Gastroesophageal Reflux Disease Symptoms after Laparoscopic Sleeve Gastrectomy: A Retrospective Study

(1) Background: Laparoscopic sleeve gastrectomy (LSG) is widely performed in bariatric surgery. However, the prevalence and risk factors of gastroesophageal reflux disease (GERD) symptoms after LSG remain unclear to date. This study aimed to identify risk factors of GERD after LSG. (2) Methods: We conducted a retrospective study at Linkou Chang Gung Memorial Hospital and reviewed 296 patients who underwent LSG from 2016 to 2019. A total of 143 patients who underwent preoperative esophagogastroduodenoscopy and completed the 12-month postoperative follow-up were enrolled. Patients’ demographic data, comorbidities, and postoperative weight loss results were recorded for analysis. The GerdQ questionnaire was used to assess GERD after LSG. (3) Results: There were eight surgical complications (5.6%) among the 143 studied patients (median age, 36 years; 56 (39.2%) men; median body weight 105.5 kg; median body mass index [BMI], 38.5 kg/m2). Twenty-three patients (16.1%) developed de novo GERD symptoms. GERD was significantly associated with older age (p = 0.022) and lower BMI (<35 kg/m2, p = 0.028). In multiple logistic regression analysis, age and BMI were significantly associated with GERD. (4) Conclusions: LSG is a safe and effective weight loss surgery. In our study, it led to 16.1% of de novo GERD symptoms, which were significantly related to older age and lower BMI (<35 kg/m2).


Introduction
Obesity is a modern-day global epidemic that is causing economic burden and a significant impact on health. In 2016, 39% of adults aged 18 years and older (39% of men and 40% of women) were overweight. Approximately 13% of the adult population (11% of men and 15% of women) were obese [1,2]. In Taiwan, a recent study showed that 47.97% of adults were overweight in 2020 [3].
Bariatric surgery is the most effective means of obtaining substantial and lasting weight loss in individuals with obesity [4]. Among the bariatric surgery options, laparoscopic sleeve gastrectomy (LSG) became the most frequently performed bariatric procedure to date because of its effectiveness in weight loss, relative simplicity, and fewer long-term nutritional problems than other bariatric procedures [5]. Sleeve gastrectomy was first introduced in 1988 by Hess as part of the biliopancreatic diversion duodenal switch procedure. In 1999, it was first performed laparoscopically and gradually became a stand-alone bariatric procedure [6].
Although LSG provides good weight loss results, there are controversial effects on de novo gastroesophageal disease (GERD) [7]. Previous studies reported that the incidence of new-onset GERD ranges from 0 % to 34.9% in Western countries [8], and the risk factors of de novo GERD after LSG are still controversial. Additionally, the incidence and risk factors of de novo GERD after LSG in current literature from the Asia-Pacific region remain unclear

Procedures
EGD was not a routinely recommended procedure in the workup for bariatric surgery. The American Society of Metabolic and Bariatric Surgery advised the use of endoscopy only for patients with significant gastrointestinal symptoms [10,11]. In our hospital, some surgeons preferred performing EGD for patients before surgery. LSG was the first considered bariatric procedure in our department for all morbidly obese patients. However, laparoscopic Roux-en Y gastric bypass (LRYGB) was recommended for patients with type II diabetes older than 8 years of age and those who needed insulin injections or a preoperative C peptide level less than 2 mmol/L. Furthermore, LRYGB was also recommended for patients with preoperative GERD symptoms, those with the need for prolonged proton pump inhibitor (PPI) treatment, those with evidence of GERD (grade C or D according to the Los Angeles [LA] classification), and those with a hiatal hernia due to preoperative EGD. Postoperatively, we measured the symptoms and signs and scheduled EGD at 12 months postoperatively if the patient had severe GERD symptoms.

Definitions
The weight loss result was calculated based on the percentage of excess weight loss (%EWL) and percentage of total weight loss (%TWL). Patients who reach %EWL ≥ 50 or %TWL ≥ 20% at postoperative month 12 were defined as achieving successful weight loss [12,13].

