LAUSR

AIM To compare renal function by several GFR formulas (particularly cystatin C eGFR-"CAPA") in relation to renal risk drugs (RRDs) in patients admitted to two geriatric wards in a university geriatric department. MATERIALS AND METHODS This was a prospective quality improvement study including 108 patients, 2/3 women, age ≥ 75 years, admitted with multimorbidity. Renal function tests were performed with Cockcroft & Gault with uncalibrated (C&Guc) and calibrated creatinine (C&Gcc), and 3 - 4 points' iohexol clearance (mGFR) in mL/min, and eGFR with MDRD4, CKD-EPI, CAPA, and BIS2 clearance in mL/min/1.73m2. Agreement was tested by Bland & Altman analysis. The number and type of RRDs were analyzed. RESULTS Measured GFR, C&Gcc, and C&Guc were mean 37, 39, and 32 mL/min, respectively. Estimated GFR by MDRD4, CKD-EPI, CAPA, and BIS2 were mean 56, 52, 45, and 40 mL/min/1.73m2, respectively. Compared to mGFR, women had significantly higher clearance for all estimates except for C&Gcc and C&Guc. C&Gcc, C&Guc, and BIS2 showed the lowest bias. 38 RRDs were identified. 96 patients used a mean of 2.3 RRDs per patient, and 1.7 RRDs needed dose adjustments. Cardiovascular drugs and analgesics were the most frequent RRDs. DISCUSSION The C&Gcc, C&Guc, and BIS2 equations gave the best estimate of kidney function in relation to mGFR for drug dosing in the elderly. The eGFR methods showed significantly higher clearance than mGFR, C&Gcc, C&Guc, and BIS2. RRDs that needed dose adjustment were common in this geriatric population. If the eGFR formulas (MDRD4, CKD-EPI, and CAPA) are used instead of C&Gcc, C&Guc, and BIS2, higher and potentially more risky doses of RRDs may be administered to geriatric patients over 75 years, women in particular.