Perioperative Glucagon-like Peptide-1 Receptor Agonist Use Is Associated With Lower Pseudoarthrosis Rates Following Long-Segment Spinal Deformity Correction: A Propensity-Matched Analysis

Abstract

Study Design

Retrospective Cohort Study

Objective

This study investigated the association between perioperative glucagon-like peptide-1 receptor agonist (GLP-1 RA) therapy and pseudoarthrosis rates in patients undergoing long-segment spinal deformity surgery.

Methods

Adult patients with a history of spinal deformity who underwent posterior segmental instrumentation spanning seven or more levels between 7/1/2010 and 7/1/2024 were identified. Patients prescribed a GLP-1 RA within six months before or after surgery comprised the experimental group; patients without GLP-1 RA use comprised the control group. Cohorts were balanced 1:1 by demographic characteristics and medical comorbidities. The primary outcome was the clinical diagnosis of pseudoarthrosis at 6 months, 1 year, 2 years, and 3 years following the index procedure; relevant 90-day medical complications were also assessed using risk ratios (RR) and 95% confidence intervals.

Results

Perioperative GLP-1 RA use was associated with a significant reduction in risk of pseudoarthrosis at 6 months (RR 0.677, P = .028), one year (RR 0.708, P = .036), two years (RR = 0.671, P = 0.010), and three years (RR = 0.714, P = .024) following the index procedure. Perioperative GLP-1 RA use was also associated with a significant reduction in hospital readmission (RR 0.863, P = .012) and sepsis (RR 0.485, P = .002) within 90 days following the index procedure.

Conclusions

This study supports the potential protective effect of GLP-1 RAs against the risk of pseudoarthrosis in patients who underwent long-segment spine deformity surgery. Further, perioperative GLP-1 RA therapy was not associated with increased risk of medical complications in the 90-day postoperative period and may reduce hospital readmission and sepsis.

Introduction

Surgical management of adult spinal deformity (ASD) is an increasingly utilized treatment option to address structural spinal abnormalities. The surgical correction of ASD is associated with superior improvements in function, pain relief, patient satisfaction, mental health, and radiographic deformity correction when compared to nonoperative management. ¹ However, these advantages come at the cost of a higher complication rate, particularly in long-segment cases that exceed seven consecutive segments.^2-5

Pseudoarthrosis, defined as the failure of bony fusion across a surgically intended fusion site, remains a significant complication following spinal fusion.^6,7 It can result in persistent axial or radicular pain and may require a more extensive second operation that utilizes a different surgical approach to the spine. ⁶ Importantly, long-segment constructs, such as those indicated by ASD, carry a higher risk of pseudoarthrosis when compared to shorter-segment fusions.^8,9 This risk is multifactorial. Longer constructs introduce a greater number of fusion interfaces, with each additional level increasing the cumulative risk of nonunion. Multiple studies have demonstrated fusion length as an independent predictor of pseudoarthrosis.^7-12 Additional risk factors include elevated body mass index (BMI), advanced age, smoking, diabetes mellitus, long-term steroid use, surgical site infection, and poor bone quality (osteoporosis or osteopenia).^7,9,12-14 Successful bony fusion also requires adequate vascularization and blood supply, which is often impaired in patients with uncontrolled blood glucose levels, leading to decreased bone strength, low bone turnover, and worse fusion outcomes.^15-17

In recent years, glucagon-like peptide-1 (GLP-1) receptor agonists have become increasingly utilized for the management of Type 2 diabetes mellitus (T2DM) and obesity. ¹⁸ Further, given the known risks of obesity on postoperative outcomes, these therapies have been suggested for preoperative optimization of obese patients while also demonstrating favorable effects on markers of bone metabolism.^19,20 Thus, it is postulated that the use of GLP-1 receptor agonists may facilitate bony healing following elective spinal fusion procedures. Recent studies have found that the use of GLP-1 receptor agonists significantly reduces the incidence of pseudoarthrosis following single-level cervical and lumbar spine surgery, supporting this hypothesis.^21,22

