Red Cell Indices and Functions Differentiating Patients with the β-thalassaemia Trait from those with Iron Deficiency Anaemia

Transcription

1 The Journal of International Medical Research 2009; 37: [first published online as 37(1) 4] Red Cell Indices and Functions Differentiating Patients with the β-thalassaemia Trait from those with Iron Deficiency Anaemia V OKAN, A CIGILOGLU, S CIFCI, M YILMAZ AND M PEHLIVAN Gaziantep University School of Medicine, Department of Haematology, Gaziantep, Turkey This study examined the diagnostic accuracy of nine indices to discriminate between patients with mild-to-moderate (haemoglobin g/dl) or moderateto-severe (haemoglobin < 8.5 g/dl) iron deficiency anaemia (IDA) from those with b-thalassaemia (b-tt) (n = 100 per group). Indices examined were red blood cell (RBC) count, RBC distribution width (RDW), Mentzer index (MI), Shine and Lal index (S&L), England and Fraser index (E&F), Srivastava index (S), Green and King index (G&K), RDW index (RDWI), and Ricerca index (R). Index sensitivity, specificity, and positive and negative prognostic values were examined. Youden s indices were calculated and showed: S&L > G&K > E&F > RBC = RDWI > MI > S > R > RDW to differentiate between b-tt and mild-to-moderate IDA; and S&L > G&K > E&F = RDWI > RBC > R > MI > S> RDW to differentiate between b-tt or moderate-to-severe IDA. For both groups, S&L and G&K offered the best discrimination and RDW the worst. S&L showed the highest Youden index for b-tt and IDA discrimination, but sensitivity and specificity were not 100%. In both mild and severe IDA, the S&L index may be used to differentiate cases of b-tt from IDA cases, but large clinical trials are needed to explore this further. KEY WORDS: IRON DEFICIENCY ANAEMIA; b-thalassaemia TRAIT; DISCRIMINATION INDICES Introduction The most frequent causes of microcytic anaemia are iron deficiency anaemia (IDA) and the β-thalassaemia trait (β-tt). Thalassaemia occurs in Turkey at a frequency of 2.1%, whereas in the Turkish town of Gaziantep it occurs at a frequency of 1.8%. 1,2 When diagnosis is based solely on data obtained by full blood count devices available in polyclinics, IDA and thalassaemia are confused, and many thalassaemia cases receive inappropriate iron treatment. Many erythrocyte indices of automatic blood count devices have been reported to discriminate between thalassaemia and IDA patients. 3 8 These studies generally contained only mildto-moderate IDA patients among the IDA cases. The aim of the current study, however, was to obtain nine erythrocyte discrimination indices for cases of both moderate-to-severe 25

2 IDA and mild-to-moderate IDA, and to seek to use these indices to discriminate between β-tt and IDA patients using simple full blood count data. Patients and methods PATIENTS This cross-sectional study comprised consecutive male and female patients with mild-to-moderate IDA (haemoglobin [Hb] g/dl), moderate-to-severe IDA (Hb < 8.5 g/dl) or the β-thalassaemia trait (n = 100 per group) who had been referred to the haematology outpatient clinic of Gaziantep University School of Medicine. A diagnosis of IDA was based on low serum ferritin levels (< 10 ng/ml). A diagnosis of β-tt was made when hypochromic microcytosis, normal serum ferritin levels and elevated serum levels of HbA 2 (> 4.0%) were found. Patients having conditions such as acute and chronic inflammatory diseases, infections, hypothroditis, acute bleeding, or any kind of malignancies that might effect ferritin or haemoglobin levels were excluded. All patients provided written informed consent, in accordance with the Ethical Principles for Human Investigations outlined in the Helsinki Declaration. BLOOD ANALYSES AND INDICES Complete blood counts were performed using an automated analyser (Sysmex XT- 2000i ; Roche Diagnostics, Indianapolis, IN, USA). Serum levels of HbA 2 were determined using Chromsystems CLC 200 devices (Chromsystems, Munich, Germany), as directed by the manufacturer, or by highperformance liquid chromatography. The following were obtained for each patient: (i) red blood cell (RBC) count; (ii) RBC distribution width (RDW); (iii) Mentzer index (MI); (iv) Shine and Lal index (S&L); (v) England and Fraser index (E&F); (vi) Srivastava index (S); (vii) Green and King index (G&K); (viii) RDW index (RDWI); and (ix) Ricerca index (R). The threshold values of these parameters that were used to differentiate between IDA and β-tt are shown in Table 1. The sensitivity and specificity, and positive and negative predictive values, were calculated as follows: (i) sensitivity = true positive/(true positive + false negative); (ii) specificity = true negative/(true negative + false positive); (iii) positive predictive value (PPV) = true positive/(true positive + false positive); and (iv) negative predictive value (NPV) = true negative/(true negative + false TABLE 1: Threshold values of the indices used to discriminate between iron deficiency anaemia (IDA) and the b-thalassaemia trait (b-tt) Indices IDA b-tt Red blood cell (RBC) count < 5 > 5 RBC distribution width > 14 <14 Mentzer index (MI) = MCV/RBC > 13 < 13 Shine and Lal (S&L) index = MCV 2 MCH 0.01 > 1530 < 1530 England and Fraser (E&F) index = MCV RBC 5Hb 3.4 Positive Negative Srivastava index (S) = MCH/RBC > 3.8 < 3.8 Green and King (G&K) index = MCV 2 RDW/100Hb > 65 < 65 RBC distribution width index (RDWI) = MCV RDW/RBC > 220 < 220 Ricerca (R) index = RDW/RBC > 4.4 < 4.4 MCV, mean cell volume; MCH, mean corpuscular haemoglobin; Hb, haemoglobin. 26

