Combination anti-obesity medications to effectively treat bariatric surgery weight regain at an academic obesity center

Abstract

Background

Combination anti-obesity medications (AOMs) to treat postoperative bariatric surgery weight regain have limited data on their use in the clinical setting. Understanding the optimal treatment protocol in this cohort will maximize weight loss outcomes.

Methods

A retrospective review of bariatric surgery patients (N = 44) presenting with weight regain at a single academic multidisciplinary obesity center who were prescribed AOM(s) plus intensive lifestyle modification for 12 months.

Results

Age: 28–76 years old, 93% female, mean weight 110.2 ± 20.3 kg, BMI 39.7 ± 7.4 kg/m², presenting 5.2 ± 1.6 years post-bariatric surgery [27 (61.4%), 14 (31.8%), and 3 (6.8%) laparoscopic Roux-en-Y gastric bypass (RYGB), laparoscopic vertical sleeve gastrectomy (VSG), and open RYGB, respectively], with 15.1 ± 11.1 kg mean weight gain from nadir. Mean weight loss after medical intervention at 3-, 6-, and 12-month time points was 4.4 ± 4.6 kg, 7.3 ± 7.0 kg, and 10.7 ± 9.2 kg, respectively. At 12 months, individuals prescribed 3 or more AOMs lost more weight than those prescribed one (−14.5 ± 9.0 kg vs. −4.9 ± 5.7 kg, p < 0.05) irrespective of age, gender, number of comorbidities, initial weight or BMI, type of surgery, or GLP1 use. RYGB patients lost less weight overall (7.4% vs. 14.8% VSG respectively; p < 0.05).

Conclusions

Combination AOMs may be needed to achieve optimal weight loss results to treat post-operative weight regain.

INTRODUCTION

Bariatric surgery initiates rapid and significant weight loss for patients with class 3 obesity or class 2 obesity with associated comorbidities.^1,2 The long-term durability of weight loss is variable; however, a systematic review suggests at least 1 in 6 bariatric surgery patients experience weight regain >10% from nadir.³ Moreover, an estimated 25% do not achieve expected weight loss outcomes⁴ and 30% fail to maintain 20% total weight loss (TWL%) at 10 years, either because of weight regain or insufficient initial loss.⁵ Other studies suggest up to 20% of patients regain the majority of their lost weight.⁶ Bariatric weight loss is typically maximal at 1–2 years but is often followed by a period of gradual regain.^2,4,7 Many factors underly post-surgical weight regain, and the presence of psychosocial conditions or grazing behavior has been shown to predict increased weight regain.^8,9 Yet, studies evaluating psychological or behavioral interventions alone have been shown to produce only minimal weight loss.^10–12 Though few, existing studies support anti-obesity pharmacotherapy in this population.^13–15 One multi-center study found pharmacotherapy facilitated >5% weight loss in more than half of their cohort, although this study did not include newer GLP-1 agonists like semaglutide.¹³

Medical treatment options for weight regain is growing, which includes pharmacotherapy targeting metabolic and nervous system pathways controlling absorption, hunger, satiety, and energy expenditure versus storage. Approved anti-obesity medications (AOMs) for long-term obesity control include: orlistat (pancreatic and gastric lipase inhibitor), phentermine-topiramate (combination sympathomimetic and antiepileptic/GABA enhancer/carbonic anhydrase inhibitor), bupropion-naltrexone (combination aminoketone antidepressant and opioid receptor antagonist), setmelanotide (melanocortin-4-receptor agonist for certain genetic obesity syndromes), and glucagon-like peptide 1 (GLP-1) receptor agonists semaglutide and liraglutide.^16,17 Collectively, AOMs can lead to 3%–9% weight loss in conjunction with intensive lifestyle and behavioral modifications, but new generation AOMs (i.e., semaglutide 2.4 mg FDA approved in June 2021) can result in >15% weight loss.^18,19 Regardless of the gains in the armamentarium of AOM options, an understanding of how these medications are best used in terms of timing, dose escalation and choice of agent in surgical patients is lacking.

