Predictors of Discrepancies Between Cockcroft-Gault and Chronic Kidney Disease-EPI Creatinine-Cystatin in Hospitalized Patients

Cockcroft-Gault (eCrCl) remains the industry standard for estimating renal function in hospitalized patients despite known limitations. Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine-cystatin C (eGFRcr-cys) equation was developed to ameliorate these limitations and is now preferred in specific outpatient scenarios. This study aims to determine patient-specific factors associated with a >25 milliliters per minute (mL/min) difference where eCrCl exceeds eGFRcr-cys. This retrospective, Institutional Review Board approved, observational review included data from June 2021 to December 2022. Adults with serum cystatin C and creatinine collected within 24-hours of admission were included. Exclusions included pregnancy or renal replacement therapy. The primary endpoint was to identify patient-specific predictors of a >25 mL/min discrepancy between eCrCl and eGFRcr-cys. The secondary outcome was to determine odds ratios for identified patient-specific characteristics. Of 235 encounters, 67 had a >25 mL/min difference where eCrCl exceeded eGFRcr-cys. Multivariate regression identified the following independent predictors of this difference: age 105 kg (OR: 9.079; 95% CI: 4.344-18.978), intensive care unit (ICU) admission within 24 hours (OR 2.779; 95% CI 1.248-6.188) and diagnosis of paraplegia or quadriplegia (OR 5.490; 95% CI 1.523-19.796). This study concluded that eCrCl may exceed eGFRcr-cys by >25 mL/min in patients presenting for hospital admission with one or more predictors: age 105 kg, intensive care unit (ICU) admission, or diagnosis of paraplegia or quadriplegia. Current literature suggests improved accuracy of renal clearance estimations with eGFRcr-cys. Accordingly, practitioners should consider eGFRcr-cys when evaluating new admissions with these characteristics.