Separating thalassemia trait and iron deficiency by even simpler inspection.

To the Editor I read with interest the correspondence relating to recent articles published in the Journal describing mathematical approaches to the differentiation of thalassemia trait (TT) and iron deficiency anemia (IDA). 1-3 I found this and the accompanying authors' responses to be thoughtful and informative. There has been a clear revival of interest in the application of discriminant functions to these areas as demonstrated by the increasing number of publications in recent years. Doubtless this reflects the interest in population screening, and from a UK perspective, in the antenatal and newborn screening programs. Ntaios 2 admirably summarizes the key issue in that mathematical formulas and simple examination of indices work extremely well only in classical cases of TT and IDA. The biologic variability seen in both conditions, and especially when TT is complicated by coexisting IDA, as is common in some populations and highlighted by Rathod, 3 or in the presence of another hemoglobinopathy such as hemoglobin (Hb) E/TT makes this relationship considerably more complex, and a numeric approach alone will be insufficient to accurately resolve these cases. An alternative approach used in preference by our laboratory is visual analysis of the cytograms produced by our hematology analyzers, specifically the Advia 120 (Siemens Medical Solutions Diagnostics, Tarrytown, NY). Classical TT and IDA exhibit characteristic RBC cytogram profiles as shown in ❚Figure 1❚. Figure 1A shows the triangular shape of the RBC population tending toward more hypochromia than microcytosis commonly seen in IDA. Conversely, Figure 1B shows the comma-shaped RBC population resulting from an excess of microcytosis rather than hypochromia so characteristic of TT and often referred to as " Cooley's comma " in recognition of the eponymous physician who first described the disorder in 1925. Subtle perturbations in the appearance of these cytograms often hint at a compound erythropoietic disorder that numeric approaches are neither sufficiently sensitive nor specific enough to detect. Cytograms of combined TT/IDA and HbE/TT and such conditions as HbH disease, HbE trait, and others differ from classical TT and IDA alone and often sufficiently from each other to alert to the requirement for further investigation. 4 My own view is that cytogram analysis should form part of routine hematology practice because it can often significantly extend the " reach " of a traditional CBC in many areas of hematology. Although the examples offered here are only applicable to the Advia 120 and 2120 series analyzers, …