry was positive only through CV increase. Protein defects were detected in 72.3% of patients; ankyrin and spectrin were the most frequently found deficiencies. The CV of the fluorescence showed significantly higher increases in moderate and severe anemia than in mild anemia (p=0.003). Severity of anemia showed no other correlation with tests results, type of deficient protein, inheritance pattern, or neonatal jaundice. CH and flow cytometry are easy methods with the highest diagnostic accuracy. Simultaneous reading of mean channel fluorescence (MCF) decrease and CV increase improve diagnostic usefulness of flow cytometry. This test seems to be a reliable predictor of severity. The type of detected protein deficiency has no predictive value for outcome. Predominant ankyrin and spectrin deficiencies agree with reports from other Latin American countries. Keywords Spherocytosis . Hemolytic anemia . Red blood cell membrane . Anemia . Hereditary anemia
Introduction Hereditary spherocytosis (HS) is the most common hereditary anemia excluding beta thalassemia in Argentina. Clinical manifestations of HS vary widely, ranging from asymptomatic patients to severely anemic, transfusiondependent patients [1]. Severity has been related to inheritance pattern and protein defects. Typical HS presents no diagnostic problem. In most cases, the association between a positive family history, physical findings (e.g., splenomegaly, jaundice), laboratory evidence of extravascular hemolysis, visualization of spherocytes on the blood smear, and a negative direct antiglobulin test usually
C. Riccheri e-mail:
[email protected] G. Miguez : S. Caldarola Servicio de Medicina Transfusional, Departamento de Diagnóstico, Hospital Nacional Alejandro Posadas, Buenos Aires, Argentina G. Miguez e-mail:
[email protected] S. Caldarola e-mail:
[email protected] R. L. Crisp (*) División Hematología Clínica, Departamento de Medicina, Hospital Nacional Alejandro Posadas, Fernandez de Enciso 4602, 1419 Buenos Aires, Argentina e-mail:
[email protected]
confirms the diagnosis [1–4]. However, in many patients, the diagnosis may be difficult due to several reasons: spherocytosis can be common to all chronic hemolytic anemias, a positive family history of HS is not evident in around 25% of patients, and specificity and sensitivity of confirmatory diagnostic tests are variable. Mild HS may be difficult to diagnose because patients may have normal hemoglobin and bilirubin levels and very few or even absent spherocytes on the blood smear [1]. Severe HS may present, in addition to spherocytes, red cells with diverse morphology (acantocytes, echinocytes, mushroom-shaped erythrocytes, poikilocytes, stomatocytes), raising the problem of differential diagnosis with other hemolytic anemias [1]. In such cases, it is essential to perform confirmatory tests for the diagnosis. However, none of them can detect all HS patients [1, 2, 4–7]. Osmotic fragility (OF), which gives positive results in around 80% of patients [8, 9], is still considered as the gold standard test by some authors [1, 6, 10] conducted simultaneously with autohemolysis (AH) test, although it cannot differentiate true HS from spherocytosis secondary to other diseases (autoimmune hemolytic anemia, ABO blood group incompatibility in newborns, hereditary spherocytic elliptocytosis) [3, 10, 11]. Two recently described tests—hypertonic cryohemolysis (CH) and eosin-5′-maleimide flow cytometry (EMA-FC)— seem to have a similar or even higher sensitivity and specificity for diagnosis of HS [4, 7, 11–15]. Densitometric quantification of membrane proteins separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDSPAGE) allows detecting isolated or combined deficiency of spectrin, ankyrin, band 3, protein 4.1, and/or protein 4.2. Nevertheless, the usefulness of the type of protein deficiency or results of diagnostic tests as predictive factors of clinical features has been poorly studied and remains controversial [1, 2, 4–6]. The aims of this study were (a) to evaluate diagnostic accuracy of CH and EMA-FC in relation to standard screening tests used in our laboratory (OF and AH) and SDS-PAGE; (b) to assess the relationship between specific membrane protein defects, clinical, and laboratory features; and (c) to determine prevalence of each membrane protein defect in Argentine people. No previous report concerning these subjects has been published in our country.
Material and methods Committees of Ethics and Research approved the study protocol. Informed consent was obtained from healthy blood donors, patients, or parents (in case of children) before entering the protocol. The study was performed according to Helsinki international ethical standards on human experimentation.