However, to our knowledge, no evidence exists regarding their use in long-segment deformity fusion, which presents a different risk profile than short-segment fusion procedures.^8,23 The objective of this study was to investigate the relationship between perioperative GLP-1 receptor agonist therapy and pseudoarthrosis rates in patients undergoing long-segment complex spinal deformity surgery spanning seven or more consecutive spinal segments. We hypothesized that perioperative use of GLP-1 receptor agonists would be independently associated with a lower incidence of pseudoarthrosis in this high-risk cohort.

Methods

Data Source

The study employed the TriNetX Research Network database (Cambridge, Massachusetts), a global healthcare data network that compiles real-time, de-identified electronic medical data from over 120 healthcare organizations (HCOs) and 250 million patient lives. ²⁴ Participating HCOs permit the use of their information in exchange for access to the platform’s research capabilities at no additional cost, which includes clinical trial protocol design, database query, and drug safety analysis.

Ethical Approval

The TriNetX Research Network is compliant with the Health Insurance Portability and Accountability Act of 1996 (HIPAA) Security Rule and certified according to the ISO 27001:2013 standard. For these reasons, the study conferred minimal risk to patient privacy and was exempt from institutional review board approval and informed consent at our institution. As recommended, this study was performed and reported in accordance with Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Cohort Selection

All data included in this analysis were obtained on July 29, 2025, from the TriNetX Research Network database within the 14-year period beginning July 1, 2010, and ending July 1, 2024. This time constraint allowed for at least one year of follow-up for all patients included in the study. Inclusion criteria, exclusion criteria, propensity score matching criteria, and study endpoints utilized classifications outlined by the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes, Current Procedural Terminology (CPT), or RxNorm codes. Patient cohorts were first established by querying the TriNetX database for all patients aged 18 years and older with a previous history of scoliosis, kyphosis, lordosis, or congenital spine malformation (ICD-10 M41, M40, or Q76) who underwent posterior segmental instrumentation of at least seven consecutive levels (CPT 22843 or 22844). The GLP-1 receptor agonist-exposed group included patients who were prescribed a GLP-1 receptor agonist within the six months before or after long-segment spinal fusion (RxNorm 1991302, 1440051, 1534763, 475968, 60548, or 1551291). The control group contained patients who did not receive a GLP-1 receptor agonist within the 12-month perioperative period (Figure 1).

Figure 1.

Flowchart Demonstrating Patient Selection in the TriNetX Database. Abbreviations - GLP-1, Glucagon-like Receptor-1; PSM, Propensity Score Matching

Propensity Score Matching

Next, the two cohorts were prepared for 1:1 propensity matching to limit the potential for confounding covariates. The characteristics chosen for matching included age at index, sex, race, ethnicity, body mass index (BMI), hemoglobin A1c (HbA1c), diabetes mellitus (DM), hypertension, nicotine dependence, osteoporosis, heart failure, and chronic kidney disease. Patients who lacked data for any of the matching criteria were excluded from the analysis. Successful cohort matching was reflected by a standard difference less than or equal to 0.1 for each of the selected covariates.

Outcomes of Interest

The primary outcome was pseudoarthrosis, as defined by ICD-10 M96.0, and was assessed at 6 months, 1 year, 2 years, and 3 years following long-segment spinal fusion. Secondary endpoints assessed within 90 days included myocardial infarction (MI), blood transfusion, acute kidney failure, pneumonia, deep vein thrombosis (DVT) or pulmonary embolism (PE), stroke, sepsis, emergency department (ED) utilization, hospital readmission, superficial surgical site infection (SSI), deep SSI, foot drop, and dural sac tear.

