Abstract
Background:
Laparoscopic sleeve gastrectomy (LSG) is the most common primary bariatric operation performed in the United States. Its relative technical ease, combined with a decreased risk for anatomic and malabsorptive complications make LSG an attractive option compared to laparoscopic gastric bypass (LGB) for many patients and surgeons. However, emerging evidence for progressive gastroesophageal reflux disease (GERD) after LSG, and the inferior weight loss in many studies, suggests that the enthusiasm for LSG requires reassessment. We hypothesized that patient satisfaction and quality of life (QoL) may be lower after LSG compared to LGB because of these differences.
Methods:
We distributed a survey querying weight-loss outcomes, complications, foregut symptoms, QoL, and overall satisfaction to patients who underwent bariatric operations at our institution between 2000 and 2020 and who had electronic mail contact information available. Mean follow-up was 2.75 ± 2.41 years for LGB patients and 3.37 ± 2.18 (P = .021) years for LSG patients. We compared these groups for weight-loss outcomes, changes in foregut symptoms, gastrointestinal QoL, postbariatric QoL, and overall satisfaction using appropriate statistical tests.
Results:
Among 323 respondents, 126 underwent LGB and 197 underwent LSG. LGB patients had larger body mass index (BMI) reduction than LSG patients (−17.16 ± 9.07 kg/m2 versus −14.87 ± 7.4 kg/m2, P = .023). LGB patients reported less reflux (P = .003), with decreased heartburn (P < .0001) and regurgitation (P = .0027). However, a greater proportion of LGB patients reported at least one complication (P = .025). Despite this, more LGB patients reported satisfaction (92.86%) than LSG patients (73.6%).
Conclusion:
LGB patients are significantly more likely to be satisfied than LSG patients. Factors contributing to the higher level of satisfaction include less GERD and better BMI decrease.
Background
According to the World Health Organization, the number of obese individuals has increased dramatically in recent years, with more than 650 million adults obese in 2016. 1 Obesity is associated with many comorbid conditions that impact the duration and quality of life (QoL). Obese patients are more prone to myocardial infarction, stroke, osteoarthritis, and cancers of several organs, including endometrium, breast, ovary, prostate, liver, gallbladder, kidney, and colon. 1 The worldwide obesity epidemic, along with the accumulating morbidity and mortality associated with the disease, makes treatment improvements imperative, whether behavioral, medical, or surgical.
Behavioral treatments of obesity include exercise, diet, and psychological support. Unfortunately, long-term weight-loss outcomes are inadequate for most, limiting the value of these measures in isolation for those needing durable solutions.2,3
Several medications for the treatment of obesity have been developed and studied. Common options include orlistat, sibutramine, and rimonobant. Unfortunately, most of the studies on such agents are limited by attrition, which calls into question the assessment of the effectiveness of these drugs in the long term. 4
Bariatric surgery is the contemporary obesity treatment option that provides the best enduring weight loss and comorbidity improvement. Therefore, the number of bariatric operations has been increasing over recent years.5,6 In addition, surgical treatment of obesity has proven more cost effective compared to other treatment approaches.7,8 Many different operative procedures have been developed, but currently the two most prominent are laparoscopic gastric bypass (LGB) and laparoscopic sleeve gastrectomy (LSG). 5
LSG, according to recent statistics derived from the International Federation for the Surgery of Obesity and Metabolic Disorders (IFSO), is now more commonly performed than LGB. 5 This increase in utilization is likely related to the procedure's comparative technical simplicity, decreased risk of dumping syndrome and micronutrient deficiencies, and lack of certain potentially catastrophic anatomic complications such as internal hernias and marginal ulcers, compared to LGB. However, LSG has been shown to have significant risk of aggravating preexisting gastroesophageal reflux disease (GERD), or in some cases inciting the development of de novo GERD. 9 In addition, weight-loss outcomes are inferior to LGB in many studies. In some situations, patients are requiring revisional procedures.
On the contrary, LGB is a more technically complex procedure requiring more extreme lifestyle limitations and carrying an increased risk of complications.6,10 Despite this, LGB is noted to have better outcomes in terms of weight loss.9,11
Therefore, it is important to continue to compare long-term outcomes and patient satisfaction to better inform patients during the shared decision-making process regarding bariatric options. Gathering data from a large single-center case series minimizes some of the biases that occur from aggregated data from multiple centers and from surgeons at various experience and skill levels.
