AAOS Reports: GLP‑1 RAs — Perioperative Associations and Musculoskeletal Safety Signals

AAOS 2026 press materials describe two retrospective, observational analyses examining GLP‑1 receptor agonists (including semaglutide) among adults with obesity or type 2 diabetes undergoing common orthopaedic procedures. The meeting-presented summaries frame the findings as associations from database analyses rather than evidence of causation. One analysis addresses perioperative patterns and short-term postoperative outcomes drawn from claims data, while the other evaluates five-year diagnosis signals using an electronic medical record–derived dataset. Taken together, the release juxtaposes near-term postoperative associations with longer-term musculoskeletal diagnosis comparisons as reported by the investigators.

In the first analysis, the press materials report a national claims–based retrospective cohort spanning 2010–2023 that identified patients with obesity-related diagnosis codes undergoing 10 common orthopaedic surgeries. Investigators described a consistent increase in preoperative GLP‑1 and semaglutide use across procedures, with a marked rise since 2019. As summarized, GLP‑1 use was associated with significantly lower odds of postoperative emergency department visits across multiple major procedures, and it was also reported to be associated with lower surgical site infection rates among patients undergoing total hip arthroplasty and total knee arthroplasty. The materials also describe a procedure-specific pattern in revision outcomes—reported as lower revision rates for total knee arthroplasty among GLP‑1 users but higher revision rates among patients undergoing carpal tunnel release. The release characterizes these as short-term, procedure-specific associations observed in claims data.

The second analysis is described as an EMR-derived, matched retrospective cohort in adults with type 2 diabetes and obesity using a large, multi-institutional electronic medical record–derived database, with matched cohort sizes reported as 73,483 patients and balanced baseline characteristics after matching. Over five years, investigators reported a higher incidence of osteoporosis among GLP‑1 RA users compared with matched controls (4.1% vs 3.2%; RR 1.29, 95% CI 1.22–1.36; p<0.001) and a higher incidence of gout (7.4% vs 6.6%; RR 1.12, 95% CI 1.08–1.16; p<0.001). The greatest reported relative risk increase was for osteomalacia (2% vs 0.1%; RR 2.55, 95% CI 1.83–3.55; p<0.001). The press materials state that these observed differences were statistically significant and emphasize the observational nature of the comparisons.

Presenter commentary in the press materials repeatedly frames both studies as meeting-presented, retrospective findings intended to inform ongoing examination of musculoskeletal outcomes associated with GLP‑1 RA exposure. In the long-term analysis summary, the primary investigator is quoted as saying that when patients are prone to osteoporosis, gout, or osteomalacia, clinicians “should consider bone health surveillance and monitor for delayed-onset complications in at-risk populations,” characterizing such steps as feasible within routine care. The same commentary situates the findings in the context of emerging follow-up windows for widely adopted medications and includes a call for continued research to better understand longer-term effects on bone and joint health. Overall, the press materials frame the results as observations from retrospective datasets and note that longer-term musculoskeletal effects of GLP‑1 RA exposure remain poorly understood.

Key Takeaways:

• AAOS meeting materials described increasing GLP‑1/semaglutide use and reported associations with selected short-term postoperative outcomes in claims data.
• Procedure-specific associations were reported for surgical site infection and revision patterns across total hip arthroplasty/total knee arthroplasty and carpal tunnel release.
• A matched EMR analysis reported higher five-year rates of osteoporosis, gout, and osteomalacia among GLP‑1 RA users, alongside investigator comments about monitoring, presented as an author statement.