A-373 Creatinine Measurement Using the Dried Blood Microsampling Device Capitainer®B50: Analytical Validation, Correction Modeling, and Stability Evaluation

Abstract

Creatinine is an essential biomarker for assessing kidney function, used widely in the diagnosis and monitoring of acute kidney injury, chronic kidney disease, and treatment related nephrotoxicity. Reliable quantification is therefore critical. Microsampling technologies offer major advantages, including minimally invasive collection, reduced sample volume, simplified logistics, and suitability for decentralised or at home sampling. However, for microsampling to be adopted clinically, analytical reliability, agreement with reference methods, and sample stability throughout transport and storage must be assured. For creatinine quantification the microsampling device Capitainer®B50 (Capitainer AB) This quantitative device collects 2 x 50µl dried whole blood into two precut paper discs. Two elution volumes (250?µL and 500?µL) were evaluated using a Methanol based elution buffer on two instrument platforms (cobas® Pro, Roche Diagnostics GmbH and Thermo Scientific™ Indiko™ Clinical Chemistry Analyzer). For each instrument–volume combination, linear regression was performed against the reference plasma Creatinine concentration. Agreement after correction was assessed using Bland–Altman analysis. In total 160 comparisons (Plasma Creatinine vs Microsampling-Creatinine) was used in the study. Analytical performance of the two platforms was compared by evaluating linearity, residual dispersion, and limits of agreement at both elution volumes. Capillary-sample validation using the optimal method: Based on optimal elution buffer and instrument was selected for correction-factor derivation. The resulting correction model was applied to 16 capillary whole-blood microsamples to assess commutability with venous EDTA plasma. Imprecision was assessed at four creatinine concentrations (˜0.19, 0.39, 0.57 and 1,13 mg/dL; six replicates per level) for both elution volumes. Stability testing was performed only on Cobas Pro, evaluating Day?9 room temperature storage, heat-stress cycle, and cold-stress cycle, with a predefined acceptance criterion of ±15% deviation from baseline. Across all four instrument–volume combinations, linearity was high, with R² ranging from 0.953 to 0.997. Intercepts spanned –1.84 to +1.41 The 250?µL protocol consistently outperformed 500?µL by demonstrating markedly lower residual variation and narrower Bland–Altman limits of agreement, indicating substantially reduced measurement dispersion. Indiko with the 250?µL elution protocol showed the lowest residual SD and narrowest LOA of all evaluated configurations and thus was selected as the optimal method for correction-factor derivation. Also, precision was consistently superior at 250?µL. with a mean CV of 7.10% (range 3.37–9.85%) compared to a mean CV 13.14% (range 3.64–18.78%) for the 500?µL protocol: Application of the validated correction model to capillary microsamples produced small positive bias?0.05 mg/dL, SD?0.06?mg/dL. These results confirm commutability between capillary microsamples and venous EDTA plasma after correction. All storage and stress conditions demonstrated high robustness, with 90–95% of measurements within the ±15% acceptance limit. Mean deviations ranged from –3.61% to +4.58% across Day?9 RT, heat cycle, and cold cycle. Capillary microsamples exhibit strong agreement with plasma after regression-based correction, supporting clinical commutability. Stability data confirm that the microsampling workflow remains reliable under typical transport and environmental stress conditions.