Reduced Complications and Improved Survival With Postoperative GLP-1 Receptor Agonist Use Following Lumbar Arthrodesis

Abstract

Study Design

Retrospective propensity-score-matched cohort study.

Objectives

To evaluate whether postoperative glucagon-like peptide-1 receptor agonist (GLP-1 RA) use is associated with fewer complications, lower pseudarthrosis, and improved survival after lumbar fusion in adults aged 50 years and older.

Methods

The TriNetX Global Collaborative Network was queried to identify adults ≥50 years who underwent lumbar arthrodesis (2010-2025). GLP-1 RA exposure was defined as an active prescription within 1 to 90 days postoperatively. The comparison group consisted of patients without a GLP-1 RA prescription during this window. Propensity score matching adjusted for demographics, BMI, hemoglobin A1c, preoperative GLP-1 RA use, and comorbidities. Outcomes: 90-day and 1-year infectious, pulmonary, and thromboembolic complications, 1-year pseudarthrosis, and 1-year mortality rates. Survival was assessed with Kaplan Meier and Cox regression.

Results

Among 169 286 patients who underwent lumbar arthrodesis, 1844 had postoperative GLP-1 RA use and 167 442 did not. After matching, 1816 patients remained in each cohort. Postoperative GLP-1 RA use was associated with lower 90-day pneumonia (Risk Ratio [RR] 0.51), surgical site infection (RR 0.58), sepsis (RR 0.64), urinary tract infection (RR 0.56), wound dehiscence (RR 0.46), and venous thromboembolism (RR 0.54) (all P < 0.05), and 1-year lower hazard of death (HR 0.38; P = 0.018).

Conclusion

In this large propensity-score-matched cohort, postoperative GLP-1 RA use was associated with fewer early infectious, pulmonary, wound, and thromboembolic complications and improved 1-year survival after lumbar spinal fusion.

Keywords

glucagon-like peptide-1 receptor agonist lumbar fusion postoperative complications mortality global database

Introduction

Obesity and Type 2 Diabetes Mellitus (T2DM) are prevalent diseases in the United States and are major drivers of escalating healthcare utilization.^1–3 Approximately 34.9%, or 78.6 million, of US adults are obese¹ and 9.5% have T2DM.³ By 2030, the prevalence of obesity is projected to be over 50%,⁴ whereas 1 in 10 US adults are expected to have T2DM in 2040.² These metabolic conditions are common among spine surgery patients. Wiener et al noted that 70% of lumbar fusion patients are overweight or obese.⁴ This is important because both these conditions are well understood to increase the risk of perioperative complications via impaired wound healing, altered immune function, and chronic systemic inflammation.^1,2,5

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have transformed the current landscape in the management of obesity and T2DM. They can provide rapid weight loss, improved glucose control, and a mortality benefit in certain patients.⁶ Roughly 12.5% of U.S. adults have been prescribed GLP-1 RA medication.³ These medications are now understood to influence various surgically relevant pathways important to tissue healing and have become the subject of significant investigation across surgical disciplines.^7,8

Prior large retrospective database studies evaluating spinal fusion cohorts have reported fewer complications and lower pseudarthrosis rates in patients taking GLP-1 RAs; however, nearly all of these studies defined GLP-1 RA exposure as a preoperative prescription. Many assumed continued postoperative use, despite little to no direct evidence that patients remained on therapy after surgery.^3,4,9–17 This methodology cannot accurately separate these medication’s preoperative metabolic optimization effects from their direct anti-inflammatory influence during the postoperative healing period. Additionally, most of these studies analyze diabetic and obese patients separately, even though these conditions frequently overlap. Furthermore, an increasing proportion of GLP-1 RA prescriptions are now written for weight loss alone, which dilutes exposure groups and complicates interpretation. These issues are particularly relevant in older adults, who represent the largest group that is prescribed GLP-1 RAs,¹⁸ and experience disproportionately higher postoperative complication rates compared to younger patients following spine surgery.¹⁹

To address these gaps, we examined postoperative GLP-1 RA use, rather than preoperative exposure, in a spinal fusion cohort utilizing the TriNetX research database. We hypothesized that active GLP-1 RA therapy during the early postoperative period would be associated with fewer infectious and pulmonary complications, lower rates of pseudarthrosis, and decreased mortality after lumbar arthrodesis in adults aged 50 years and older.

Methods

Study Design & Source

We performed a retrospective, observational cohort study using the TriNetX research database, specifically the Global Collaborative Network. TriNetX is a federated electronic health records database that compiles real-time data from participating healthcare systems across the United States and other parts of the world.²⁰ The platform contains de-identified data from over 160 healthcare organizations worldwide and is compliant with HIPAA regulations, and thus this study was exempt from institutional review board approval.