Roux-en-Y gastric bypass (RYGB) and vertical sleeve gastrectomy (VSG) result in comparable weight loss,^20,21 but there is limited research on how each procedure may uniquely alter responsiveness to weight loss medications. A randomized controlled trial by Fatima et al²² found that RYGB surgery recipients had higher postprandial levels of GLP-1 than VSG recipients.²² Given these previous findings and because GLP-1 receptor agonists are highly effective anti-obesity therapeutics,²³ it was hypothesized that weight loss response to GLP-1 agonist therapy versus other AOMs might differ depending on type of surgery.

In the present study, outcomes of a single multidisciplinary specialized medical obesity treatment center in the management of weight regain post-bariatric surgery are reported. Weight loss at 3, 6, and 12 months is reported and associations with number of prescribed AOMs and type of bariatric surgery are analyzed.

METHODS Study design and approvals

To determine the medical weight loss program's efficacy, a chart review was conducted on weight loss outcomes and utilization data from September 2019 to March 2020 for patients presenting with weight regain greater than 10% of nadir weight after bariatric surgery. De-identified clinical outcomes for each patient were extracted from the electronic health record. All patients were treated at the institution's Weight Loss Center, a multidisciplinary outpatient obesity care center in Nashville, TN, which is staffed by obesity medicine specialists, bariatric surgeons, registered dietitians, and social workers.

Medical weight loss assessments

Assessments at the center consisted of (1) height and weight measurements, (2) laboratory screenings, (3) nutrition, and (4) psychosocial evaluations. Lifestyle interventions included counseling on nutrition, physical activity, stress, sleep hygiene (including screening for sleep apnea if indicated) and psychotherapy for maladaptive eating behaviors. Patients were treated with one or more FDA-approved or off-label AOMs, including topiramate, phentermine, metformin, setmelanotide (if genetic obesity testing positive for LEPR, POMC or PCSK1 gene mutation), lisdexamfetamine (if binge eating disorder was present), bupropion, naltrexone, or GLP-1 receptor agonists. Selection of AOMs was determined based on clinical decision-making as per obesity medicine specialists. Important factors dictating choice in pharmacotherapy included presence of indicating or contraindicating comorbidities and previous use or side effects of AOMs. Based on response and side effects, the number of AOMs a patient is prescribed may change over time. The number of AOMs was defined as the total number of prescribed AOMs at the point the clinician felt that the patient had achieved an optimal treatment strategy as per clinical standard of care in the practice of obesity medicine. If vitamin deficiencies were noted on laboratory screenings, appropriate oral or injectable supplements were prescribed. All patients were encouraged to continue to take daily multivitamins and supplements per bariatric surgery recommendations. Obesity-related comorbidities were collected from the patient chart based on ICD-10 code. Follow-ups were recommended at 3, 6, and 12 months.

Data visualization and statistics

Excess weight was calculated as body weight in excess of ideal weight that would result in BMI of 25 kg/m². Values are presented as means ± standard deviation unless indicated otherwise. Repeated measures analyses were used to assess changes in weight pre- and post-medical intervention. Weight loss comparisons were determined using unpaired t-tests and an alpha level of 0.05 was considered statistically significant. Statistical analysis was performed using R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria). This study was approved by the university's Institutional Review Board (IRB) and a waiver of informed consent was applicable.

RESULTS

A total of 44 patients were included in the analysis (Table 1). The patients were 28–76 years old, 93% female, with initial mean weight of 110.2 ± 20.3 kg, BMI 39.7 ± 7.4 kg/m². The patients were 70.4% African American, 25.0% white, 2.2% American Indian or Alaskan native, and 2.2% unknown. The participants had an average of 3.6 obesity-related medical conditions including type 2 diabetes, fatty liver, and sleep apnea. In terms of previous operation type, 27 (61.4%), 14 (31.8%), and three (6.8%) patients underwent laparoscopic RYGB, laparoscopic VSG, and open RYGB, respectively. The number of prescribed AOMs over the study period ranged from one to four, with an average of 1.9. The most commonly prescribed AOMs were GLP-1 receptor agonists (24 patients, 54.5%), phentermine (19 patients, 43.2%), topiramate (16 patients, 36.4%), and metformin (10 patients, 22.7%). Of those prescribed GLP-1 receptor agonists, 19 (79.2%) were on combination therapy with another AOM, while 5 (20.8%) were on monotherapy.

