Accuracy of Red Blood Cell Parameters in Predicting α⁰-Thalassemia Trait Among Non-Anemic Males

Background/Objectives: Red blood cell (RBC) parameters are routinely used to screen for α- and β-thalassemia traits as part of prenatal diagnosis for severe fetal thalassemia in countries with a high prevalence of the disease. In clinical practice, the same cut-off values for these parameters are applied to both females and males. However, given that the normal reference ranges for some RBC parameters differ significantly between sexes, sex-specific cut-off values may be more appropriate, especially in combination. To date, the effectiveness of RBC indices in males for predicting α- and β-thalassemia traits has not been evaluated. The objectives of this study are to assess the diagnostic performance of individual and combined RBC parameters in detecting α⁰-thalassemia traits among non-anemic males. Methods: This diagnostic study is a secondary analysis of prospectively collected data from our project on prenatal control of severe thalassemia. The study population comprised male partners of pregnant women who underwent thalassemia screening during their first antenatal visit. RBC parameters, including hemoglobin (Hb), hematocrit (Hct), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), and RBC count, were measured for each participant. Carrier status for the α0-thalassemia Southeast Asian (SEA) genotype was confirmed by using a polymerase chain reaction (PCR)-based method. The diagnostic performance of each RBC parameter and their combinations, based on predictive models generated using logistic regression, was evaluated and compared using receiver operating characteristic (ROC) curves. Results: A total of 486 Thai males were recruited for the study, including 137 individuals with the α⁰-thalassemia trait and 349 with a normal α-thalassemia genotype (control group). All RBC parameters, except for Hct, differed significantly between the two groups. Among the individual indices, MCH exhibited the highest diagnostic accuracy, followed by MCV, with areas under the curve (AUCs) of 0.981 and 0.973, respectively. An MCH cut-off value of 26 pg and an MCV cut-off value of 80 fL provided a sensitivity of 100% for both indices, with specificities of 88.5% and 86.8%, respectively. The combination predictive model provided the best diagnostic performance, achieving an AUC of 0.987, which was slightly but significantly higher than that of any individual parameter. This model yielded a sensitivity of 100% and a significantly higher specificity of 90.8% at a cut-off probability of 7.0%. Conclusions: MCH and MCV demonstrated excellent screening performance for identifying α0-thalassemia carriers in males. However, the combination model exhibited even greater accuracy while reducing the false-positive rate. Implementing this model could minimize the need for unnecessary PCR testing, leading to substantial cost savings.