Study Population

A query for adult patients ≥50 years who underwent a lumbar spine fusion between January 2010 and November 2025 was performed using CPT codes for lumbar arthrodesis. Eligible procedure codes included 22533, 22534, 22558, 22585, 22612, 22614, 22630, 22632, 22633, 22634 (Supplemental Table 1). Patients with any prior lumbar spine fusion code were excluded to ensure the study cohort was comprised of only index operations.

We restricted the cohort to adults aged ≥50 years to create a more homogeneous population in order to limit residual confounding, particularly confounding by indication. This older demographic is most commonly prescribed GLP-1 RAs,¹⁸ are more likely to suffer from degenerative pathology,⁹ and face higher rates of postoperative complications.^5,21,22 Thus, the analysis focuses on patients who could potentially derive the most benefit from these medications.

Exposure cohorts were defined based on postoperative GLP-1 RA use. The GLP-1 RA cohort included patients with an active prescription for exenatide, liraglutide, dulaglutide, semaglutide, or tirzepatide between 1 to 90 days postoperatively. The comparison cohort consisted of patients without any recorded GLP-1 RA prescription during that same period.

Outcomes

The outcomes of interest were 90-day and 1-year postoperative complications: pneumonia, surgical site infection (SSI), sepsis, urinary tract infection (UTI), debridement and irrigation (D&I), wound dehiscence, venous thromboembolism (VTE), and pulmonary embolism (PE). Pseudarthrosis and mortality occurring within 1 year of the index procedure were also evaluated. Supplemental Table 1 lists the ICD-10 and CPT codes utilized to define each outcome.

Covariates and Matching

A 1:1 propensity score–matched (PSM) analysis was performed to limit confounding. Covariate selection followed recommendations from Brookhart et al, as we prioritized variables most likely associated with the outcomes of interest in order to reduce residual confounding.²³ These included age at index event, sex, diabetes mellitus, diseases of liver, nicotine dependence, diseases of arteries, arterioles, and capillaries, heart failure, chronic lower respiratory diseases, tobacco use, chronic kidney disease, inflammatory polyarthropathies, systemic connective tissue disorders, chronic ischemic heart disease, preoperative prescription for GLP-1 analogues, body mass index (further divided into: <20 kg/m², 20 to <25 kg/m², 25 to <30 kg/m², 30 to <35 kg/m², 35 to <40 kg/m², 40 to <45 kg/m², and ≥45 kg/m².), and hemoglobin A1c (further divided into: <6.5, 6.5 to <7.5%, 7.5 to <8.5%, 8.5 to <9.5%, 9.5 to <10.5%, 10.5 to <11.5%, 11.5 to <12.5%, 12.5 to <13.5%, ≥13.5%). The covariate time window was any instance of documentation of these variables in the patient’s medical record from 3 months to 1 day before surgery.

Statistical Analysis

Analyses were conducted within TriNetX. Propensity score matching (1:1 nearest neighbor; caliper 0.1) was applied to balance covariates between exposure groups, with standardized mean differences (SMD) less than 0.10 considered acceptable balance. Continuous variables were compared using t-tests and categorical variables using χ² tests. Risk ratios (RR) and absolute risk differences were calculated with 95% confidence intervals (CI) to estimate associations between GLP-1 RA use and 90-day postoperative complications. One-year all-cause mortality was evaluated using Kaplan–Meier survival analysis. Survival was compared using the log-rank test. A Cox proportional hazards model was then used to estimate the hazard ratio and 95% CI for mortality associated with postoperative GLP-1 RA use. Statistical significance was defined as P < 0.05.

Results

Pre-match cohort characteristics can be found in Table 1. The initial query found 169 286 patients, 1844 of which had postoperative GLP-1 RA use (Cohort A) and 167 442 of which did not (Cohort B). Before propensity score matching, Cohort A had a mean age at index surgery of 64.4 years and was primarily composed of females (56.7%), compared to 65.6 years and 50.3% female in Cohort B. Additionally, significant differences were noted across all covariates (SMD >0.1), aside from tobacco use. Post-match patient characteristics are found in Table 2. After matching, 1816 patients remained in each cohort and there was excellent covariate balance between cohorts (all SMD <0.1).

Table 1.

