Toward the harmonization of result presentation for the eosin-5'-maleimide binding test in the diagnosis of hereditary spherocytosis.

The eosin-5'-maleimide (EMA) Binding test measures reduced mean channel fluorescence (MCF) reading of EMA-labeled red cells (EMA-RBCs) from patients with hereditary spherocytosis (HS). Reporting test results can be either in the actual MCF reading or as a ratio by normalization of the test MCF result to the mean MCF value of six normal controls. The latter format has potential for universal reporting. We analyzed three years' archival MCF data from HS and non-HS patient groups for establishment of reference ranges of ratios for normal adults and HS. A prospective study used FC500 and FACS Canto II cytometers to analyze contemporaneously EMA-RBCs from several patient groups and normal donors. Statistical analyses of the prospective data determined the cut-off values, and the sensitivity and specificity for HS respectively for the MCF and the ratio result presentations. The effect of using fewer than six normal controls for the ratio denominator was explored. The FC500 gave a mean ratio of 0.782 (SD=0.086) in HS patients with an optimal cut-off ratio of 0.918 (98.7% specificity, 95.6% sensitivity), and gray area ratio of 0.868-0.918. The Canto II gave a mean ratio of 0.774 (SD=0.085) with an optimal cut-off ratio of 0.925 (97.1% specificity and 100% sensitivity), and gray area ratio of 0.859-0.925. Harmonization of result presentation is feasible with no apparent constraint by instrument design. Interpretation of gray-area data requires an assessment of patient's clinical presentation and family history or performing a family study.