Evaluation of Renal Function Assessment Methods across Different Drug Administration Designs

Le résumé fourni par la source

Background Glomerular filtration rate (GFR) with inulin is a standard indicator for assessing renal function, and measured creatinine clearance (mCLCr) is also used. As GFR assessment requires urine collection, renal function is sometimes estimated using serum creatinine levels with equations such as the Cockcroft-Gault equation (CG). However, the drug dosing in the package insert often contains inconsistent descriptions of estimation formulas and units for evaluating renal function for each drug. Additionally, the measured renal function values may differ from the estimated results. This retrospective study examined the agreement between the actual measured values and the results obtained from renal function estimation equations and assessed the accuracy of these equations. Methods Approved by the Institutional Review Board, a retrospective single-center analysis was conducted on hematopoietic stem cell transplantation recipients at our institution between January 1, 2019, and December 31, 2023. Agreement between mCLCr and the three estimation equations (CG enzyme method [enz], CG enzyme method + 0.2 correction [enz+0.2], and personalized estimated GFR [eGFR]) and between converted GFR (cGFR) and the three estimation equations was assessed using Bland–Altman analysis. Factors influencing the relationship between the measured and estimated values were identified using multiple regression analysis. The predictive accuracy of each estimation equation was calculated as |Estimated equation - Measured value|/ Measured value × 100 (%). Equations with deviations greater than 30% from the actual measurements were classified as overestimating or underestimating renal function. Results The study included 59 males and 34 females. The age at transplantation was 54.0 (18–70 years). Bland–Altman analysis showed that the CG (enz) tended to agree with the mCLCr, with a limit of agreement (LOA) of approximately 30 mL/min. The predictive accuracy of each estimation equation was over 80% for mCLCr with the CG (enz) and personalized eGFR and for cGFR with CG (enz+0.2) and personalized eGFR. Multiple regression analysis revealed that the three estimation equations and skeletal muscle mass were significant factors for mCLCr and cGFR (P <0.01). Conclusions CG (enz) provided the most consistent results when assessing renal function using CLCr, whereas CG (enz+0.2) was the most consistent with GFR. In addition, it is better to pay attention to skeletal muscle mass when assessing by eGFR. However, when using the estimation equation, it should be noted that an LOA of 30 mL/min.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.