Methods
From July 21, 2000 to April 1, 2020, 2328 patients had bariatric surgery at the University of North Carolina at Chapel Hill. Of these, 808 had an electronic mail (e-mail) contact recorded in the medical record. With Institutional Review Board approval, a 20-minute electronic survey, which consisted of 78 questions, was electronically distributed to all of them on April 22, 2020 (Supplementary File S1). Twenty-six e-mails were undeliverable, leaving a total of 782 patients whose response was pending. Reminder e-mails were sent on April 25, 2020, May 2, 2020, and May 9, 2020 to nonresponders. The survey closed 2 weeks later, on October 4, 2020. At the end of the data-collection phase, 323 patients responded to the electronic survey (126 LGB, 197 LSG) for a response rate of 41.3%. All respondents had a laparoscopic operation, with no open conversions.
Of the 126 respondents in the LGB group, 110 (87.30%) were female. Of the 197 respondents in the LSG group, 170 (86.29%) were female. The preoperative body mass index (BMI) in the LGB arm was 47.36 ± 6.94 kg/m2, and in the LSG arm 49.66 ± 9.79 kg/m2. Patient characteristics, BMI, and weight variations are reported in Table 1. Patient follow-up time was 2.75 ± 2.41 years for the LGB respondents and 3.37 ± 2.18 years for the LSG respondents. The survey assessed preoperative and postoperative foregut symptoms, as well as gastrointestinal (GI) QoL12,13 and bariatric QoL. 14
Demographic Characteristics, Body Mass Index, and Weight Changes
BMI, body mass index; LGB, laparoscopic gastric bypass; LSG, Laparoscopic sleeve gastrectomy.
To compare the treatment effect between the two operations, we used the Welch's two-sample t-test for continuous responses, such as BMI, the Cochrane-Mantel-Haenszel test for ordinal responses and the chi-square test for binary responses such as satisfaction. Regression analysis was carried out to compare the weight loss and BMI reduction controlling for the preoperative measurements. All continuous values are expressed as mean ± standard deviation (SD), whereas all categorical values are expressed as total number and percentage (%).
The upper GI symptom survey looked specifically at esophageal symptoms such as regurgitation, reflux, and dysphagia. We collected information regarding frequency of symptoms before and after the bariatric procedure. We then categorized the change of symptoms using ordinal variables with three levels: “worse,” “the same” and “improved.”
For GI-QoL, we incorporated in our survey the questionnaire proposed by Eypasch et al 13 and by Decker et al. 12 That questionnaire consists of 37 questions assessing GI symptoms (Q43–Q78 in our survey). Possible responses range from “never” to “all of the time” and were scored on a scale of 1 to 5. We calculated a cumulative GI-QoL score for each patient, and the mean and SD was determined for each arm and compared by Welch's t-test.
For postbariatric QoL, we integrated into the survey the tool proposed by Moorehead et al, 14 which asked respondents to assess the change they experienced postoperatively in five crucial domains of everyday life (Q38–Q42 in our survey). Answers to these items ranged from “much less” to “much more” scored on a scale of 1 to 5. We calculated a cumulative score for each patient and then summarized the mean and SD for each arm. The mean scores between groups were compared using the chi-squared Welch's t-test. 15
Results
Our study demonstrates several differences that favor LGB over LSG. The results of the questionnaire are cumulatively presented in Table 2, whereas the most important findings of our study are reported in Table 3. LGB patients lost nearly 12 pounds more (P = .036) and 3 BMI units more (P = .022) than LSG patients. Regression analysis demonstrated that after controlling for the preoperative weight variation, LGB is associated with significantly more weight loss (P < .0001) and BMI reduction (P < .00010) compared to LSG (Tables 4 and 5). However, the percentage of patients reporting any complication was higher in the LGB group (P = .025). Despite that, LGB patients reported a significantly higher satisfaction rate (P < .0001), with 92.86% of patients satisfied compared to the 73.6% of those in the LSG group.
Proton Pump Inhibitor Use, Change in Comorbidities, and Change in Foregut Symptoms
LGB, laparoscopic gastric bypass; LSG, Laparoscopic sleeve gastrectomy; PPI, proton pump inhibitor.
Important Study Findings
BMI, body mass index; GI, gastrointestinal; LGB, laparoscopic gastric bypass; LSG, Laparoscopic sleeve gastrectomy; QoL, quality of life; SD, standard deviation.
Regression Analysis on Body Mass Index
Multivariate linear regression results associating BMI and surgery type, controlling for the confounding effect of the baseline BMI. We found that controlling for the baseline BMI, LGB further decreases the BMI by 3.25 U (P < .0001) compared to LSG.
LGB, laparoscopic gastric bypass; LSG, laparoscopic sleeve gastrectomy; Pre-BMI, preoperative body mass index.