Statistical Analysis

Baseline characteristics were evaluated using descriptive statistics. Binary outcomes for each endpoint were assessed with relative risk (RR) and 95% confidence intervals (CI). All statistical procedures were automatically performed within the TriNetX research platform. In situations when a patient cohort contained fewer than 10 patients, TriNetX rounded up to 10 to maintain patient confidentiality, and statistical analyses were executed using the rounded value. A two-sided P-value < 0.05 was considered to be statistically significant.

Results

The analysis yielded 328 matched pairs. The average age at index after matching was 64.4 ± 9.7 years in the GLP-1 receptor agonist-exposed group compared with 66.1 ± 9.8 years in the control (Table 1). Patients in the experimental group were 48.8% female compared with 49.4% in the control group. Approximately 83.2% of patients in the GLP-1 receptor agonist-exposed group were diagnosed with diabetes mellitus. There were no other differences between the two groups in medical comorbidities after matching (Table 1). Similar proportions across the selected BMI and hemoglobin A1c categories were observed (Table 2).

Table 1.

Baseline Characteristics and Propensity Matching

	Before matching				After matching
Characteristic	GLP-1 RA (+)	GLP-1 RA (−)	P-value	Std diff	GLP-1 RA (+)	GLP-1 RA (−)	P-value	Std diff
Demographic
Age at index	64.4 (9.7)	43.0 (25.6)	<0.001	1.109	64.4 (9.7)	66.1 (9.8)	0.026	0.174
Female	161 (48.8)	12 633 (56.5)	0.005	0.155	160 (48.8)	162 (49.4)	0.876	0.012
Black or African American	35 (10.6)	2117 (9.5)	0.484	0.038	35 (10.7)	38 (11.6)	0.710	0.029
White	227 (68.8)	16 117 (72.1)	0.185	0.072	227 (69.2)	218 (66.5)	0.452	0.059
Hispanic or Latino	19 (5.8)	1365 (6.1)	0.793	0.015	18 (5.5)	19 (5.8)	0.866	0.013
Comorbidity
Diabetes mellitus	275 (83.3)	2919 (13.1)	<0.001	1.978	273 (83.2)	275 (83.8)	0.833	0.016
Hypertension	275 (83.3)	8318 (37.2)	<0.001	1.069	273 (83.2)	278 (84.8)	0.594	0.042
Nicotine dependence	112 (33.9)	3381 (15.1)	<0.001	0.448	110 (33.5)	111 (33.8)	0.934	0.006
Osteoporosis	68 (20.6)	2873 (12.8)	<0.001	0.209	68 (20.7)	71 (21.6)	0.774	0.022
Heart failure	66 (20.0)	1315 (5.9)	<0.001	0.430	65 (19.8)	69 (21.0)	0.698	0.030
Chronic kidney disease	85 (25.8)	1583 (7.1)	<0.001	0.521	85 (25.9)	82 (25.0)	0.788	0.021

Legend – Continuous variables represented as quantity (standard deviation); categorical variables represented as n (% in cohort); Std diff <0.1 after matching reflects no difference between the two groups. Abbreviations - GLP-1RA (+), glucagon-like receptor-1 receptor agonist prescribed within 6 months before or after surgery; GLP-1RA (−), glucagon-like receptor-1 receptor agonist not prescribed within 6 months before or after surgery; Std diff., standard difference.

Table 2.