Patient Characteristics Before Propensity Score Matching

Variable	Postop GLP-1 RA (n = 1844)	No postop GLP-1 RA (n = 167 442)	P-value	SMD
Age at index (yrs, mean ± SD)	64.4 ± 7.7	65.6 ± 8.6	<0.001	0.145
Female	1741 (56.7%)	83, 709 (50.3%)	<0.001	0.129
Diabetes mellitus	1570 (85.5%)	30 749 (18.5%)	<0.001	0.808
Diseases of liver	172 (9.4%)	5266 (3.2%)	<0.001	0.258
Nicotine dependence	215 (11.7%)	13 935 (8.4%)	<0.001	0.111
Arterial disease	186 (10.1%)	10 469 (6.3%)	<0.001	0.140
Heart failure	234 (12.7%)	7434 (4.5%)	<0.001	0.298
Chronic lower respiratory diseases	549 (29.9%)	20 469 (12.3%)	<0.001	0.442
Tobacco use	35 (1.9%)	2816 (1.7%)	0.479	0.016
Chronic kidney disease	417 (22.7%)	11 244 (6.8%)	<0.001	0.462
Inflammatory polyarthropathies	238 (13.0%)	11 368 (6.8%)	<0.001	0.206
Systemic connective tissue disorders	51 (2.8%)	2266 (1.4%)	<0.001	0.100
Chronic ischemic heart disease	464 (25.3%)	18 658 (11.2%)	<0.001	0.370
Glucagon-like peptide-1 analogues	842 (45.8%)	1839 (1.1%)	<0.001	1.243
Body Mass index (mean ± SD)	34.0 ± 6.5	30.1 ± 6.3	<0.001	0.613
<20 kg/m2	19 (1.0%)	4846 (2.9%)	<0.001	0.135
20 - 25 kg/m2	117 (6.4%)	25 089 (15.1%)	<0.001	0.284
25 - 30 kg/m2	406 (22.1$)	45 585 (27.4%)	<0.001	0.122
30 - 35 kg/m2	610 (33.2%)	37 462 (22.5%)	<0.001	0.241
35 - 40 kg/m2	486 (26.5%)	19 464 (11.7%)	<0.001	0.383
40 - 45 kg/m2	248 (13.5%)	7550 (4.5%)	<0.001	0.317
>45 kg/m2	117 (6.4%)	3214 (1.9%)	<0.001	0.224
Hemoglobin A1c (mean ± SD)	7.0 ± 1.3	6.3 ± 1.2	<0.001	0.555
6.50 - 7.50 %	434 (23.6%)	7957 (4.8%)	<0.001	0.561
7.50 - 8.50 %	262 (14.3%)	3400 (2.0%)	<0.001	0.458
8.50 - 9.50 %	90 (4.9%)	1388 (0.8%)	<0.001	0.246
9.50 - 10.50 %	31 (1.7%)	562 (0.3)%	<0.001	0.135
10.50 - 11.50 %	21 (1.1%)	276 (0.2%)	<0.001	0.121
11.50 - 12.50 %	10 (0.5%)	125 (0.1%)	<0.001	0.085
12.50 - 13.50 %	10 (0.5%)	76 (0.04%)	<0.001	0.092
≥13.50 %	10 (0.5%)	57 (0.03%)	<0.001	0.095

Continuous variables are reported as mean ± SD; categorical variables as n (%).

GLP-1 RA = Glucagon-like peptide-1 receptor agonist; SD = standard deviation.

Table 2.

Patient Characteristics After Propensity Score Matching

Variable	Postop GLP-1 RA (n = 1816)	No postop GLP-1 RA (n = 1816)	P-value	SMD
Age at index (yrs, mean ± SD)	64.5 ± 7.6	64.3 ± 7.8	0.221	0.041
Female	1024 (56.4%)	1048 (57.7%)	0.221	0.041
Diabetes mellitus	1549 (85.3%)	1590 (87.6%)	0.047	0.066
Diseases of liver	167 (9.2%)	167 (9.2%)	1	<0.001
Nicotine dependence	212 (11.7%)	197 (10.8%)	0.431	0.026
Arterial disease	184 (10.1%)	166 (9.1%)	0.311	0.034
Heart failure	229 (12.6%)	195 (10.7%)	0.079	0.058
Chronic lower respiratory diseases	533 (29.4%)	535 (29.5%)	0.942	0.002
Tobacco use	34 (1.9%)	32 (1.8%)	0.804	0.008
Chronic kidney disease	410 (22.6%)	395 (21.8%)	0.549	0.020
Inflammatory polyarthropathies	233 (12.8%)	208 (11.5%)	0.204	0.042
Systemic connective tissue disorders	48 (2.6%)	48 (2.6%)	1	<0.001
Chronic ischemic heart disease	458 (25.2%)	438 (24.1%)	0.441	0.026
Glucagon-like peptide-1 analogues	821 (45.2%)	773 (42.6%)	0.108	0.053
Body Mass index (mean ± SD)	34.0 ± 6.5	34.1 ± 6.1	0.503	0.025
<20 kg/m2	19 (1.0%)	14 (0.8%)	0.382	0.029
20 - 25 kg/m2	117 (6.4%)	97 (5.3%)	0.159	0.047
25 - 30 kg/m2	405 (22.3%)	379 (20.9%)	0.294	0.035
30 - 35 kg/m2	604 (33.3%)	620 (34.1%)	0.574	0.019
35 - 40 kg/m2	472 (26.0%)	518 (28.5%)	0.087	0.057
40 - 45 kg/m2	239 (13.2%)	245 (13.5%)	0.770	0.010
>45 kg/m2	116 (6.4%)	102 (5.6%)	0.328	0.032
Hemoglobin A1c (mean ± SD)	7.0 ± 1.3	7.1 ± 1.3	0.064	0.078
6.50 - 7.50 %	424 (23.3%)	451 (24.8%)	0.295	0.035
7.50 - 8.50 %	254 (14.0%)	270 (14.9%)	0.450	0.025
8.50 - 9.50 %	89 (4.9%)	102 (5.6%)	0.334	0.032
9.50 - 10.50 %	31 (1.7%)	30 (1.7%)	0.897	0.004
10.50 - 11.50 %	21 (1.2%)	25 (1.4%)	0.553	0.020
11.50 - 12.50 %	10 (0.6%)	10 (0.6%)	1	<0.001
12.50 - 13.50 %	10 (0.6%)	10 (0.6%)	1	<0.001
≥13.50 %	10 (0.6%)	10 (0.6%)	1	<0.001

