Use of Blood Indices for Β Thalassemia Trait Screening with a Population in Western India

Context: β thalassemia is one of the most common hereditary single gene disorders. β thalassemia trait (BTT) causes mild hypochromic microcytic anemia (HMA), so erythrocyte indices are used to identify BTT presumptively followed by the confirmation with the abnormal elevation of hemoglobin A2 (HbA2). Mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and formulas based on erythrocyte indices are used to differentiate BTT from other causes of HMA. The discriminatory cut-offs vary across regions due to different prevalence rates of disorders causing HMA and the mutation frequency of BTT, hence programs should derive local cut-offs. Objectives: Develop receiver operating curves (ROC) for BTT screening specific to the population of Western India, using MCH and MCV values. Methods: This retrospective analysis (2007-2016) used data of 397,554 individuals. Complete blood count (CBC) and hemoglobin (Hb) analysis were done by the three-part differential cell counter (Sysmex K-4500, Japan) and VARIANT IITM (Bio-Rad Laboratories, USA) respectively. Sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of MCV and MCH, using MCV \u3c78 fl or MCH \u3c 27 pg/cell as the cut-off, were calculated for six groups (males and females of age \u3c 1 year, ≥ 1 year & \u3c 10 years, and ≥ 10 years). The definitive diagnosis of BTT was made based on a HbA2 value of ≥ 3.5 and ≤ 7% in absence of other abnormal hemoglobins. A new range of HbA2 was derived from the BTT cases confirmed with genetic mutation analyses (n = 963). ROC curves were created with new HbA2 value and optimal cut-offs for MCV and MCH were derived. Sensitivity, specificity, accuracy, PPV, and NPV were calculated and compared with the previous values. Results: HbA2 values were higher in males compared to females in the BTT group (p \u3c 0.001). Sensitivity, specificity, NPV and accuracy improved with new cut-offs. There was no improvement in PPV. Specificity was low in 1 to10-year group. Conclusion: BTT screening program should derive gender specific HbA2 values. While deriving optimum cut-offs of MCV and MCH using ROC, one should consider the cost of false positive and negative results also