1 Introduction

The global rise in obesity and diabetes mellitus (DM) has profound implications for spinal surgery, as these metabolic disorders have been independently linked to elevated perioperative morbidity, prolonged recovery, and increased healthcare costs.^1,2 Over 80% of patients undergoing spinal procedures are classified as overweight, obese, or diabetic, underscoring the urgency for targeted perioperative optimization strategies in this population.³ Obesity and hyperglycemia synergistically exacerbate postoperative risks, including surgical site infections (SSIs), delayed wound healing, venous thromboembolism (VTE), and revision surgeries, with associated costs exceeding those of non-comorbid cohorts by 30–50%.^4–7 Consequently, preoperative interventions aimed at weight reduction and glycemic control have become integral to mitigating complications, particularly in elective spine surgery.⁸ For example, morbidly obese patients undergoing bariatric surgery before thoracolumbar fusion demonstrate significantly lower rates of deep vein thrombosis (DVT), mechanical failure, and neurologic deficits compared to patients without perioperative intervention.⁹ Similarly, stringent glycemic control, evidenced by hemoglobin A1c levels <6%, correlates with reduced infection rates, fewer readmissions, and improved fusion success.^10–12 Emerging evidence also suggests that pharmacological agents, such as metformin, may offer additional benefits through pleiotropic anti-inflammatory mechanisms, potentially leading to improved postoperative outcomes.¹³

However, the perioperative impact of glucose-lowering therapies remains heterogeneous. Certain medications, including thiazolidinediones and sodium-glucose cotransporter-2 inhibitors, raise concerns due to adverse effects on bone metabolism, fracture risk, and calcium-phosphate homeostasis, potentially compromising spinal fusion and hardware integration.¹⁴ As a result, glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have garnered attention for their dual utility in managing DM and obesity while exhibiting a favorable safety profile. Originally approved for type 2 diabetes, agents such as semaglutide and liraglutide now hold FDA indications for chronic weight management, supported by evidence of sustained weight loss (10–15% body weight reduction) and cardiometabolic benefits.^1,2,15 Preclinical studies further suggest that GLP-1 RAs may also enhance bone mineral density, osteoblast activity, and trabecular microarchitecture, potentially mitigating osteoporosis-related complications in spinal fusion.^16–19 Despite these promising attributes, clinical data evaluating GLP-1 RA use in elective lumbar spine procedures as a collective remains unexplored.

This gap in knowledge is clinically significant considering that approximately 42.8% of U.S. adults are diagnosed with obesity, 9.5% with diabetes, and the rates of elective lumbar surgeries are rising. Furthermore, about 12.5% of American adults have been prescribed glucagon-like peptide-1 receptor agonists (GLP-1 RAs), with 6% actively maintained on these medications.^1,20–22 Postoperative complications following lumbar procedures, including pseudoarthrosis, hardware failure, SSIs, and VTE, disproportionately burden metabolically compromised patients, extending rehabilitation timelines and elevating reoperation rates to 18%.^23–26 While GLP-1 RAs have demonstrated favorable perioperative safety profiles in total joint arthroplasty, reducing 90-day adverse events by 22–34% in prior studies,^27,28 their effects on spine-specific outcomes remain unexplored. Important outcomes such as medical complications, effectiveness of bone fusion, revision surgery rates, 90-day readmissions, and the risk of surgical site infections in spinal patients have not been thoroughly investigated.

This study aims to address this gap in the literature and evaluate the association between GLP-1 RA use and postoperative outcomes in patients undergoing elective lumbar arthrodesis, facetectomy, foraminotomy, laminectomy, laminotomy, or disc/joint excision, comparing medical and mechanical complications between GLP-1 RA users and non-users. We hypothesize that GLP-1 RA use will correlate with equivalent or superior outcomes, attributable to its beneficial effects on weight, glycemic stability, and bone metabolism.

2 Methods

3 Results

4 Discussion

Metabolic comorbidities, notably obesity and DM, are known to significantly impact outcomes following orthopaedic procedures, specifically spinal surgery.^1,2,4–7 With obesity projected to affect over half of the U.S. population by 2030 and the concurrent rise in type 2 diabetes prevalence, interventions targeting these modifiable risk factors are increasingly relevant in elective lumbar spine surgery.¹ Both obesity and DM are established predictors of postoperative complications such as SSIs, DVT, PE, wound healing disturbances, hospital readmissions, and pseudarthrosis.^4–7 Consequently, GLP-1 RAs, due to their beneficial effects on glycemic control, weight reduction, and potentially bone health, have emerged as valuable perioperative therapeutic agents.^8,10–12 However, their safety profile and long-term implications for elective lumbar spine surgery outcomes have not been thoroughly examined.