TABLE 1 Baseline characteristics of post-bariatric surgical patients (N = 44)

	Mean±SD or Value
Age	51.7±11.9
Gender
Male	6.8%
Female	93.2%
Race
Black or African American	70.4%
White	25.0%
American Indian or Alaskan Native	2.2%
Unknown	2.2%
BMI at initial visit	39.7±7.4
Initial program weight (kg)	110.2±20.3
Number of weight-related comorbidities	3.6±1.8
Prevalences of diabetes	31.8%
Prevalences of fatty liver	6.8%
Prevalences of sleep apnea	59.1%
Pre-operative weight (kg)	137.3±24.1
Pre-operative BMI	49.5±8.8
Nadir weight (kg)	95.1±19.3
Nadir BMI	34.2±6.5
Surgical weight loss (kg)	42.2±14.3
Percent surgical weight loss	30.0%
Time since surgery (years)	5.2±1.6
Weight gain, nadir to program weight (kg)	15.1±11.1
Percent surgical weight loss regained	35.9%

In patients with at least 3 months of follow-up with medical management (N = 29), mean weight loss was 4.4 ± 4.6 kg (3.9 total weight loss (TWL)%, p < 0.001 from baseline). In patients with at least 6 months of follow-up with medical management (N = 28), mean weight loss was 7.3 ± 7.0 kg (6.9 TWL% p < 0.001 from baseline). In patients with at least 12 months of follow-up with medical management (N = 25), mean weight loss was 10.7 ± 9.2 kg (9.5 TWL%, p < 0.001 from baseline).

The number of prescribed AOMs correlated with greater weight loss (Figure 1). For patients with at least 12 months of follow-up with medical management (N = 25), individuals prescribed three or more AOMs lost an average of 14.5 ± 9.0 kg, whereas individuals prescribed one AOM lost an average of 4.9 ± 5.7 kg (p < 0.05). At 12 months, individuals prescribed two AOMs lost an average of 13.1 ± 10.3 kg, but this did not significantly differ from weight loss in those who were prescribed one AOM (p = 0.06). There was no statistically significant difference between age, gender, number of comorbidities, starting weight, and starting BMI across the AOM cohorts.

FIGURE 1. Weight Loss by Number of AOMs. Weight loss at each assessed time point based on total number of AOMs prescribed. All weight loss was significant (p [less than] 0.001) from baseline starting weight. At 12 months, individuals prescribed three or more AOMs had greater weight loss than those prescribed only one AOM (*p [less than] 0.05). Data are mean ± SEM. At the three-, six-, and 12-month follow-up, N = 8, 8, and 9, respectively, for one AOM; N = 12, 12, and 7, respectively, for two AOMs; and N = 9, 8, and 9, respectively, for three or more AOMs

Patients who underwent VSG lost more weight (16.6 ± 9.4 kg) compared to those who underwent RYGB (8.2 ± 8.4 kg) (p < 0.05) after 12 months of medical management (Figure 2). TWL% was 7.4 for RYGB recipients and 14.8 for VSG recipients.

FIGURE 2. Weight Loss by Surgery-Type. Initial cohorts sizes were n = 27 for RYGB and n = 14 for VSG. At the three-, six-, and 12-month follow-up, N = 18, 15, and 16, respectively, for RYGB and N = 9, 11, and 7, respectively, for VSG. Initial cohort size for open gastric bypass recipients was n = 3 and this cohort was not included in this analysis. Weight loss was statistically significantly lower than baseline weight regardless of time period (p [less than] 0.001). At 12 months, VSG patients had greater weight loss than RYGB patients (*p [less than] 0.05). Data are mean ± SEM

At 12 months, RYGB patients taking GLP-1 receptor agonists lost 8.4% total weight compared to 4.2% in those who were not prescribed GLP-1 receptor agonists. VSG patients taking GLP-1 receptor agonists lost 14.1% of total body weight compared to 15.9% in those not prescribed GLP-1 receptor agonists. Though there was a trend toward greater weight loss weight on GLP1 agonists, these differences were not statistically significant, which could be explained by the small size of respective cohorts at 12 months (n = 16 for RYGB and n = 7 for VSG), limiting statistical power. Between these two cohorts, there were no statistically significant differences between number of prescribed AOMs, age, gender, number of comorbidities, starting weight, and starting BMI. Overall, 47% of the cohort lost 50% or more of regained weight, and 57% of patients lost >5 kg while in the program.