Continuous variables are reported as mean ± SD; categorical variables as n (%).

GLP-1 RA = Glucagon-like peptide-1 receptor agonist; SD = standard deviation.

Table 3 summarizes the 90-day and 1-year outcomes, while Figures 1 and 2 provide visual representations of the associated risk differences. Within 90 days after lumbar fusion surgery, patients with postoperative GLP-1 RA use (from 1 to 90 days after lumbar fusion surgery) had significantly lower rates of pneumonia, surgical site infection (SSI), urinary tract infection (UTI), wound dehiscence, sepsis, and venous thromboembolism (VTE). Pulmonary embolism and debridement and irrigation (D&I) were not significantly associated with GLP-1 RA use at any timepoint.

Table 3.

Post-match Analysis of Outcomes after Lumbar Fusion Surgery

Outcome	Timeframe	Postop GLP-1 RA	No postop GLP-1 RA	RR (95%CI)	P-value
Pneumonia	90 days postop	15/1537 (0.98%)	29/1528 (1.90%)	0.51 (0.28, 0.96)	0.032
Pneumonia	1 year postop	40/1537 (2.60%)	35/1520 (2.30%)	1.13 (0.72, 1.77)	0.592
SSI	90 days postop	44/1687 (2.61%)	79/1759 (4.49%)	0.58 (0.40, 0.86)	0.003
SSI	1 year postop	49/1687 (2.90%)	89/1754 (5.07%)	0.57 (0.41, 0.81)	0.001
Sepsis	90 days postop	33/1629 (2.03%)	52/1633 (3.18%)	0.64 (0.41, 0.98)	0.038
Sepsis	1 year postop	55/1629 (3.37%)	80/1634 (4.90%)	0.69 (0.49, 0.97)	0.029
UTI	90 days postop	33/1305 (2.53%)	63/1384 (4.55%)	0.56 (0.37, 0.84)	0.005
UTI	1 year postop	72/1305 (5.52%)	99/1410 (7.02%)	0.79 (0.59, 1.05)	0.107
D&I	90 days postop	43/1674 (2.57%)	63/1717 (3.67%)	0.70 (0.48, 1.03)	0.066
D&I	1 year postop	67/1674 (4.00%)	83/1714 (4.84%)	0.83 (0.60, 1.13)	0.235
Wound dehiscence	90 days postop	38/1698 (2.24%)	87/1767 (4.92%)	0.46 (0.31, 0.66)	<0.001
Wound dehiscence	1 year postop	48/1698 (2.82%)	101/1771 (5.70%)	0.50 (0.35, 0.69)	<0.001
VTE	90 days postop	23/1626 (1.42%)	44/1665 (2.64%)	0.54 (0.33, 0.88)	0.013
VTE	1 year postop	36/1626 (2.21%)	59/1659 (3.56%)	0.62 (0.41, 0.94)	0.022
PE	90 days postop	12/1712 (0.70%)	23/1746 (1.32%)	0.53 (0.27, 1.07)	0.070
PE	1 year postop	22/1712 (1.29%)	28/1744 (1.61%)	0.80 (0.46, 1.39)	0.430
Pseudarthrosis	1 year postop	62/1638 (3.79%)	122/1665 (7.33%)	0.52 (0.38,0.70)	<0.0001

Differences in cohort sizes reflect the exclusion of patients who experienced the outcome prior to the 90-day postoperative window following their initial lumbar fusion surgery.

GLP-1 RA = Glucagon-like peptide-1 receptor agonist; RR = risk ratio; CI = confidence interval; SSI = surgical site infection; UTI = urinary tract infection; D&I = debridement and irrigation; VTE = venous thromboembolism; PE = pulmonary embolism. Bolded values indicate statistical significance with P < 0.05.

Figure 1.