BMI and HbA1c Propensity Matching

	Before matching				After matching
BMI category	GLP-1 RA (+)	GLP-1 RA (−)	P-value	Std diff	GLP-1 RA (+)	GLP-1 RA (−)	P-value	Std diff
>18.5	13 (3.9)	3857 (17.3)	<0.001	0.443	13 (4.0)	15 (4.0)	0.699	0.030
18.5-24.9	33 (10.0)	8012 (35.8)	<0.001	0.646	33 (10.1)	28 (8.5)	0.501	0.053
25-29.9	115 (34.8)	6648 (29.7)	0.044	0.110	114 (34.8)	104 (31.7)	0.407	0.065
30-34.9	176 (53.3)	4823 (21.6)	<0.001	0.695	174 (53.0)	161 (49.1)	0.310	0.079
>35	165 (50.0)	3053 (13.7)	<0.001	0.847	163 (49.7)	156 (47.6)	0.585	0.043
HbA1c Category
>5.7	75 (22.7)	2990 (13.4)	<0.001	0.245	75 (22.9)	70 (21.3)	0.598	0.048
5.7-6.4	121 (36.7)	2239 (10.0)	<0.001	0.664	120 (36.6)	109 (33.2)	0.368	0.070
>6.4	194 (58.8)	1437 (6.4)	<0.001	1.347	192 (58.5)	176 (53.7)	0.208	0.098

Legend - Patient counts represented as n (% of cohort); Std diff <0.1 after matching reflects no difference between the two groups. Abbreviations - GLP-1RA (+), glucagon-like receptor-1 receptor agonist prescribed within 6 months before or after surgery; GLP-1RA (−), glucagon-like receptor-1 receptor agonist not prescribed within 6 months before or after surgery; Std diff., standard difference; BMI, body mass index; HbA1c; hemoglobin A1c.

Perioperative GLP-1 receptor agonist use was associated with a significant reduction in risk of pseudoarthrosis diagnosis at 6 months (RR 0.677, 95% CI (0.477 - 0.961)), one year (RR 0.708, 95% CI (0.512 - 0.980)), two years (RR = 0.671, 95% CI (0.537 - 0.950)) and three years (RR = 0.714, 95% CI (0.532 - 0.958)) following the index procedure (Table 3).

Table 3.

Pseudoarthrosis Rates

Time point	Risk in GLP-1 (+)	Risk in GLP-1 (−)	Risk difference	Risk ratio (95% CI)	P-value
6 Months	0.134	0.198	−0.064	0.677 (0.477, 0.961)	0.028
1 Year	0.155	0.220	−0.064	0.708 (0.512, 0.980)	0.036
2 Year	0.168	0.250	−0.082	0.671 (0.494, 0.910)	0.010
3 Year	0.183	0.256	−0.073	0.714 (0.532, 0.958)	0.024

Legend – risk is the proportion of patients with the outcome in the cohort. Bold face type indicates statistical significance. Abbreviations - GLP-1RA (+), glucagon-like receptor-1 receptor agonist prescribed within 6 months before or after surgery; GLP-1RA (−), glucagon-like receptor-1 receptor agonist not prescribed within 6 months before or after surgery; CI, confidence interval.

Interestingly, patients with GLP-1 receptor agonist exposure were more than 50% less likely to develop sepsis within the 90-day postoperative period (RR 0.485, 95% CI (0.272 - 0.863)) (Table 4). Patients in the experimental group were also admitted to the hospital at lower rates compared with matched controls (RR 0.863, 95% CI (0.784 - 0.949)). Rates of all other 90-day postoperative complications assessed were similar between the two cohorts.

Table 4.

90-Day Postoperative Complications

Complication	Risk in GLP-1 (+)	Risk in GLP-1 (−)	Risk difference	Risk ratio (95% CI)	P-value
MI	0.046	0.030	0.015	1.500 (0.684, 3.290)	0.308
Transfusion	0.113	0.070	0.043	1.609 (0.978, 2.645)	0.058
Acute kidney failure	0.110	0.125	−0.015	0.878 (0.577, 1.337)	0.544
Pneumonia	0.061	0.067	−0.006	0.909 (0.506, 1.633)	0.750
DVT/PE	0.131	0.098	0.034	1.344 (0.873, 2.068)	0.177
Stroke	0.030	0.037	−0.006	0.844 (0.365, 1.902)	0.664
Sepsis	0.049	0.101	−0.052	0.485 (0.272, 0.863)	0.012
ED Visit	0.180	0.223	−0.043	0.808 (0.595, 1.099)	0.173
Readmission	0.671	0.777	−0.107	0.863 (0.784, 0.949)	0.002
Superficial SSI	0.030	0.030	0.000	1 (0.422, 2.370)	1
Deep SSI	0.037	0.030	0.006	1.200 (0.526, 2.739)	0.664
Foot drop	0.030	0.030	0.000	1 (0.422, 2.370)	1
Dural sac tear	0.030	0.030	0.000	1 (0.422, 2.370)	1

