BACKGROUND: Evidence on whether glucagon-like peptide 1 receptor agonists (GLP-1RA) influence outcomes after total knee arthroplasty (TKA) is mixed. We evaluated the association between perioperative use of GLP-1RA and postoperative outcomes in patients with and without diabetes.
METHODS: In this retrospective cohort from a single U.S. health system (2019-2023), adults undergoing primary TKA were stratified by GLP‑1RA use at the time of surgery. Primary outcomes included length-of-stay (LOS), 30-day medical complications, 90-day readmissions and surgical site infection (SSI), 1-year SSI, medical complications, and TKA implant complications, and revision TKA. Multivariable logistic regression and negative binomial regression were adjusted for age group, sex, BMI, smoking status, diabetes, and the Elixhauser Comorbidity Index. Unadjusted subgroup analyses examined outcomes by diabetes status and specific GLP-1RA agent.
RESULTS: Among 26,154 TKA patients, 914 (3.5 %) used a GLP-1RA (73 % of these had diabetes). GLP-1RA use was associated with 45 % lower odds of 1-year implant complications (odds ratio = 0.55, 95 % CI 0.33-0.90, P = 0.02) and 10 % shorter LOS (incidence rate ratio = 0.90, 95 % CI 0.85-0.94, P < 0.001) (absolute difference of 0.17 days). No significant differences in 30-day complications, 90-day SSI, or readmissions were observed after adjustment.
CONCLUSIONS: Preoperative GLP-1RA use was associated with reduced one-year implant complications and slightly shorter hospital stays following TKA. These findings, although encouraging, are exploratory. Larger, adjusted analyses are needed to confirm the benefits before recommending changes in perioperative management.