Ninety-day postoperative complication rates by postoperative GLP-1 receptor agonist use. Bars show the percentage of patients with each complication within 90 days of index surgery. Abbreviations: GLP-1 RA = Glucagon-like peptide-1 receptor agonist; SSI = surgical site infection; UTI = urinary tract infection; D & I = debridement and irrigation; VTE = venous thromboembolism; PE = pulmonary embolism. Black Bars = GLP-¹ RA use, Gray Bars = No GLP-1 RA use

Figure 2.

One-year postoperative complication rates by postoperative GLP-1 receptor agonist use. Bars show the percentage of patients with each complication within 1 year of index surgery. Abbreviations: GLP-1 RA = Glucagon-like peptide-1 receptor agonist; SSI = surgical site infection; UTI = urinary tract infection; D & I = debridement and irrigation; VTE = venous thromboembolism; PE = pulmonary embolism. Black Bars = GLP-1 RA use, Gray Bars = No GLP-1 RA use

At 90 days postoperatively, GLP-1 RA users were 49% less likely to develop pneumonia (RR 0.51, 95% CI 0.28-0.96; P = 0.032), 42% less likely to develop an SSI (RR 0.58, 95% CI 0.40-0.86; P = 0.003), 36% less likely to develop sepsis (RR 0.64, 95% CI 0.41-0.98; P = 0.038), and 44% less likely to develop a UTI (RR 0.56, 95% CI 0.37-0.84; P = 0.005). Postoperative GLP-1 RA use was also associated with a 54% reduction in wound dehiscence (RR 0.46, 95% CI 0.31-0.66; P < 0.001) and a 46% reduction in VTE (RR 0.54, 95% CI 0.33-0.88; P = 0.013).

At 1 year following lumbar fusion, GLP-1 RA use remained associated with significantly lower rates of SSI (RR 0.57, 95% CI 0.41-0.81; P = 0.001), wound dehiscence (RR 0.50, 95% CI 0.35-0.69; P < 0.001), VTE (RR 0.62, 95% CI 0.41-0.94; P = 0.022), and pseudarthrosis (RR 0.52, 95% CI 0.38-0.70; P < 0.001).

Table 4 presents the 1-year survival analysis. Postoperative GLP-1 RA use was associated with reduced 1-year mortality and improved survival compared to no GLP-1 RA use. Kaplan–Meier analysis demonstrated significantly greater survival at 1 year (log-rank P < 0.001). In Cox proportional hazards modeling, GLP-1 RA use was associated with a 62% lower hazard of death (HR 0.38, 95% CI 0.24-0.58, P = 0.018).

Table 4.

One-Year Survival and Mortality after Spine Surgery by GLP-1 RA Use

	GLP-1 RA use	No GLP-1 RA use	P-value
Deaths	28/1816 (1.54)	73/1816 (4.02)	–
KM survival probability at 1 year	98.24%	95.61%	<0.001*
Cox proportional hazards model	HR = 0.38 (0.24-0.58)		0.018

KM = Kaplan-Meier; GLP-1 RA = Glucagon-like peptide-1 receptor agonist. Bolded values indicate statistical significance with P < 0.05. *Log-rank χ² = 20.754 (df = 1).

Discussion

In this large, multicenter PSM cohort study within the TriNetX research database, postoperative exposure to GLP-1 RAs within the first 3 months following lumbar fusion was associated with a significantly lower incidence of early postoperative medical complications. This study is novel due to the fact that GLP-1RA exposure was defined as use in the postoperative period, whereas prior large database studies have only assessed preoperative or a combined pre and postoperative exposure to these medications.^{3,4,9–11,13,14,24–26} By isolating active use during the recovery period, our findings directly address a clinically relevant question for spine surgeons regarding the safety and even the potential benefit of initiating or resuming GLP-1 RA therapy immediately after lumbar fusion. Importantly, propensity score matching achieved excellent balance across BMI and hemoglobin A1c, allowing for comparison of patients with similar metabolic phenotypes who did or did not receive a GLP-1 RA. The persistence of these significant outcome associations despite matching suggests that the observed benefit of these medications is not solely due to differences in weight or glycemic control, but instead may relate to GLP-1 RA effects beyond metabolic regulation, including anti-inflammatory pathways that could mitigate postoperative complications.