Regression Analysis on Weight
Multivariate linear regression results associating weight and surgery type, controlling for the confounding effect of the baseline weight. We found that controlling for the baseline weight, LGB further decreases the weight by 18.9 pounds (P < .0001) compared to LSG.
LGB, laparoscopic gastric bypass; LSG, laparoscopic sleeve gastrectomy; preweight, preoperative weight.
A greater percentage of LSG patients reported the existence of reflux symptoms (P = .0027), whereas more LGB patients reported improvement of heartburn (P < .0001), chest pain (P = .0011), and regurgitation (P < .0001) after operation. However, no statistically significant difference was noted in postoperative proton pump inhibitor use (P = .7279). Interestingly, both groups reported similarly improved GI-QoL (P = .8966) and postbariatric QoL (P = .082).
Discussion
Bariatric surgery improves QoL in patients, as they experience rapid and sustainable weight loss and reduction of comorbidities.9,11,16,17 LSG, according to recent statistics, has become the most commonly done bariatric operation. 5 This procedure was initially described as the first stage of a two-stage biliopancreatic diversion in the most challenging morbidly obese patients. 18 It was recognized that the second stage was not always required, as LSG alone offered satisfactory results.19,20 The lack of data regarding the long-term outcomes of LSG remain the main source of uncertainty regarding the right degree of implementation for this operation.
LGB has long been the “gold standard” in obesity treatment. The complexity and high technical proficiency required by surgeons has resulted in limitation of its utilization. 5 In addition, LGB has been associated with an increased incidence of both early and late postoperative complications.19,21,22 Nevertheless, it is believed that technical advancements, such as the standardization of the mesenteric defect closure as part of the procedure, and the improvements in training paradigms for bariatric surgeons, are reducing complication rates.23,24 Furthermore, LGB has shown better weight-loss outcomes in many studies.9,16,25,26
Several meta-analyses have compared the two most commonly performed bariatric operations; however, there are still many discrepancies within the existing literature that limit strong conclusions. Some of these variances in the outcomes reported in component studies raise concern for ongoing bias from surgeons and centers that are at different skill and quality levels due to the rapid expansion in the field.
In our single-center series, where a small group of surgeons has maintained a bariatric surgery practice over 20 years, there was a statistically significant difference in weight-loss outcomes favoring LGB. These findings are in accordance with the results of the recent meta-analysis performed by Sharples and Mahawar, 9 a study which analyzed randomized control trials comparing gastric bypass to sleeve gastrectomy.
Improvements in GI QoL and postbariatric QoL were high in both groups. These results are again consistent with the current literature. Specifically, a comprehensive meta-analysis performed by Hu et al 16 comparing LGB to LSG demonstrated that QoL was similarly improved after both operations.
A higher proportion of our LGB patients reported complications compared to our LSG patients. This difference is also in accordance with the current literature comparing complication rates after the two procedures. The meta-analysis led by Hu et al 16 found that LGB patients had more early (odds ratio [OR] = 2.11, 95% confidence interval [CI] = 1.51–2.91, P < .001) and late (OR = 2.60, 95% CI = 1.93–3.49, P < .001) complications. Similar results were reported in the meta-analysis conducted by Han et al 27 (OR = 1.59, 95% CI = 1.22–2.06).
An important finding of our study is the higher level of satisfaction in the LGB group compared with the LSG group (92.86% versus 73.6%). We believe that this variation relates to the difference in the symptoms of heartburn (P < .0001), chest pain (P = .0011), regurgitation (P < .0001), and increased satisfaction with superior weight loss. A meta-analysis performed by Sharples and Mahawar, 9 a randomized controlled trial performed by Peterli et al, 28 and a comparative study performed by Chouillard et al 29 all demonstrated that LSG may lead to either progression of existing GERD symptoms or de novo development of GERD in those not previously afflicted. Similar claims have been reported in one-armed studies assessing the long-term outcomes of LSG by Genco et al 30 and Felsenreich et al. 31
In summary, patients having LGB are more satisfied than LSG patients despite higher report of complications. Better weight loss and reduced GERD likely account for this difference. Improved patient selection, ongoing optimization of technical elements, and better perioperative management of LGB patients will likely bring complication rates even closer to those of LSG. Examples of process modifications implemented in our practice include (1) extended thromboprophylaxis using a risk calculator 32 to limit VTE, (2) a 6-month course of antisecretory medications to reduce the occurrence of marginal ulcers, and (3) standardized closure of all mesenteric defects with barbed, running suture to limit internal hernia risk, which is supported in the literature. 33 Since most of these practice improvements are directed at complications specific to LGB, such measures will likely amplify the difference in patient satisfaction noted in this report.