GerdQ Questionnaire
Most of the patients did not want to undergo invasive examination such as EGD or functional studies (pH-manometry) after surgery. We used more convenient tools to evaluate GERD, considering the cost effectiveness and time efficiency. GERD symptoms were assessed by administering the GerdQ questionnaire during the postoperative outpatient follow-up. The GerdQ questionnaire is a simple, self-administered and patient-centered questionnaire that includes six items [14]. It asks patients to score the number of days with symptoms and use of over-the counter (OTC) medications during the previous 7 days. It uses a four-point Likert scale (0-3) to score the frequency of four positive predictors of GERD (heartburn, regurgitation, sleep disturbance due to reflux symptoms, or use of OTC medications for reflux symptoms) and a reversed Likert scale (3-0) for two negative predictors of GERD (epigastric pain and nausea); the total GerdQ score ranges from 0 to 18 [14]. Patients completed the GerdQ questionnaire at 12 months after LSG. De novo GERD is defined as a GerdQ score ≥9 as per literature [14].

Laparoscopic Sleeve Gastrectomy
LSG was performed using the same technique between January 2016 and December 2019. A 32-French bougie was applied for calibration of the gastric tube. Dissection was started 4 cm proximal to the pylorus and proceeded to 1 cm from the angle of His. We did not check the hiatus routinely since the presence of a hiatal hernia was excluded by preoperative EGD. All patients received seromuscular suturing for staple line reinforcement, and no leak test was performed routinely intra-and postoperatively. Patients were instructed to follow a clear liquid diet on the first postoperative day, followed by a pureed diet on the second day. In general, patients were discharged on postoperative day 3, if they had stable vital signs and adequate pain control.

Statistical Analysis
Data are presented as median (interquartile range) or percentages. Patients' characteristics were compared using the Mann-Whitney U test for nonparametric data and the chi-square test for categorical data. Logistic regression analysis was performed to calculate odds ratios (ORs) for predictive factors. When the data reached statistical significance in invariant analysis, multivariate analysis with logistic regression was conducted to determine the relationship between predictive factors and de novo GERD symptoms. Data were analyzed using SPSS Statistics, version 26 (IBM Corp.). Variables with a 95% CI for the OR that did not include 1.0 and p-values < 0.05 were considered to be significant.

Patient Characteristics
In total, 384 patients who received bariatric surgery at Linkou CGMH between January 2016 and December 2019 were screened. Among those patients, 143 patients were eligible for the study. The study flow chart is shown in Figure 1.
medications for reflux symptoms) and a reversed Likert scale (3-0) for two negative predictors of GERD (epigastric pain and nausea); the total GerdQ score ranges from 0 to 18 [14]. Patients completed the GerdQ questionnaire at 12 months after LSG. De novo GERD is defined as a GerdQ score ≥9 as per literature [14].

Laparoscopic Sleeve Gastrectomy
LSG was performed using the same technique between January 2016 and December 2019. A 32-French bougie was applied for calibration of the gastric tube. Dissection was started 4 cm proximal to the pylorus and proceeded to 1 cm from the angle of His. We did not check the hiatus routinely since the presence of a hiatal hernia was excluded by preoperative EGD. All patients received seromuscular suturing for staple line reinforcement, and no leak test was performed routinely intra-and postoperatively. Patients were instructed to follow a clear liquid diet on the first postoperative day, followed by a pureed diet on the second day. In general, patients were discharged on postoperative day 3, if they had stable vital signs and adequate pain control.

Statistical Analysis
Data are presented as median (interquartile range) or percentages. Patients' characteristics were compared using the Mann-Whitney U test for nonparametric data and the chi-square test for categorical data. Logistic regression analysis was performed to calculate odds ratios (ORs) for predictive factors. When the data reached statistical significance in invariant analysis, multivariate analysis with logistic regression was conducted to determine the relationship between predictive factors and de novo GERD symptoms. Data were analyzed using SPSS Statistics, version 26 (IBM Corp.). Variables with a 95% CI for the OR that did not include 1.0 and p-values < 0.05 were considered to be significant.
Regarding the results of the GerdQ questionnaire in postoperative month 12, 23 patients (16.1%) were diagnosed with de novo GERD symptoms after LSG. The risk factors for predicting de novo GERD symptoms are shown in Table 2. In invariant analysis, de novo GERD symptoms were not associated with tobacco use and alcohol consumption. However, de novo GRED symptoms were associated with older age (40 years versus [vs.] 35 years, p = 0.005) and lower BMI (<35 kg/m 2 vs. ≥35 kg/m 2 , p = 0.019). Preoperative comorbidities, such as hypertension, diabetes mellitus, and fatty liver, and the postoperative weight loss result were not associated with de novo GERD symptoms. Age and preoperative BMI were included in multivariable analysis with logistic regression. The results are shown in Table 3. De novo GERD symptoms were significantly associated with older age (OR = 1.1; p = 0.022) and lower BMI (OR = 3.1; p = 0.028).

