GLP-1 Receptor Agonists and Longer-Term Outcomes After Shoulder Replacement
A preoperative GLP-1 receptor agonist signal has now extended into shoulder arthroplasty. In a retrospective cohort drawn from the TriNetX global database, patients who used these agents within 30 days before total shoulder arthroplasty were followed for early medical complications and longer-term surgical outcomes, with the strongest separation emerging only after 2 years.
The study focused on adults who underwent total shoulder arthroplasty between 2008 and 2022. Investigators used one-to-one propensity score matching to balance baseline demographics and comorbidities, a necessary step in an observational analysis where diabetes, obesity, cardiovascular disease, and other chronic conditions can strongly shape both medication exposure and postoperative risk. After matching, the 90-day cohort included 1,951 patients, while the 2-year cohort included 1,917 patients.
Short-term results were flat. Within 90 days, GLP-1 receptor agonist use did not separate the cohorts for medical complications, readmission, emergency department utilization, or all-cause mortality. The absence of an early safety or efficacy signal is clinically important because it suggests these agents were not associated with an obvious perioperative penalty in the immediate postoperative window examined here.
The longer follow-up told a different story.
Two-Year Divergence in Emergency Department Use and Mortality
Among matched patients followed for 2 years, preoperative GLP-1 receptor agonist use was associated with lower emergency department utilization and lower all-cause mortality. Emergency department visits occurred in 26.7% of the GLP-1 receptor agonist group versus 31.0% of the comparison group, yielding a relative risk of 0.86 with a 95% confidence interval of 0.78 to 0.95 and a _p_ value of 0.012. All-cause mortality was 1.8% versus 4.3%, respectively, corresponding to a relative risk of 0.43 with a 95% confidence interval of 0.29 to 0.63 and a _p_ value of 0.008.
Those are not trivial differences. In a postoperative population, fewer emergency department encounters may reflect better chronic medical stability, fewer decompensations of metabolic disease, or less downstream healthcare use after the surgical episode has passed. The mortality signal is even more striking, but it also demands caution because death after arthroplasty can be influenced by the full burden of cardiometabolic illness, not just the operation itself. An observational association can align with clinical plausibility without proving causation.
Why a Metabolic Drug Might Affect Arthroplasty Recovery
The biological rationale is credible. GLP-1 receptor agonists have demonstrated metabolic benefits, and their anti-inflammatory and cardiovascular effects may matter long after a joint replacement incision has healed. In patients undergoing arthroplasty, improved glycemic control, reduced systemic inflammation, and better cardiovascular risk profiles could all translate into fewer medical crises during recovery and the broader postoperative interval.
Shoulder arthroplasty places special demands on patients with diabetes and other chronic illnesses. Pain control, rehabilitation adherence, wound healing, and general cardiometabolic reserve all interact during recovery. A medication class that lowers inflammatory tone and improves metabolic physiology could plausibly reduce the chance of later destabilization, even if it does not change the immediate 90-day postoperative course. The separation at 2 years, rather than 90 days, fits that kind of delayed systemic effect better than a direct surgical effect on the implant itself.
What the Neutral Early Results Mean for Surgeons
The 90-day null finding deserves equal attention. No differences emerged in the early medical complications examined, and no signal appeared for readmission, emergency department use, or mortality in that shorter interval. For surgeons and perioperative teams, that pattern argues against expecting a rapid protective effect from preoperative GLP-1 receptor agonist exposure during the immediate postoperative phase.
It also avoids an overinterpretation trap. A medication linked to lower long-term emergency department use and mortality after shoulder arthroplasty should not be treated as a perioperative substitute for standard optimization, blood pressure management, diabetes care, infection prevention, or rehabilitation planning. The data support an association, not a directive to prescribe GLP-1 receptor agonists solely to improve arthroplasty outcomes.
Observational Strengths and the Limits of Causal Inference
Propensity matching strengthens internal balance, but it cannot eliminate unmeasured confounding. Patients taking GLP-1 receptor agonists may differ in healthcare engagement, weight trajectories, adherence to chronic medications, or access to multidisciplinary care. Those factors can improve outcomes independently of the drug itself and are not fully captured in a database study.
The study also leaves several clinically relevant questions unanswered. The exact indication for GLP-1 receptor agonist use was not dissected here, and the analysis did not establish whether the benefits were concentrated among patients with type 2 diabetes, obesity, or other high-risk phenotypes. The database design also cannot determine whether the timing, duration, or specific agent within the GLP-1 receptor agonist class altered the observed associations. Without prospective confirmation, the mortality result should be viewed as hypothesis-generating rather than practice-changing.
Still, the pattern is hard to ignore. A drug class developed for metabolic disease may be exerting measurable downstream effects in orthopedic surgery populations, not by changing the mechanics of implantation, but by altering the physiology that surrounds recovery. That possibility now deserves targeted prospective testing in patients undergoing total shoulder arthroplasty, where the gap between early postoperative neutrality and later divergence may prove clinically important.