GLP-1 RAs Reduce Joint Replacements, In-Hospital Mortality in Osteoarthritis

Use of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) is associated with significantly lower odds of knee replacement among hospitalized adults with osteoarthritis (OA). This is according to research presented at ACR Convergence 2026, held November 6 to 11 in Orlando, Florida.

Additionally, GLP-1 RA use was also associated with lower crude in-hospital mortality among patients hospitalized with OA or rheumatoid arthritis (RA).

The retrospective, cross-sectional study, led by Naina Kumari, MD, of the Brooklyn Hospital Center in Brooklyn, New York, evaluated data from the National Inpatient Sample (NIS) spanning 2016 through 2023. The research team assessed hospitalizations among adults with a primary or secondary ICD-10-CM diagnosis of OA (22,868,684 weighted hospitalizations) or RA (4,463,113 weighted hospitalizations), stratifying cohorts by concurrent GLP-1 RA therapy alongside type 2 diabetes (T2D) or obesity. Overall, GLP-1 RA use was documented in 0.11% of OA-related and 0.13% of RA-related hospital encounters.

Among patients with OA, GLP-1 RA users were younger on average compared with nonusers (mean age, 65.9 vs 70.4 years) and were more frequently covered by private insurance (26.9% vs 19.7%).

These findings suggest benefits extending beyond glycemic control, potentially mediated through anti-inflammatory, weight-modifying, and cardioprotective mechanisms.

Following multivariable survey-weighted logistic regression adjusting for sociodemographic factors and clinical comorbidities, GLP-1 RA use in the OA cohort was associated with a 31% reduction in the odds of knee replacement (adjusted odds ratio [aOR], 0.69).

In-hospital mortality was significantly lower in unadjusted analyses among OA patients treated with GLP-1 RAs compared with nonusers (0.35% vs 1.60%); however, mortality was omitted from the multivariable logistic regression model because the coefficient was not estimable due to perfect prediction (zero events in the GLP-1 RA group).

GLP-1 RA use was also associated with a shorter mean length of stay (4.02 vs 4.43 days). However, mean total charges were higher among GLP-1 RA users ($84,174 vs $63,947), which the authors noted likely reflects a heavier baseline comorbidity burden. Crude hip replacement rates were also lower with GLP-1 RA exposure (5.3% vs 10.4%), but this association did not retain statistical significance after multivariable adjustment (aOR, 0.90).

In the RA cohort, GLP-1 RA use demonstrated parallel unadjusted reductions in in-hospital mortality (0.34% vs 2.76%) — which was likewise excluded from adjusted models — hip replacement rates (1.34% vs 2.58%), knee replacement rates (1.93% vs 3.30%), and mean hospital stay (4.49 vs 5.29 days). However, adjusted odds ratios for hip replacement (aOR, 0.47) and knee replacement (aOR, 0.69) did not achieve statistical significance, which the authors attributed to limited statistical power given the small GLP-1 RA subgroup size.

“These findings suggest benefits extending beyond glycemic control, potentially mediated through anti-inflammatory, weight-modifying, and cardioprotective mechanisms,” the study authors concluded. “As GLP-1 RA prescribing continues to expand rapidly across cardiometabolic indications, prospective and longitudinal studies are warranted to better define their role in modifying disease trajectory and reducing surgical burden in arthritis populations.”