Ann Hematol
Patients A prospective study on 50 families was carried out between January 2007 and January 2010. A total of 68 individuals with presumptive diagnosis of HS and 48 asymptomatic relatives (parents and/or siblings) were studied. Age ranged from 14 days to 77 years; 49 individuals were under 18 years of age. Ten patients were already splenectomized at the time of study. Healthy blood donors (n=263) were used as normal controls. Methods Clinical history, physical examination, and results of laboratory tests were evaluated for the diagnosis of HS. The following tests were performed: complete blood cell count, blood smear examination, reticulocytes count, DAT, unincubated and incubated OF tests, AH test, hypertonic CH test, EMA-FC method, and SDS-PAGE of red cell membrane proteins. Blood samples from patients or relatives were collected simultaneously with samples from six normal controls, and processed within 48 h. None of patients had been transfused within the 2 months preceding the blood sampling. Blood cell counts were determined on an automated hematology analyzer (Sysmex XT 1800, Roche Diagnostics). Blood smears were independently examinated by two experts (RLC and HD). OF and AH tests were performed using the usual standard methods [16, 17]. CH test was performed according to the method described by Streichman et al. [12] slightly modified. Differences with the original procedure were the use of manual shaking instead of vortex and a sucrose concentration slightly lesser than the original method (0.6 M vs. 0.7 M, respectively). To avoid contamination of the reactive material, buffered sucrose was obtained at the time of use, adding 10 ml of 50 mM phosphate buffer pH 7.4 to 2.4 g sucrose and yielding a solution density 1.092 g/ml (0.62 M). EMA-FC test was performed with the following method. Red blood cells were washed thrice with phosphatebuffered saline (PBS) pH 7.4. Washed packed cells (5 μl) were incubated with 5 μl of 5′EMA (0.5 mg/ml PBS) in the dark at 20°C for 15 min. The excess of EMA was removed by washing the cells three times with PBS. Finally, the red cell pellet was suspended in 1.0 ml PBS. The fluorescence intensity was determined as the geometric MCF and its coefficient of variation (CV) for 20,000 events in the FL1 channel using a FacScalibur cytometer and CELLQuest software (Becton Dickinson). Results were expressed as percentage of decrease of MCF and increase of CV in relation to the means of the same parameters obtained in six simultaneously performed control samples.
Ghost preparation and electrophoresis of membrane proteins was carried out using the method described by Vittori et al. [18]. Triplicate lanes for the patient and six different normal controls were subjected to SDS-PAGE in the slab mini gel (gel size 7 cm×8 cm×1 mm) using Laemmli buffer system. The monomer concentration (%T) in the slab mini gels was 4%, 0.125 M Tris–HCl, pH 6.8 in the stacking gel and 8%, 0.375 M Tris–HCl, pH 8.8 in the running gel. Samples were dissociated by mixing them and boiling for 1 min in SDS sample buffer (62.5 mM Tris– HCl, pH 6.8, containing 2% SDS, 5% 2-mercaptoethanol, 10% glycerol, and bromophenol blue as tracking dye). Amounts of 15 μg of protein were applied in each lane. Protein concentrations had been previously determined by the method developed by Lowry et al. [19]. Constant current of 20 mA during the isotachophoretic process in the stacking gel and 25 mA throughout the running gel were applied with running buffer Tris 0.025 M, glycine 0.192 M, SDS 0.1%, pH 8.3; the whole electrophoretic procedure lasted approximately 1 h. A molecular weight protein standard was including in each slab gel. At the end of running a staining using Coomassie brilliant blue, R-250 was developed. Amounts of spectrins, ankyrin, band 3, protein 4.1, and protein 4.2 in the stained gels were quantified by densitometry and analyzed with Gel Pro Analyzer 4.0. For each protein in each gel, a 90% prediction interval was performed using the six normal controls. Mean values of the studied subject were compared with those of reference samples. Protein deficiency was considered significant whenever the mean value of that protein level was lower than the minimum value of the prediction interval. Diagnostic criteria for HS were signs and symptoms of hemolytic anemia plus two positive diagnostic laboratory tests. According to the criteria by Mariani et al. [5], anemia was defined as severe (hemoglobin10 g/dl) in nonsplenectomized patients; healthy family members, showing any protein membrane defect but otherwise asymptomatic and with negative results for other diagnostic tests, were considered as HS silent carriers. Statistical analysis Statistical analysis was performed using software InfoSTAT (Release 2009, InfoStat Group, FCA, and University of Córdoba, Argentina). A receiver–operator characteristic (ROC) curve derived from a logistic regression analysis was used to determine an optimum cut-off point for discriminating between normal subjects and HS patients for CH and EMA-FC tests. Results of laboratory tests in relation to severity of anemia, inheritance pattern, neonatal jaundice, and protein membrane deficiency were evaluated
Ann Hematol
by nonparametric methods (Kruskal–Wallis test). Spearman’s rank correlation coefficient was used to evaluate relationships between sets of data. Results are presented as mean± SD or as median. Statistical significance was defined as a p value18 years 14 (51.9%) Patients
