The relationship between kidney function and body mass index before and after bariatric surgery in patients with chronic kidney disease.

Abstract

Improvements in kidney function post-bariatric surgery may be related to weight loss-independent effects. To characterize the dynamic relationship between body mass index (BMI) and estimated glomerular filtration rate (eGFR) before and after bariatric surgery in patients with chronic kidney disease (CKD). Kaiser Permanente Southern California (KPSC) health system. We conducted an observational, retrospective cohort study of patients with CKD stage 3 or higher who received bariatric surgery at the KPSC health system between 2007-2015. Bariatric surgery procedures included primary Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) procedures. Outcomes consisted of mean trajectory estimates and correlations of BMI and eGFR taken between 2 years before and 3 years after surgery. Multivariate functional mixed models were used to estimate how BMI and eGFR trajectories evolved jointly. A total of 619 RYGB and 474 SG patients were included in the final analytic sample. The measurements were available before surgery for a median time of 1.9 years for SG and 1.8 years for RYGB patients. Median follow-up times after surgery were 2.8 years for both SG and RYGB patients. The mean age at the time of surgery was 58 years; 77% of patients were women; 56% of patients were non-Hispanic White; the mean BMI was 44 kg/m2; 60% of patients had diabetes mellitus; and 84% of patients had hypertension. Compared to the presurgery eGFR declines, the postsurgery declines in eGFR were 57% slower (95% credible interval [CrI], 33%-81%) for RYGB patients and 55% slower (95% CrI, 25%-75%) for SG patients. The mean correlation between BMI and eGFR was negligible at all time points. Though bariatric surgery slowed declines in eGFR up to 3 years after surgery, changes in eGFR tracked poorly with changes in BMI. This study provides evidence that the kidney-related benefits of bariatric surgery may be at least partly independent of weight loss. Confirming this hypothesis could lead to mechanistic insights and new treatment options for CKD.