Objective: Analysis of the interference of ketoacids on various routine plasma creatinine assays during a clinical episode of diabetic ketoacidosis (DKA).
Design: Observational study. Blood samples were drawn before, during and after standard in-hospital treatment. Plasma creatinine was measured with two dissimilar enzymatic assays (creatininase PAP + and creatinine iminohydrolase Serapak), a kinetic alkaline picrate method (Jaffé) and a high-performance liquid chromatography (HPLC) procedure. Acetoacetate and beta-hydroxybutyrate were analysed by enzymatic methods.
Setting: Department of Medicine, University Hospital.
Subjects: Nine patients who experienced 10 episodes of DKA.
Main outcome measures: Agreement of the routine plasma creatinine assays with HPLC and analysis of possible interferents.
Results: At presentation, the Jaffé assay gave falsely high values of plasma creatinine (median 99 micromol L(-1)), in contrast to the PAP+ (median 60.5 micromol L(-1)) and HPLC assays (median 67.5 micromol L(-1)). This positive error decreased during treatment. This was due to a decrease in acetoacetate, as the positive error by the Jaffé method correlated with the acetoacetate concentration (r = 0.79, P < 0.0001). In the multiple regression analysis, beta-hydroxybutyrate caused no additional interference by the Jaffé assay, confirmed by in vitro experiments. Analysis of agreement showed that the difference between PAP+ and HPLC creatinine was -4.6 +/- 3.0 micromol L(-1) (mean +/- SD), and 2.0 +/- 5.3 micromol L(-1) between Serapak and HPLC. This was statistically significant, but clinically negligible.
Conclusion: Acetoacetate caused severe interference of the alkaline picrate (Jaffé) assay, which might influence therapeutic decisions at the start of diabetic ketoacidosis. Enzymatic assays lack this interference.