Assessment of sereum creatine-based equations for estimating glomerular filtration rate in critically ill patients with or without augmented renal clearance

Farm Hosp. 2026 Mar 24:S1130-6343(26)00030-9. doi: 10.1016/j.farma.2026.02.018. Online ahead of print.

ABSTRACT

BACKGROUND: Serum creatinine-based equations are commonly used to estimate glomerular filtration rate (eGFR) in critically ill patients, despite not having been specifically developed for this population. This study aimed to assess and compare the performance of three widely used serum creatinine-based equations in this setting.

METHODS: Observational retrospective study conducted in four intensive care units of a tertiary university hospital. The most commonly used serum creatinine-based equations, the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), Modification of Diet in Renal Disease (MDRD-4), and Cockcroft-Gault (CG) equations were compared to the creatinine clearance from 24-hour urine collection (CrCl_24h), used as the reference method. Bland and Altman plots, bias and precision were performed to contrast CrCl_24h values with estimated GFR. Bias and its 95% confidence interval were calculated as the mean difference between the eGFR estimated by each equation and the measured CrCl_24h. Precision was reported as one standard deviation of the bias.

RESULTS: A total of 261 patients were included. In patients with CrCl_24h between 0-129 mL/min, no significant differences were observed between equations. However, in patients with augmented renal clearance (15.7%), with a mean CrCl_24h of 180 mL/min, there was a statistically significant bias between the CKD-EPI equation (66.4 mL/min/1,73m²) and both the CG and MDRD-4 equations (24.8 mL/min/m² and 29.3 mL/min/1,73 m², respectively; p < 0.01).

CONCLUSIONS: This study highlights that the most commonly used equations to estimate glomerular filtration rate in critically ill patients have remarkable limitations compared to creatinine clearance from 24-hour urine collection. In critically ill patients with CrCl_24h between 0-129 mL/min, no significant differences were found between the CG, MDRD-4, and CKD-EPI equations. However, for patients with augmented renal clearance, the CG and MDRD-4 equations performed statistically better than the CKD-EPI equation.

PMID:41881746 | DOI:10.1016/j.farma.2026.02.018