Legend – risk is the proportion of patients with the outcome in the cohort. Bold face type indicates statistical significance. Abbreviations - GLP-1RA (+), glucagon-like receptor-1 receptor agonist prescribed within 6 months before or after surgery; GLP-1RA (−), glucagon-like receptor-1 receptor agonist not prescribed within 6 months before or after surgery; CI, confidence interval; MI, myocardial infarction; DVT, deep vein thrombosis; PE, pulmonary embolism; ED, emergency department; SSI, surgical site infection.

Discussion

As the prevalence of large-segment spinal deformity surgery, a high-risk procedure type, continues to rise, it is critically important to understand how evolving pharmacologic therapies and baseline comorbidities influence surgical outcomes. There has been a rapid increase in use and expansion of indications for the prescription of GLP-1 receptor agonists.^25,26 As a result, many patients who historically would not have been eligible for elective spine surgery due to high-risk profiles, including high BMI, diabetes mellitus, and metabolic syndrome, are increasingly becoming viable surgical candidates due to the effects of these medications. This shift underscores the vital importance of understanding the perioperative and long-term effects of GLP-1 therapies on fusion outcomes, including complications such as pseudoarthrosis, which is especially relevant for this high-risk cohort. This study identified a significant association between perioperative GLP-1 receptor agonist use and reduced odds of pseudoarthrosis at six-month, one-year, two-year, and three-year follow-up in patients undergoing long-segment spinal deformity fusion involving seven or more levels. Importantly, GLP-1 receptor agonist use was not associated with an increased risk of postoperative complications. These findings remained robust following PSM for demographic characteristics, BMI, HbA1c, DM, and nicotine use. To our knowledge, this is the first large-scale analysis evaluating the impact of GLP-1 receptor agonists in the setting of complex adult spinal deformity surgery.

The absolute rate of pseudoarthrosis in adult spinal deformity procedures is approximately 6.3%, although it has ranged as high as 24% in some studies; pseudoarthrosis typically presents within four years postoperatively.^7,10 Other challenges surrounding long-segment adult spinal deformity surgery include a patient population with a high prevalence of medical comorbidities, prolonged hospitalization, increased healthcare costs, and increased complication rates, ranging from 34.8% to as high as 70%, when compared to patients undergoing less invasive spine procedures.^27-29 Spinal deformity surgery is also subject to extremely high revision rates, of which pseudoarthrosis is a major indication, with approximately 20% of ASD patients requiring reoperation.^29-31 For example, Ledesma et al. reported a revision surgery rate of 25% following long-segment fusion (≥3 levels), compared to 11% for short-segment fusions (<3 levels) in patients with adult degenerative scoliosis. ³² Revision surgery is known to be associated with a higher rate of complications compared to primary surgery, while also necessitating additional costs and utilization of hospital resources.^33,34 Given this, there is growing interest in whether pharmacologic optimization of metabolic, microvascular, and skeletal health through GLP-1 receptor agonists may mitigate the risk of pseudoarthrosis and enhance long-term fusion outcomes in this vulnerable population.