The analysis demonstrated that postoperative GLP-1 RA use was associated with lower 90-day rates of several significant medical complications such as pneumonia, SSI, sepsis, UTI, wound dehiscence, and VTE (Table 3). These associations were most pronounced in the 90-day postoperative window, which aligns with the period when systemic inflammation, immune dysregulation, and endothelial dysfunction most strongly contribute to morbidity after lumbar spine fusion.^7,27,28

At 1 year postoperatively, pneumonia and UTI rates were similar between the 2 groups. In contrast, the lower rates of SSI, sepsis, wound dehiscence, and VTE for GLP-1 RA users persisted at 1 year (Table 3). These later findings should be interpreted cautiously, as GLP-1 RA exposure was defined only within the first 90 postoperative days, and events at 1 year are more likely influenced by baseline comorbidities and other exposures unrelated to the index surgery. However, the sustained differences in many complication rates at 1 year postoperatively may reflect the broad systemic effects of GLP-1 RAs on these outcomes, independent of surgical exposure. Preclinical and clinical evidence has shown that GLP-1 RAs directly influence critical inflammatory and immunomodulatory pathways.⁷ Thus, although GLP-1 RAs limit chronic hyperglycemia, a known risk factor for postoperative infection and pulmonary complications,^22,29 the observed reductions in these complications may also reflect anti-inflammatory effects beyond glycemic control alone. Our methods also support this interpretation, as we included non-diabetic patients in the analysis and most recent HbA1c in the matching process, suggesting that the observed associations are not explained by improved glycemic control alone.

Importantly, postoperative GLP-1 RA use was associated with roughly a 50% lower 1-year risk of pseudarthrosis (RR 0.52; P < 0.0001; Table 3). Multiple studies have demonstrated reduced pseudarthrosis rates among patients prescribed GLP-1 RAs prior to spine surgery,^3,9,10,13 and recent systematic reviews and meta-analyses have supported these findings.^25,26 However, these studies largely defined exposure as preoperative prescription only, or included a range of 6 months pre- and postoperatively.^9,10 Therefore, these data cannot support the conclusion that continued GLP-1 RA therapy during the postoperative period influenced fusion success.

As previously mentioned, the present study’s exposure definition suggests that continued or new GLP-1 RA therapy during the postoperative period may play a direct role in promoting successful fusion, rather than attributing the effects on fusion to preoperative metabolic and glycemic optimization. This is critical for spine surgeons who may face decisions of whether patients should resume or are safe to start a new GLP-1 RA in the postoperative period. Our findings support their safety and even favorable complication profile after lumbar spine surgery.

Most notably, postoperative GLP-1 RA exposure was associated with lower 1-year mortality and higher Kaplan-Meier survival in this analysis (Table 4). Kaplan-Meier estimates found significantly higher survival at 1 year (log-rank P < 0.001) and Cox proportional hazards modeling found a 62% lower hazard of death (HR 0.38, 95% CI 0.24-0.58, P = 0.018). This is consistent with prior literature that reports clinically significant cardiovascular and mortality benefits of GLP-1 RAs in non-surgical populations.^6,13 Our findings suggest an even larger protective mortality signal in this lumbar fusion cohort, but this should be interpreted cautiously given the retrospective design of the present study compared with the high-quality randomized controlled trials included in Rivera et al.⁶ Regardless, spine surgeons should recognize that these medications provide substantial benefits that extend far beyond postoperative complication reduction.

In addition to appetite suppression and glycemic control, GLP-1 RA’s exert direct biologic effects via various anti-inflammatory pathways. Basic science and clinical evidence demonstrate that GLP-1 RAs suppress nuclear factor–κB signaling, reduce pro-inflammatory cytokine production, and enhance anti-inflammatory pathways.⁷ These agents can drastically improve endothelial function and microvascular perfusion, aiding immune system function, tissue oxygenation, and wound healing. Direct reductions in cytokines such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) have been found in patients using these medications, independent of weight loss.⁷ Collectively, these mechanisms provide biologic plausibility for the reductions in infection, wound complications, VTE, and mortality observed in our study.

This study has several strengths. It represents one of the largest multicenter analyses evaluating GLP-1 RA use in spine surgery and is the first to define exposure specifically and only during the postoperative period, which is most relevant to complications that occur in the postoperative setting. Rigorous propensity score matching was used to minimize baseline differences, and outcomes were assessed at two clinically meaningful timepoints. Finally, inclusion of both diabetic and non-diabetic patients allowed evaluation of GLP-1 RA effects beyond glycemic control alone and thus improves the generalizability of our findings.

Importantly, a recent meta-analysis of GLP-1 RA use in spine surgery noted that prior studies did not ensure metabolic comparability between exposure groups, particularly with respect to BMI thresholds, glycemic control, concurrent metabolic therapies, and complications of obesity and diabetes.²⁴ The present study addressed several of these concerns by incorporating detailed BMI and hemoglobin A1c stratification into matching, adjusting for diabetes status and preoperative GLP-1 analogue use, and matching on multiple comorbid conditions commonly associated with metabolic disease. This approach improves comparability between these groups and strengthens confidence that the observed associations are not simply attributable to baseline metabolic differences.

This study also has limitations. Given its retrospective nature, causality cannot be inferred, and residual confounding persists despite matching. TriNetX lacks granular procedural detail, limiting adjustment for surgical complexity, fusion technique, and surgeon-specific or institution-specific factors. Medication adherence, dosage, and duration cannot be verified in TriNetX. Nonetheless, the consistency of increased complications across multiple categories, different timepoints, and the analytic approach utilized in this study strengthens confidence in the validity of observed findings.

