The influence of pharmaceutically induced weight changes on estimates of renal function: A patient-level pooled analysis of seven randomised controlled trials of glucose lowering medication.

BACKGROUND: Estimation of kidney function (eGFR) is essential in monitoring of patients with kidney disease. Estimates of kidney function based on serum creatinine are derived from cross-sectional studies. If body weight (BW) changes, this might affect creatinine and eGFR. The Cockcroft-Gault (CG) equation includes creatinine and BW, whereas the Modification of Diet in Renal Disease (MDRD) and Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations only include creatinine. METHODS: Data were pooled from the six LEAD (Liraglutide Effect and Action in Diabetes) trials and the LIRA-DPP4 trial. The trials were conducted in patients with type 2 diabetes and of 26weeks duration. We investigated changes in eGFR for patients treated with liraglutide, and for patients treated with glucose-lowering medications with less weight-reducing effects (insulin glargine, glimepiride, exenatide and rosiglitazone). RESULTS: We included 5100 patients (liraglutide n=3173, comparator n=1927). Mean (SD) CKD-EPI eGFR was 81.2 (20.6) ml/min/1.73m(2) for liraglutide and 81.6 (20.3) ml/min/1.73m(2) for comparator. For liraglutide, BW changed -1.9 (95% CI (-2.0; -1.8)) kg, for comparator BW changed 0.2 (95% CI (0.03; 0.3)) kg. Using regression modelling, a 10% BW decrease yielded no change in creatinine, MDRD eGFR or CKD-EPI eGFR for both liraglutide and comparator, but was associated with a 10.2% (-11.3%; -9.1%) decrease in CG eGFR for liraglutide, and a 10.6% (-12.0%; -9.1%) decrease for comparator. CONCLUSIONS: A liraglutide-induced weight reduction of 1.9kg was not associated with change in creatinine. Accordingly, there was no change in weight-independent estimates of GFR, whereas weight-dependent estimates were changed. The MDRD and CKD-EPI equations can be used in patients experiencing pharmaceutically induced weight reductions.