In our propensity-matched cohort analysis of 5722 patients, GLP-1 RA use was not associated with any elevated risks in medical complications within 90 days or mechanical-related complications within 1 year or 3-years following lumbar facetectomy, foraminotomy, laminectomy, laminotomy, and disc or joint excision procedures. Notably, however, we observed a significant reduction in pseudarthrosis rates at both one-year and three-year follow-ups among GLP-1 RA users undergoing arthrodesis procedures. Overall, these results align with an expanding body of evidence indicating the perioperative benefits and safety associated with GLP-1 Ras use following orthopaedic procedures.^19,20,27,28

The sustained reduction in pseudarthrosis at one-year and three-year intervals (6.54% vs. 8.53%, P < 0.05), and (8.88% vs. 10.77%, P < 0.05), respectively, is a noteworthy finding. This approximate 20% relative risk reduction corroborates findings from Ghali et al., who reported similar pseudarthrosis reduction among diabetic patients undergoing lumbar fusion (12% vs. 16%, P = 0.002).¹ Comparable risk reductions have also been documented in cervical spine procedures.²⁹

Preclinical evidence supports a plausible biological mechanism to explain these findings, demonstrating GLP-1 receptor expression in osteoblasts, with receptor activation promoting osteogenesis through the MAPK/ERK and Wnt signaling pathways, subsequently increasing the expression of key osteogenic markers such as RUNX2 and osteocalcin.^30,31 Concurrently, GLP-1 RAs also suppress osteoclast-mediated bone resorption by modulating the RANKL/osteoprotegerin ratio and stimulating calcitonin release.³⁰ Such dual mechanisms in these agents are likely responsible for the net positive effect on bone remodeling, facilitating improved graft integration. Clinical trials further substantiate this, demonstrating that GLP-1 RAs increase bone formation markers, including a notable rise in P1NP in patients treated with liraglutide³²; additionally, meta-analyses suggest these medications enhance bone density without increasing fracture risk.^33–35 Moreover, the anti-inflammatory effects of GLP-1 RAs, demonstrated by reductions in CRP and inflammatory cytokines,³⁶ could also diminish inflammation-driven delays in bone healing. While improvements in glycemic control and weight loss likely play contributory roles, the persistence of reduced pseudarthrosis after propensity matching for diabetes and obesity suggests direct osteogenic actions of GLP-1 RAs.

While the observed reduction in pseudarthrosis underscores the potential long-term benefits of GLP-1 RAs in spinal fusion, evaluating their overall perioperative safety profile in elective lumbar spine procedures remains equally important. Our analysis of 90-day medical complications, including SSIs, pneumonia, blood transfusions, readmission rates, sepsis, acute kidney failure, wound dehiscence, DVT, PE, myocardial infarction, and stroke, revealed comparable outcome rates between GLP-1 RA users and non-users. These findings hold clinical significance given that patients with obesity and diabetes are predisposed to systemic inflammation, endothelial dysfunction, and immunosuppression—factors that amplify risks for thromboembolic events, infections, and cardiometabolic complications.^{4–7,37–40} Despite these inherent risks, our analysis revealed no excess morbidity in GLP-1 RA users, suggesting these agents may counteract the pathophysiological mechanisms driving adverse perioperative complications. Our findings align with emerging evidence in arthroplasty studies as Magruder et al. demonstrated that GLP-1 RAs attenuated medical outcomes, infection rates, and readmissions in arthroplasty patients, even among metabolically high-risk cohorts.^19,20 These favorable outcomes observed in patients treated with GLP-1 RAs are likely attributable to improved glycemic control and weight reduction, both of which are established contributors to decreased systemic morbidity, lower infection risk, and improved wound healing.^2,41

Clinically, as GLP-1 RAs become more widely prescribed, now used by over 12.5% of U.S. adults,^20–22 assessing their perioperative safety, efficacy, and therapeutic value has become increasingly relevant for spine surgeons. Our findings support the safe use of GLP-1 RAs in the context of elective lumbar surgery, providing reassurance that these agents do not heighten complication risks and may enhance fusion outcomes following arthrodesis procedures.

5 Conclusions

Our study demonstrated that the use of GLP-1 receptor agonists does not elevate the risk of medical or mechanical complications following elective lumbar spine surgery after adjusting for pertinent comorbidities. Notably, patients receiving GLP-1 RAs exhibited significantly lower and sustained rates of pseudarthrosis following spinal arthrodesis procedures. As one of the first studies evaluating both short- and intermediate-term outcomes associated with GLP-1 RA use in elective lumbar spine surgery, our findings provide foundational evidence of the safety profile and potential therapeutic benefits of GLP-1 RAs. Prospective randomized controlled trials are warranted to validate our findings and further explore potential synergistic effects when combined with established osteoanabolic agents such as romosozumab, teriparatide, or abaloparatide. Nonetheless, existing evidence in spinal literature supports the consideration of GLP-1 RAs as part of preoperative optimization strategies for patients at elevated metabolic risk undergoing lumbar spinal surgery.

CRediT authorship contribution statement

Sri Tummala: Writing – original draft, Data curation, Formal analysis, Investigation, Conceptualization, Methodology, Validation, Writing – review & editing, Prepared, Project administration. David C. Gibbs: Writing – review & editing, Validation, Software. Joseph Chavarria: Writing – review & editing, Validation. Jason Alder: Writing – review & editing, Validation. Ioannis Avramis: Supervision, Project administration, Writing – review & editing, Supervision, Validation. James M. Rizkalla: Conceptualization, Methodology, Supervision, Project administration, Writing – review & editing, Supervision, Validation, All authors have read and approved the final manuscript.

Informed consent statement

This retrospective study is exempt from informed consent. The data reviewed represent a secondary analysis of existing data, do not involve intervention or interaction with human subjects, and are de-identified per the de-identification standard defined in Section §164.514(a) of the HIPAA Privacy Rule. The process by which the data are de-identified is attested to through a formal determination by a qualified expert as defined in Section §164.514(b)(1) of the HIPAA Privacy Rule. This formal determination by a qualified expert was refreshed on December 2020.

Ethical statements

This study was exempt from IRB approval since the data were de-identified and publicly available.

Funding and sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.