Conclusion

In this propensity-matched cohort study, postoperative GLP-1 RA use demonstrated a protective association with early complications after lumbar spine fusion. Postoperative GLP-1 RA use was associated with lower rates of 90 day infectious and pulmonary complications and venous thromboembolism, as well as reduced 1-year pseudarthrosis and lower 1-year mortality. These findings support continuation of GLP-1 RAs in the immediate postoperative period and suggest that postoperative GLP-1 RA use may be a potential strategy to reduce morbidity in an increasingly metabolically complex spine surgery population. Future prospective studies could evaluate these medications in postoperative spine surgery patients to determine whether they causally reduce early infectious, pulmonary, and thromboembolic complications.

Supplemental Material

Supplemental material - Reduced Complications and Improved Survival With Postoperative GLP-1 Receptor Agonist Use Following Lumbar Arthrodesis

Supplemental material for Reduced Complications and Improved Survival With Postoperative GLP-1 Receptor Agonist Use Following Lumbar Arthrodesis by Emma Proffitt, BA, Samuel W. Rice, MD, Ryan Schiedo, MD and Michael Stauff, MD in Global Spine Journal

Footnotes

ORCID iDs

Emma B. Proffitt

Samuel W. Rice

Ryan M. Schiedo

Michael P. Stauff

Ethical Considerations

This study used a HIPAA-compliant, de-identified database and was therefore exempt from institutional review board approval.

Author Contributions

Emma Proffitt: Conceptualization, Background Research, Data Analysis, Figure Creation, Manuscript Writing & Editing. Samuel Rice: Conceptualization, Background Research, Data Analysis, Manuscript Writing & Editing. Ryan Schiedo: Conceptualization, Methodology, Manuscript Writing & Editing, Supervision. Michael Stauff: Conceptualization, Methodology, Manuscript Writing & Editing, Supervision.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data are available through TriNetX Global Collaborative network and are not publicly available.*

Supplemental Material

Supplemental material for this article is available online.

References

Jackson

Devine

. The effects of obesity on spine surgery: a systematic review of the literature. Glob Spine J. 2016;6(4):394-400. doi:10.1055/s-0035-1570750

Moazzeni

Kazemi

Khanmohammad

Eslamian

Rostami

Faghih-Jouibari

. Comparison of surgical outcome between diabetic versus nondiabetic patients after lumbar fusion. Int J Spine Surg. 2018;12:5064-5532. doi:10.14444/5064. Published online.

Tummala

Gibbs

Chavarria

Alder

Avramis

Rizkalla

. GLP-1 receptor agonist use in elective lumbar spine surgery: reduced pseudarthrosis rates and favorable safety profile. J Orthop. 2025;65:227-232. doi:10.1016/j.jor.2025.05.018

Wiener

Sanghvi

Vlastaris

, et al. Glucagon-like Peptide-1 receptor agonist medications alter outcomes of spine surgery: a study among over 15,000 patients. Spine. 2025;50(13):871-880. doi:10.1097/BRS.0000000000005283

Patel

Bagan

Vadera

, et al. Obesity and spine surgery: relation to perioperative complications. J Neurosurg Spine. 2007;6(4):291-297. doi:10.3171/spi.2007.6.4.1

Rivera

Cruz

LLA

Magalong

, et al. Cardiovascular and renal outcomes of glucagon-like peptide 1 receptor agonists among patients with and without type 2 diabetes mellitus: a meta-analysis of randomized placebo-controlled trials. Am J Prev Cardiol. 2024;18:100679. doi:10.1016/j.ajpc.2024.100679

Alharbi

. Anti-inflammatory role of glucagon-like peptide 1 receptor agonists and its clinical implications. Ther Adv Endocrinol Metab. 2024;15:20420188231222367. doi:10.1177/20420188231222367

Gatto

Liu

Milan

Wong

. The effects of GLP-1 agonists on musculoskeletal health and orthopedic care. Curr Rev Musculoskelet Med. 2025;18(10):469-480. doi:10.1007/s12178-025-09978-3

Agrawal

Amasa

Karabacak

Margetis

. Perioperative glucagon-like Peptide-1 agonist use and rates of pseudarthrosis after single-level lumbar fusion: a large retrospective cohort study. Neurosurgery. 2025;97(1):91-97. doi:10.1227/neu.0000000000003291

10.

Chang

Chi

Song

Lin

. Association between glucagon-like peptide-1 receptor agonists and nonfusion risk after single-level anterior cervical discectomy and fusion. Spine J. 2026;26(1):158-162. doi:10.1016/j.spinee.2025.07.010

11.

Ghali

Lawand

Mitchell

, et al. The use of semaglutide in patients undergoing lumbar fusion does not increase 90-Day medical or 1-Year implant complications. Clin Spine Surg. 2025;38(10):E579-E584.doi:10.1097/BSD.0000000000001800

12.

