Abstract
BACKGROUND: Modifications in Jaffe serum creatinine (sCr) assays question the suitability of the results for direct comparison. METHODS: sCr in adult in-patients was routinely measured either by SRM 909-standardized/noncompensated (method A) or isotope dilution mass spectrometry traceable/compensated method (reference). We converted values by method A into values by the reference using a formula provided by the manufacturer [Beckman Coulter (BC)] and traditional equating methods. RESULTS: The BC-based conversion and linear equating resulted in underestimated sCr values, whereas equipercentile equating (EE) provided sCr with not significantly different distribution from the reference values. Proportions of patients with renal impairment did not differ between the reference and EE-converted sCr, as opposed to BC-recalculated values. Three percent of patients were classified into better renal function category when applying BC versus EE conversion. CONCLUSIONS: Equipercentile equation was a more accurate method for recalculation of sCr obtained from different Jaffe reaction assays than the linear equating or the BC linear formula. This study emphasizes the importance of the derivation sample specificity when applying research results to other real-world populations.