3 negative). Youden s index was calculated as (sensitivity + specificity) 100. STATISTICAL ANALYSIS The means ± SD were calculated for patients demographic data. The indices data were analysed using Microsoft Office Excel Results This study included 100 β-tt cases (50 women and 50 men) with a mean ± SD age of 31.8 ± 11.8 years (range years), 100 mild-to-moderate IDA cases (14 men and 86 women) aged 37.3 ± 13.1 years (range years) and a further 100 moderate-to-severe IDA cases (15 men and 85 women) aged 37.7 ± 16.9 years (range years). The sensitivity, specificity, PPV, NPV and Youden s index for the parameters that were used to discriminate between cases of β-tt and mild-to-moderate IDA and between cases of β-tt and moderate-to-severe IDA are listed in Tables 2 and 3, respectively. For TABLE 2: Predictive values of evaluated indices for patients with mild-to-moderate iron deficiency anaemia (IDA; Hb g/dl) and patients with the b-thalassaemia trait (b-tt) Sensitivity Specificity PPV NPV Youden s Indices (%) (%) (%) (%) index Red blood cell (RBC) count IDA β-tt RBC distribution width (RDW) IDA β-tt Mentzer index (MI) IDA β-tt Shine and Lal (S&L) index IDA β-tt England and Fraser (E&F) index IDA β-tt Srivastava (S) index IDA β-tt Green and King (G&K) index IDA β-tt RBC distribution width index (RDWI) IDA β-tt Ricerca (R) index IDA β-tt Sensitivity = true positive/(true positive + false negative); specificity = true negative/(true negative + false positive); positive predictive value (PPV) = true positive/(true positive + false positive); negative predictive value (NPV) = true negative/(true negative + false negative); Youden s index = (sensitivity + specificity)

4 TABLE 3: Predictive values of evaluated indices for patients with moderate-to-severe iron deficiency anaemia (IDA; Hb < 8.5 g/dl) and patients with the b-thalassaemia trait (b-tt) Sensitivity Specificity PPV NPV Youden s Indices (%) (%) (%) (%) index Red blood cell (RBC) count IDA β-tt RBC distribution width (RDW) IDA β-tt Mentzer index (MI) IDA β-tt Shine and Lal (S&L) index IDA β-tt England and Fraser (E&F) index IDA β-tt Srivastava (S) index IDA β-tt Green and King (G&K) index IDA β-tt RBC distribution width index (RDWI) IDA β-tt Ricerca (R) index IDA β-tt Sensitivity = true positive/(true positive + false negative); specificity = true negative/(true negative + false positive); positive predictive value (PPV) = true positive/(true positive + false positive); negative predictive value (NPV) = true negative/(true negative + false negative); Youden s index = (sensitivity + specificity) 100. differentiating patients with β-tt or mild-tomoderate IDA, the Youden index showed: S&L > G&K > E&F > RBC = RDWI > MI > S> R > RDW and, for differentiating patients with β-tt or mild-to-moderate IDA, the Youden index showed: S&L > G&K > E&F = RDWI > RBC > R > MI > S > RDW. The S&L and G&K indices were, therefore, the best indicators in both comparisons and the RDW index was the worst. Discussion Several studies have derived discriminatory functions based on RBC indices that can be used to differentiate between patients with IDA and those with β-tt. 3 8 The differential diagnosis of these hypochromic microcytic anaemias is an everyday concern for physicians; accurate diagnosis is essential for appropriate treatment, prevention of disease and minimization of expense. The aim of the 28