Goldman

Mani

Scharfenberger

, et al. Glucagon-like peptide-1 receptor agonist use prior to spinal surgery results in reduced postoperative length of stay: a propensity-score matched analysis. North Am Spine Soc J NASSJ. 2025;22:100612. doi:10.1016/j.xnsj.2025.100612

13.

Kesaria

Marzook

Glover

Alijanipour

. Association between preoperative GLP-1 receptor analog use and postoperative complications and mortality following lumbar fusion surgery. Glob Spine J. 2025;25:21925682251391693. doi:10.1177/21925682251391693. Published online October.

14.

Khalid

Mirpuri

Massaad

, et al. The impact of semaglutide use for weight loss on transforaminal lumbar interbody fusion outcomes. Clin Neurol Neurosurg. 2025;254:108952. doi:10.1016/j.clineuro.2025.108952

15.

Mastrokostas

, et al. Semaglutide use is associated with higher rates of pseudarthrosis and dysphagia in patients undergoing posterior cervical fusion. Spine J. 2025;25(9):1974-1980. doi:10.1016/j.spinee.2025.03.023

16.

Seddio

Gouzoulis

Vasudevan

, et al. Semaglutide utilization associated with reduced 90-day postoperative complications following single-level posterior lumbar fusion for patients with type II diabetes. Spine J. 2025;25(3):485-493. doi:10.1016/j.spinee.2024.10.011

17.

Tao

Ranganathan

Van Halm-Lutterodt

, et al. No difference in short-term surgical outcomes from semaglutide treatment for type 2 diabetes mellitus after cervical decompression and fusion: a propensity score-matched analysis. Spine. 2025;50(8):515-521. doi:10.1097/BRS.0000000000005099

18.

Varghese

Shah

, et al. Prescribing trends of glucagon-like peptide 1 receptor agonists for type 2 diabetes or obesity. JAMA Netw Open. 2025;8(10):e2540890. doi:10.1001/jamanetworkopen.2025.40890

19.

Saleh

Thirukumaran

Mesfin

Molinari

. Complications and readmission after lumbar spine surgery in elderly patients: an analysis of 2,320 patients. Spine J. 2017;17(8):1106-1112. doi:10.1016/j.spinee.2017.03.019

20.

Palchuk

London

Perez-Rey

, et al. A global federated real-world data and analytics platform for research. JAMIA Open. 2023;6(2):ooad035. doi:10.1093/jamiaopen/ooad035

21.

Epstein

. Predominantly negative impact of diabetes on spinal surgery: a review and recommendation for better preoperative screening. Surg Neurol Int. 2017;8(1):107. doi:10.4103/sni.sni_101_17

22.

Olsen

Nepple

Riew

, et al. Risk factors for surgical site infection following orthopaedic spinal operations. J Bone Jt Surg-Am. 2008;90(1):62-69. doi:10.2106/JBJS.F.01515

23.

Brookhart

Schneeweiss

Rothman

Glynn

Avorn

Stürmer

. Variable selection for propensity score models. Am J Epidemiol. 2006;163(12):1149-1156. doi:10.1093/aje/kwj149

24.

Ibrahim

Durrani

Ibrahim

, et al. Glucagon-like Peptide-1 receptor agonist use does not impact spine surgery outcomes: a systematic review and meta-analysis. Glob Spine J. 2026;21925682251415347. doi:10.1177/21925682251415347. Published online January 3, 2026.

25.

Younus

Garcia De Oliveira

Fujii

, et al. Are GLP-1 receptor agonists safe in spine surgery: a systematic review. Spine. 2025. doi:10.1097/BRS.0000000000005466. Published online July 28.

26.

Zahed

Elmesalmi

Al-Kharouf

, et al. Efficacy and safety of glucagon-like Peptide-1 (GLP-1) receptor agonists for spinal fusion outcomes: a comprehensive meta-analysis. Cureus. 2025;17:e93065. doi:10.7759/cureus.93065. Published online 23 September 2025.

27.

Chen

Mei

Wei

, et al. GLP-1 receptor agonist as a modulator of innate immunity. Front Immunol. 2022;13:997578. doi:10.3389/fimmu.2022.997578

28.

Luna-Marco

de Marañon

Hermo-Argibay

, et al. Effects of GLP-1 receptor agonists on mitochondrial function, inflammatory markers and leukocyte-endothelium interactions in type 2 diabetes. Redox Biol. 2023;66:102849. doi:10.1016/j.redox.2023.102849

29.

Tang

Yao

Wang

Ding

. Association between admission hyperglycemia and postoperative pneumonia in geriatric patients with hip fractures. BMC Muscoskelet Disord. 2023;24(1):700. doi:10.1186/s12891-023-06829-5

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.01 MB

0.00 MB