In addition to robust improvements in glycemic and weight control, GLP-1 receptor agonists have demonstrated numerous positive effects on bone metabolism. Preclinical data suggest that GLP-1 RAs activate the Wnt/GSK-3β/β-catenin signaling pathway, promoting osteoblast differentiation and bone formation. ³⁵ Clinically, Cai et al. (2021) reported that the use of GLP-1 receptor agonists in patients with T2DM was associated with increased bone mineral density (BMD) of the total hip. ³⁶ Further, a systematic review and meta-analysis published by Li et al 2024 found that GLP-1 receptor agonist use increased serum levels of calcium, bone alkaline phosphatase, osteocalcin, and procollagen type N-terminal propeptide (P1NP), all of which reflect increased bone deposition. ²⁰ These findings have translated to positive results in clinical studies, as GLP-1 receptor agonist use has been associated with significant reductions in fracture incidence and pseudoarthrosis rates following elective foot and ankle surgery.^37,38 Similarly, our study demonstrated that GLP-1 receptor agonist use is associated with a significant reduction in pseudoarthrosis following posterior segmental instrumentation for adult spinal deformity. While the exact mechanism by which GLP-1 receptor agonists exert these effects is unclear and under ongoing investigation, it is likely related to improvement of the systemic metabolic state. A systematic review and meta-analysis by Starup-Linde and Vestergaard 2015 found that patients with T2DM had decreased levels of bone turnover markers, likely reflective of chronic inflammatory processes and impaired glucose utilization by osteoblasts. ³⁹ Thus, correction of glycemic parameters to appropriate ranges should reverse these effects and restore appropriate bone turnover, which may explain why perioperative GLP-1 receptor agonist use was associated with reduced rates of pseudoarthrosis.

Alternatively, the relationship between perioperative GLP-1 receptor agonist use and pseudoarthrosis may be related to positive effects upon microvasculature. Adequate local perfusion is a well-known prerequisite for bone healing.^40,41 Conversely, both preclinical and clinical studies have demonstrated that microangiopathy and the presence of advanced glycation end products in diabetes mellitus impair bone vasculature, leading to osteopenia. ⁴⁰ As a result, GLP-1 receptor agonists have been extensively investigated in various vascular contexts. In a meta-analysis of randomized controlled trials by Dicembrini 2017, GLP-1 receptor agonist use was shown to reduce the incidence of diabetic nephropathy. ⁴² Further, these agents lowered rates of major cardiovascular events vs matched controls. ⁴³ Mechanistically, it is believed that these protective effects are related to improvements in endothelial oxidative stress and cell senescence. ⁴⁴ Thus, it is plausible that GLP-1 receptor agonists improve blood flow to fusion sites, which facilitates adequate healing and eventual union.

While the widespread adoption of GLP-1 receptor agonists for osteoporosis or other conditions related to bone quality and healing is unlikely at this time, the observed relationship between their use and pseudoarthrosis is still clinically relevant. Given their robust weight loss effects, GLP-1 receptor agonists have been proposed as an optimization tool when BMI precludes elective surgery. ¹⁹ Although there is currently no published evidence regarding GLP-1 receptor agonist use in this capacity, the present analysis is consistent with previous studies that suggest perioperative use does not confer additional risk of postoperative complications. ⁴⁵ A large, retrospective cohort study by Tao et al 2025 found no difference in the rates of 30-day and 90-day postoperative complications between patients with and without semaglutide exposure prior to undergoing cervical decompression and fusion. ⁴⁵ In fact, our study found that patients with perioperative GLP-1 receptor agonist use experienced lower rates of sepsis and all-cause hospital readmission, which may be linked to improved glycemic control. ⁴⁶ Nonetheless, the use of these agents for surgical optimization still requires heightened vigilance for on-target adverse events, which include nausea, vomiting, gallbladder disease, reduced fat-free mass, and in rare instances, acute pancreatitis^47,48

Taken together, these data suggest that GLP-1 receptor agonists may be advantageous for surgical optimization prior to long-segment fusion. In this high-risk cohort, GLP-1 receptor agonists have exhibited no additional risk of postoperative complications while demonstrating a protective effect against pseudoarthrosis. Future research should prioritize prospective studies to validate these findings and elucidate the underlying biological mechanisms. Moreover, a substantial gap remains in our understanding of how GLP-1 receptor agonists may influence functional outcomes following long-segment fusion.