Potential limitations
Patients were solicited by e-mail to complete an online questionnaire. E-mail has been collected in our medical record for the last 10 years; therefore, unless earlier patients had continuing care at our institution during that time, they would not have been included. However, skewing the study toward the latter period in our series has allowed the comparison to exclude the learning curve periods of the surgeons involved and to incorporate the benefits of our multidisciplinary center of excellence.
Individual comorbidity improvement was not considered a primary or secondary endpoint of this survey study. Such comparative outcomes between the common bariatric procedures have been extensively reported in the literature. The purpose of this article was to evaluate patient-reported outcomes, satisfaction, and QoL. Patient-reported data are not the best means to capture the nuances of improvements in comorbidities, and therefore were not queried. During this period, the average BMI and rates of comorbidities among our bariatric patients were similar to, or higher than, averages for peer institutions reporting through the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP), with average BMI of 47.58 kg/m2, and 31% type 2 diabetes, 54% hypertension, 28% hyperlipidemia, 22% sleep apnea, and 31% GERD. At various follow-up intervals, comorbidities responded similar to our peer institutions.
Likewise, patient-reported data are not a reliable way to associate complications and outcomes. Inaccuracies and omissions are possible. When we reviewed 89 charts of patients reporting a complication, we found 40 (45%) did not have a true complication by Clavien-Dindo 34 classification, but instead minor issues such as postoperative nausea, low vitamin levels, or minor GERD symptoms. Review of the other charts found 11 (12.4%) had Clavien-Dindo Grade 1 complications, 6 (6.7%) had Clavien-Dindo Grade 2 complications, and 32 (35.9%) had Clavien-Dindo Grade 3 complications. The Clavien-Dindo Grade 3 complications included 26 reoperations. Of these, 18 were months-to-years later for bariatric revision for GERD or weight regain (11), internal hernia (2), bowel obstruction or intussusception (2), perforated marginal ulcer (1), ventral hernia (1), and sleeve stricturoplasty (1).
In addition, there were eight early reoperations (within 1 month) for obstruction at the jejunojejunostomy (3), bleeding (2), sleeve leak (1), superior mesenteric vein thrombosis associated jejunal stricture (1), and surgical remnant gastrostomy for nutrition support (1). There were also five endoscopic dilations of gastrojejunal strictures. There was one radiologic decompression of an acute gastric volvulus in the remnant stomach after LGB. Since it is possible that events may have taken place outside of our institution and would not have been captured in our electronic medical record, we could not accurately assess associations between operation performed and complications.
Over the years, there has been an evolution in the shared decision-making process such that patients with more severe diabetes, GERD, or sweet-eating tendencies are generally encouraged toward LGB, whereas those with prior complex abdominal surgical history, super obesity with central distribution, present or future need for immunosuppression, intended future pregnancies, and poor medical literacy are directed toward LSG. Therefore, we make no claim that comorbidities and baseline characteristics distribute evenly across the study groups. Our finding that GERD symptoms were more prevalent in the LSG patients is therefore even more important given this selection strategy.
The 2.75-year mean follow-up might be conceived as a potential limitation of our study, however, according to recent data, symptoms and endoscopic manifestations of reflux tend to present before 18 months after operation.35,36
Another potential limitation of the study is that the patients were asked to evaluate the presence of GERD symptoms based on their subjective experience. Patients with “silent GERD,” could be missed by this approach. We are presently gathering one-year postoperative endoscopy data to capture additional objective evidence of silent GERD.
Conclusions
Despite the current trends in utilization favoring LSG over LGB, our data suggest that there are poorer weight-loss outcomes and higher rates of GERD symptoms, with associated relatively lower levels of patient satisfaction. Even though a larger proportion of patients report complications after LGB, the higher degree of patient satisfaction persists. Therefore, we have increased our utilization of LGB in recent years, targeting individuals with any GERD history and higher BMIs, and using protocol-driven risk reduction strategies such as rigorous closure of mesenteric windows and aggressive thromboprophylaxis.
Footnotes
Authors' Contributions
D.N.V.: writing—original draft and writing—review and editing. J.N.L.: conceptualization, writing—original draft and writing—review and editing. L.L.: formal analysis, data curation, writing—original draft and writing—review and editing. M.S.C., R.Z., and M.C.D.: writing—review and editing. T.M.F.: conceptualization, writing—original draft, writing—review and editing, supervision, and project administration.
Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received for this article.
References
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.