5 current study was, therefore, to compare the sensitivity, specificity, PPV and NPV of nine indices and to calculate Youden s indices in order to rank the differentiating indices and be able to determine which of them might best differentiate between IDA and β-tt cases. Unlike previous comparative studies which only included IDA patients with Hb g/dl, the present study also included IDA patients with Hb <8.5 g/dl. No index has yet been found to offer both sensitivity and specificity of 100%. Youden s index provides a measure of validity by considering both sensitivity and specificity, and was first used by Demir et al. 9 For IDA patients with Hb g/dl and β-tt the top four indices ranked according to Youden s index were found to be S&L > G&K > E&F > RDWI = RBC and for IDA patients with Hb < 8.5 g/dl and β-tt the top four indices were S&L > G&K > E&F = RDWI. The S&L index was first defined by Shine and Lal in and was reported to have a sensitivity of 100%, a specificity of 11.8% and an efficiency value of 59.5% for differentiating between β-tt and IDA patients. Yeo et al. 10 found the S&L index and mean cell volume (MCV) to be applicable when applied to pregnant women in Singapore. Lafferty et al. 11 found the S&L index, MI and MCV to be valuable in distinguishing IDA and β-tt minor cases; the RDW and the E&F indices were not useful. In studies including schoolchildren in Jordan, MI, MCV 72 fl, and E&F and S&L indices correctly identified 91.6%, 82.4%, 81.3% and 62.6%, respectively, of microcytosis cases as having or not having the β-tt trait 12. Using the Youden index, Demir et al. 9 found RBC and RDWI to be of good predictive value in paediatric patients; they set the Youden index value of the S&L index to zero. Beyan et al. 13 determined the Youden index values to be 73.7 and 65.5 for the RBC and G&K indices, respectively, in a study comparing IDA and β-tt cases; the S&L index ranked six out of nine indices, at Ntaios et al. 14 found the Youden index values for the G&K and E&F indexes to be and 63.23, respectively, in the differentiation of β-tt and IDA patients; the RDW index had a Youden index value of 7.32 with the lowest discriminatory power. The S&L index was not used in this study. AlFadhli et al. 15 found the E&F index to be the most discriminatory and the S&L index the least when comparing patients with IDA to those with α-tt or TT minor. The present study found the S&L and G&K indices to be best at differentiating IDA from β-tt patients, and the RDW index to be the worst. In particular, the S&L index had the highest Youden index value in discriminating β-tt cases from those with moderate-to-severe IDA and also from those with mild IDA. To the best of our knowledge this is the first work to study patients with moderate-to-severe IDA; other research has typically studied mild-to-moderate IDA. Further work with more patients is required. Conflicts of interest The authors had no conflicts of interest to declare in relation to this article. Received for publication 8 September 2008 Accepted subject to revision 16 September 2008 Revised accepted 25 November 2008 Copyright 2009 Field House Publishing LLP References 1 Cavdar AO, Arcasoy A: The incidence of β- thalassemia and abnormal hemoglobins in Turkey. Acta Haematol 1971; 45: Gurbak M, Sivasli E, Coskun Y, et al: Prevalence and hematological characteristics of β- 29

6 thalassemia trait in Gaziantep urban area, Turkey. Pediatr Hematol Oncol 2006; 23: Shine I, Lal S: A strategy to detect β- thalassaemia minor. Lancet 1977; i: England JM, Bain BJ, Fraser PM: Differentiation of iron deficiency from thalassaemia trait. Lancet 1973; i: Green R, King R: A new red cell discriminant incorporating volume dispersion for differentiating iron deficiency anemia from thalassemia minor. Blood Cells 1989; 15: Ricerca BM, Storti S, d Onofrio G, et al: Differentiation of iron deficiency from thalassaemia trait: a new approach. Haematologica 1987; 72: Srivastava PC, Bevington JM: Iron deficiency and-or thalassaemia trait. Lancet 1973; i: Mentzer WC Jr: Differentiation of iron deficiency from thalassaemia trait. Lancet 1973; i: Demir A, Yarali N, Fisgin T, et al: Most reliable indices in differentiation between thalassemia trait and iron deficiency anemia. Pediatr Int 2002; 44: Yeo GS, Tan KH, Liu TC: The role of discriminant functions in screening for betathalassaemia traits during pregnancy. Singapore Med J 1995; 36: Lafferty JD, Crowther MA, Ali MA, et al: The evaluation of various mathematical RBC indices and their efficacy in discriminating between thalassemic and non-thalassemic microcytosis. Am J Clin Pathol 1996; 106: Sirdah M, Tarazi I, Al Najjar E, et al: Evaluation of the diagnostic reliability of different RBC indices and formulas in the differentiation of the β-thalassaemia minor from iron deficiency in Palestinian population. Int J Lab Hematol 2008; 30: Beyan C, Kaptan K, Ifran A: Predictive value of discrimination indices in differential diagnosis of iron deficiency anemia and betathalassemia trait. Eur J Haematol 2007; 78: Ntaios G, Chatzinikolaou A, Saouli Z, et al: Discrimination indices as screening tests for β- thalassemic trait. Ann Hematol 2007; 86: AlFadhli SM, Al-Awadhi AM, AlKhaldi D: Validity assessment of nine discriminant functions used for the differentiation between iron deficiency anemia and thalassemia minor. J Trop Pediatr 2007; 53: Author s address for correspondence Associate Professor Vahap Okan Gaziantep University School of Medicine, Department of Haematology, Gaziantep, Turkey. vahapokan@yahoo.com 30