Limitations

This retrospective study has several inherent limitations due to the design and use of a federated electronic health record database. First, its retrospective, observational design precludes causal claims, and thus, these results should only be interpreted as associations between two variables. Although propensity-score matching was performed to balance baseline characteristics between cohorts, the risk of residual confounding remains. The matching process is unable to account for unmeasured, relevant variables such as postoperative compliance with surgeon recommendations, nutritional status, physical activity, or improvements in metabolic parameters among GLP-1 receptor agonist users. Likewise, multivariate regression could not be performed within the TriNetX platform, limiting our ability to assess whether perioperative GLP-1 receptor agonist exposure independently predicts pseudoarthrosis after adjustment for potential confounders. In addition, the analysis does not account for changes in surgical techniques that may influence pseudoarthrosis rates, such as the utilization rates of bone morphogenic protein or interbody cages. Further, although our analysis strictly included cases spanning seven or more levels, we are unable to report specific operative characteristics, such as the exact number of levels fused, length of construct, or the spinal regions involved, which can alter pseudoarthrosis rates.

Second, the use of TriNetX is wholly dependent upon medical billing codes, which can be used incorrectly or heterogeneously across different institutions, reducing data granularity and accuracy. Thus, our analysis may underestimate or overestimate the true relationship between perioperative GLP-1 receptor use and pseudoarthrosis following long-segment spinal deformity surgery. Electronic health record data also does not provide information on patient adherence to medication, dosage, exact initiation date, or duration of therapy, which may confound results. Specifically, the exact length of GLP-1 receptor agonist use through the perioperative period could not be determined. Additionally, based on providers’ preferences, GLP-1 receptor agonists may be suspended in the immediate pre- or postoperative period, potentially confounding postoperative outcome data. The sole use of medical billing codes to define pseudoarthrosis may also impair translation of our results, as clinical and radiographic criteria were not available. Further, it remains unknown whether the protective relationship between GLP-1 receptor agonist use and pseudoarthrosis extends to reduced incidence of revision surgery, which could not be extracted from the present dataset.

Finally, patients on GLP-1 receptor agonists may represent a more health-conscious population. This cohort may be more likely to engage in behaviors suggested to improve postoperative recovery and fusion outcomes. Although these extraneous variables introduce potential sources of bias, the use of propensity score matching of key clinical variables for the incidence of pseudoarthrosis mitigates some of this concern. Nevertheless, prospective studies are required to validate these findings and further elucidate the relationship between GLP-1 receptor agonist use and spinal fusion outcomes.

Conclusion

Our study found that patients who were prescribed glucagon-like peptide-1 receptor agonists within six months of spine deformity fusion surgery had significantly lower rates of pseudoarthrosis at six months, one year, two years, and three years postoperatively. Further, perioperative GLP-1 receptor agonist therapy was not associated with increased risk of medical complications in the 90-day postoperative period. These findings suggest that GLP-1 receptor agonists may have pleiotropic effects on bone healing following arthrodesis procedures, as propensity matching successfully paired the two cohorts according to baseline metabolic status. While our findings must be interpreted within the inherent confines of a retrospective database study, we believe that a prospective study investigating the protective effects of GLP-1 receptor agonists in adults who undergo long-segment spinal deformity surgery is warranted. This therapeutic class may prove substantially beneficial for adult patients with spinal deformities, particularly those requiring preoperative optimization secondary to underlying comorbidities such as obesity and Type 2 diabetes mellitus.

ORCID iDs

Kyle Stump https://orcid.org/0009-0006-3878-5183

Bong-Soo Kim https://orcid.org/0000-0002-8376-2005

Ethical Considerations

This study was waived for Institutional Review Board approval due to the use of a de-identified database.

Data Availability Statement

Data is available upon request.*