Enzymatic creatinine versus alkaline picrate: evidence from a multicenter human-pool study

Background: Despite isotope dilution mass spectrometry (IDMS) traceability, creatinine harmonization across platforms remains imperfect, particularly for Jaffe methods at physiological concentrations. Leveraging patient-like human serum pools measured across multiple routine platforms, we quantified method-dependent differences and their impact on derived estimated glomerular filtration rate (eGFR). Methods: Fifteen human serum pools were measured in triplicate across major vendors (Roche, Abbott, Beckman, Siemens; enzymatic and Jaffe; Werfen, Ortho enzymatic only). Triplicates were averaged per pool and platform. We summarized (I) cross-platform dispersion [per-pool coefficient of variation (CV)] within method families, (II) within-vendor relative bias (Jaffe vs. enzymatic), (III) concentration dependence, and (IV) eGFR impact using Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) 2021 (creatinine-only) in fixed scenarios (a 60-year-old female and a 60-year-old male). Results: Cross-platform dispersion was lower for enzymatic vs. Jaffe [median CV 3.21% (2.47–3.49%) vs. 5.05% (2.94–6.72%)]. Vendor-wise median relative bias (Jaffe vs. enzymatic) across pools: Roche +5.84%, Abbott +4.64%, Beckman −1.52%, Siemens +10.85%. Bias tended to be larger at low creatinine concentration. Depending on vendor and patient example (60-year-old female or male), method-dependent differences between Jaffe and enzymatic creatinine led to Kidney Disease: Improving Global Outcomes (KDIGO) G-stage reclassification in 7–40% of serum pools, with absolute ΔeGFR values up to 12–13 mL/min/1.73 m2 near decision thresholds. Conclusions: Enzymatic creatinine assays exhibit better inter-platform alignment and smaller method-dependent bias than Jaffe procedures, particularly in the low-concentration range, where small absolute differences may carry large clinical leverage thus reducing the risk of clinically meaningful eGFR shifts.