DISCUSSION

This medical weight loss center's weight management intervention program achieved clinically meaningful weight loss outcomes at 3, 6, and 12 months in post-bariatric surgery patients who had experienced weight regain. Patients lost 4.4 kg, 7.28 kg, and 10.7 kg at each respective time point, representing 13.0%, 24.3%, and 28.9% of excess body weight, respectively. As might be expected, patients prescribed three or more AOMs had greater weight loss than those individuals prescribed only one AOM, with this trend reflected at all time points (Figure 1). Interestingly, VSG patients achieved greater weight loss with AOM treatment at 12 months than RYGB patients, though the small number of patients at this time point may weaken the meaningfulness of this comparison. The findings from this analysis suggest that a multidisciplinary approach incorporating behavioral and nutritional counseling with pharmacologic therapy is effective at inducing weight loss in bariatric surgery weight regain patients.

The racial demographics in the cohort of patients referred for weight regain or suboptimal response after bariatric surgery differs from the bariatric surgery population in the same hospital system and deserves mention. The patients who undergo bariatric surgery in this surgical weight loss center are 84% white and 16% African American according to institutional data. The cohort in this study is comprised solely of patients referred from bariatric surgery. The patients who were referred from bariatric surgery comprising the cohort of this study were 70.4% African American, and 25.0% white. This difference was surprising, and suggests a racial disparity among those who experience weight regain or suboptimal response after bariatric surgery. Several studies have noted racial differences and ethnic variation in initial weight loss response to bariatric surgery and that African Americans achieve less weight loss after Roux-en-Y gastric bypass compared to Caucasians.^24–28 However, a race-associated risk factor to post-surgical weight regain has not been reported in the literature and warrants further research as African Americans might benefit from early AOM intervention in the post-operative time course.

Overall, patients lost a TBW of 3.9%, 6.9%, and 9.5% at the 3-, 6-, and 12-month time points with obesity medicine intervention. Whether above average responders to bariatric surgery within the first postoperative year also respond favorably to anti-obesity pharmacotherapy after weight regain is not known and requires further investigation. Weight loss between 5% and 10% of initial weight is clinically significant and is associated with improvements in cardiometabolic risk factors, hemoglobin A1c, fatty liver, and obstructive sleep apnea.^29,30 Notably, weight loss achieved in this cohort was greater than what has been reported in previous studies which focused on behavioral or psychosocial interventions alone.^10–12 This suggests that although bariatric surgery improves metabolic factors associated with obesity,³¹ weight regain has both behavioral and biological causes, and AOMs are often necessary to achieve clinically significant loss. This is supported by the finding that patients prescribed a greater number of AOMs achieved superior weight loss compared to those prescribed only one AOM (Figure 2). While this study did not compare bariatric patients with non-bariatric patients, another study making that comparison found that bariatric patients require more aggressive pharmacotherapy to achieve the same weight loss results as non-bariatric patients.¹⁴ Together, these findings may suggest that monotherapy is less effective in bariatric surgery patients, and more aggressive pharmacotherapy is indicated in this population.

Of note, there were some AOM weight loss differences by surgical operation (RYGB vs. VSG) that were unexpected and require further study. In the study cohort, weight loss among VSG recipients was greater at 12 months than that of RYGB recipients (open gastric bypass was excluded due to small n = 3 for that population). Research on the physiological effects of each type of surgery is still developing, and physiologic differences may explain the greater weight loss response found in VSG compared to RYGB patients in this study.³² Despite current research indicating both RYGB and VSG procedures have statistically similar weight loss and type 2 diabetes mellitus resolution, RYGB procedures are associated with superior management of dyslipidemia, hypertension, and gastroesophageal reflux disease.^20,21,33 These differences, along with the findings in this study, suggest the specific type of bariatric procedure may lead to distinctly different metabolic and physiologic profiles. One biologically apparent manifestation is differing postprandial levels of GLP-1; RYGB recipients have greater levels of postprandial GLP-1 than VSG recipients.²² Additionally, RYGB and VSG procedures may have differing effects on other satiety and gastrointestinal hormones such as ghrelin, cholecystokinin, glucose-dependent insulinotropic polypeptide (GIP), oxyntomodulin, or peptide YY, but the clinical and pharmacological implications of these differences are not fully understood.^34,35 Studies have also shown that gut hormone secretion differs in bariatric patients who have successful weight loss compared to those who have unsatisfactory loss or weight regain.^36,37 However, existing research has reported contradictory findings that show RYGB patients have superior weight loss after weight regain compared to VSG patients.^13,38,39 There are many confounding variables when examining these effects, including differences in pre-surgery BMI, nadir BMI, and time after surgery. How surgery type and its influence on gut hormone expression affects weight loss attempts and pharmacologic response after bariatric surgery weight regain is an area that requires further research. One hypothesis is that different post-surgical levels of GLP-1 in RYGB versus VSG might influence the weight loss effect with AOMs, but analysis did not find a statistically significant difference in the response of each surgery type to GLP-1 receptor agonists.

