GLP-1 Receptor Agonist Outcomes, Safety, and Body Mass Index Change in a National Cohort of Patients on Dialysis

Key Points GLP-1 receptor agonists (GLP-1 RAs) in diabetes and dialysis are associated with 23% lower mortality and 66% higher chance of transplant waitlisting. GLP-1 RAs are not associated with increased risk of acute pancreatitis, biliary complications, or medullary thyroid cancer. GLP-1 RAs are associated with a 32% increased risk of diabetic retinopathy in patients with diabetes on dialysis. Background Of the 808,000 US patients on dialysis, 60% have diabetes and are eligible for glucagon-like peptide-1 receptor agonists (GLP-1 RAs). Safety and outcomes in this population are unknown. We sought to examine GLP-1 RA real-world safety, efficacy, and weight loss in people with diabetes on dialysis. Methods In this observational national cohort study (2013–2021), we identified adults with type 2 diabetes on dialysis. The exposure of interest was GLP-1 RA use. Body mass index (BMI) change after dialysis initiation was quantified among patients with two measurements ( N =6474). Extended Cox models with inverse probability of treatment weights (censoring for kidney transplant waitlisting) were used to quantify all-cause mortality associated with GLP-1 RAs. Specific safety outcomes (acute pancreatitis, biliary complications, medullary thyroid cancer, and diabetic retinopathy) were assessed. Results The study included 151,649 patients on incident dialysis with type 2 diabetes. Mean BMI and weight change among GLP-1 RA users were greater than those among nonusers (−1.47 versus −0.61 kg/m 2 ; −4.03 versus −1.47 kg; P < 0.001 for both). The mortality incidence rate was lower among GLP-1 RA users (219.0 versus 279.5 patients/1000 person-years; P < 0.001). GLP-1 RA use was associated with a 23% lower risk of mortality (adjusted hazard ratio [aHR], 0.77; 95% confidence interval [CI], 0.70 to 0.85; P < 0.001); results were consistent among initiates with BMI ≥30 kg/m 2 . GLP-1 RA use was associated with a 66% higher chance of waitlisting (aHR, 1.66; 95% CI, 1.28 to 2.13; P < 0.001). There was an increased association with diabetic retinopathy (aHR, 1.32; 95% CI, 1.12 to 1.56; P = 0.001), but not with any other safety outcomes. Inferences were consistent across multiple sensitivity analyses. Conclusions GLP-1 RA use in patients with type 2 diabetes on dialysis was associated with weight loss, reduced mortality risk, and increased likelihood of kidney transplant waitlisting. These real-world data are the strongest evidence to date supporting GLP-1 RA use in this population.