Reevaluating the Icterus Index Cutoff of a Jaffe Creatinine Method

Background: The aim of this study was to redefine the icterus index cutoff for the Roche Jaffe creatinine method using both conjugated and unconjugated bilirubin on 3 Roche cobas modules (c311, c501, and c701/c702) at laboratories across our hospital network. Methods: Interference was evaluated by adding conjugated bilirubin (as bilirubin conjugate, ditaurate) and unconjugated bilirubin to pooled remnant plasma. The effects of conjugated and unconjugated bilirubin were tested separately to assess the contribution of each species. The magnitude of interference was calculated as both absolute and percentage error with total allowable error limits set at 0.1 mg/dL (8 mu mol/L or 8%). Results: Analysis of interference data across the 3 Roche modules did not show bias exceeding our total allowable error limits for plasma creatinine up to a conjugated bilirubin icterus index of 16.2 (approximately 16.2 mg/dL or 277 mu mol/L) or an unconjugated bilirubin icterus index of 18.5 (approximately 18.5 mg/dL or 316 mu mol/L), the highest concentrations tested. Conclusions: Our results demonstrate that the Roche Jaffe method exhibits acceptable performance in the presence of icterus at icterus indexes above the manufacturer's current recommendations of 5 (approximately 5 mg/dL or 86 mu mol/L) and 10 (approximately 10 mg/dL or 171 mu mol/L) for conjugated and unconjugated bilirubin, respectively. We have updated the icterus index in our hospital system to 16 for conjugated bilirubin and 18 for unconjugated bilirubin.