An important consideration in this study is medication access. GLP-1 receptor agonists were the most commonly prescribed AOM in our study. However, GLP-1 receptor agonists, especially higher dosage semaglutide, are expensive, non-generic medications frequently not covered by insurance. Since lower socioeconomic status is associated with greater incidence of obesity, this is a challenge to providing effective care.⁴⁰ Future research should examine this and other obstacles preventing access to weight-loss care in these patient cohorts.

This study has limitations. The study cohort was small (n = 44) and lacked the statistical power to conduct a meaningful multivariate analysis. One important factor in this multivariate analysis would be the relative efficacies of AOMs. GLP-1 receptor agonist based therapy has been shown to be more effective in post-bariatric weight regain than non-GLP-1 agonist therapy.⁴¹ Without a larger cohort for multivariate analysis, this study was unable to account for this effect. Further research with larger cohorts is required to comprehensively examine which factors predict the greatest weight loss among bariatric surgery patients presenting with weight regain. Additionally, the attrition rate (43.1% at 12 months) may have inflated weight loss results, as individuals who have less weight loss in the program may be more likely to drop out. This attrition rate also highlights the importance of emphasizing continuously supervised medical treatment. Moreover, socioeconomic factors such as drive time or transportation access have been shown to influence attrition among bariatric patients.⁴² The differing socioeconomic background of patients remaining in the study and patients who attritted may limit the applicability of the results. Finally, although patients were counseled on vitamin and mineral deficiencies and provided supplementation, adherence to anti-obesity medications was not assessed as a factor contributing to the patients' weight loss results. More research examining the modifiable factors that affect post-operative weight regain or subsequent attempts at weight loss is needed.

CONCLUSIONS

Continuous medically supervised multidisciplinary weight management can effectively induce weight loss in post-operative bariatric surgery patients presenting with weight regain. Aggressive pharmacotherapy may be needed to achieve optimal weight loss in this cohort. The unique hormonal balance of bariatric surgery patients, especially differences in baseline GLP-1 expression, and the potential differences in this balance depending on surgery type is an important target of further research. Understanding the metabolic and hormonal physiology of bariatric surgery patients is essential to understanding weight regain in this population and subsequently tailoring appropriate pharmaceutical treatment strategies.

Abbreviations
• AOM
• anti-obesity medications
• FDA
• Food and Drug Administration
• GIP
• glucose-dependent insulinotropic polypeptide
• GLP1
• Glucagon-like-1 agonist
• RYGB
• Roux-en-Y-Gastric Bypass
• VSG
• Vertical Sleeve Gastrectomy
• TBW
• total body weight
• TWL
• total weigh loss

AUTHOR CONTRIBUTIONS

Gunther Wong and Gitanjali Srivastava conceptualized, wrote and edited the manuscript. Gunther Wong participated in the design of the study, retrospective chart review and carried out the statistical analysis. All authors helped the design, coordination, drafting and editing of the manuscript.

ACKNOWLEDGMENT

We thank our Vanderbilt Weight Loss patients for including us part of their weight loss journey.

CONFLICT OF INTEREST

GS reports advisory/consultant fees from Novo Nordisk and Rhythm, outside the submitted work. GS is a Diplomate of the American Board of Obesity Medicine. GW, EMG, ST, MDS, WE, and VA have nothing to disclose.