Discussion
The main findings of our study are as follows.
(1) After LSG, the incidence of de novo GERD symptoms was 16.1%. (2) Older patients and those with a lower preoperative BMI have significant risk of de novo GERD postoperatively. LSG is effective in terms of weight loss, relative simplicity, and fewer long-term nutritional complications than other bariatric surgeries [6]. Therefore, LSG became the current leading bariatric procedure [6].
Various modalities for evaluating GERD after LSG were reported, and the incidence of GERD after LSG is up to 34.9% [8]. In 2012, Chopra et al. reported a postoperative GERD rate of 3.7% by chart review [15]. In a postoperative EGD study, Soricelli et al. reported an incidence of 15.9% in 2013 [16]. In another study, the barium swallow, GerdQ questionnaire, and 24 h pH-manometry were applied, and the incidence was approximately 11% [17][18][19]. In our study, we evaluated de novo GERD by the GerdQ questionnaire and found an incidence of 16.1%. The GerdQ questionnaire is a simple, non-invasive, self-administered, patient-centered tool for the diagnosis of GERD. The implementation of GerdQ could reduce the need for upper endoscopy and improve resource utilization [14].
The putative pathophysiological mechanisms of de novo GERD after LSG include hypotensive lower esophageal sphincter (LES), an increased gastroesophageal pressure gradient, intrathoracic sleeve migration, and relative gastric stasis in the proximal remnant [2,20,21]. De novo hiatal hernia of the gastric tube after sleeve gastrectomy, which is also named intrathoracic sleeve migration, confirmed causes of GERD [22,23]. We did not routinely perform EGD or barium swallowing for the diagnosis of de novo GERD. However, in some patients with persistent and severe symptoms of de novo GERD, EGD and barium swallowing showed a newly developed hiatal hernia, which was not noted preoperatively.
Similar to the treatment of GERD patients without bariatric procedure, the initial management of those with de novo GERD include dietary and lifestyle modification, alcohol and smoking cessation, and PPI prescription, although most symptoms of de novo GERD symptoms cannot be controlled by PPIs. In our study, 87 patients needed PPIs in postoperative month 3, and 33 patients stopped taking them in postoperative month 12. Fourteen patients had GERD symptoms and five patients were well controlled with PPIs over the next few months.
For patients with severe symptoms who failed conservative management, several endoscopic procedures, such as plication, bulking the LES with inert biopolymers, and thermal ablation, were used for treating de novo GERD [24]. These procedures targeted the muscular layer of the LES through endoscopic means to decrease esophageal sensitivity to acid and reduce gastroesophageal junction compliance [24,25]. When it comes to surgical options, revision surgery can correct anatomic abnormalities (i.e., hiatal hernia, intragastric migration of the sleeve, distal stenosis or torsion angulation of the sleeve, and retained fundus) [26,27]. LRYGB is the most effective surgical treatment for GERD in patients with obesity [25]. LRYGB can almost completely resolve GERD and de novo GERD symptoms [25,28].
The aforementioned treatment should be openly discussed with the patients when making decisions about surgery, medications, or self-management. The opportunity to prevent such patients from having persistent gastroesophageal reflux should be seized, and the available evidence of treatment should be openly discussed with the patients.
The main finding of our study was that preoperative BMI and age were risk factors for de novo GERD after LSG. Population-based studies also found an association between GERD and higher BMI. Similar to our study, Althuwaini et al. reported that 213 patients who underwent LSG and had a high preoperative BMI were less likely to develop newonset or worsening symptoms of GERD [29]. However, two other studies showed no significant association between the preoperative BMI and development of de novo GERD after LSG [30,31]. The association with preoperative BMI has to be interpreted with caution, since it was not the primary endpoint of our study. Nonetheless, this finding should be explored in future studies.
One study that included 326 patients reported the same result for older age and the occurrence of postoperative GERD symptoms that our study showed [32]. Age was associated with an increase in esophageal acid exposure. Aging is attributed with a progressively decreased abdominal LES length and esophageal motility. The severity of GERD increased in the elderly population due to degradation of the gastroesophageal junction and impaired esophageal clearance [33]. The physiological change supports ours finding and addresses the effect of age on developing de novo GERD after SLG.

Limitations
This study has some limitations. First, we used the GerdQ questionnaire to evaluate postoperative GERD symptoms. It is not objective and is not the gold standard test for diagnosing GERD. Patients with silent reflux or delayed onset of reflux were potentially missed. Although EGD or 24 h pH-manometry is the gold standard examination, considering its cost effect and time efficiency, we did routinely perform such examinations, except in patients with severe GERD symptoms. Second, inconsistent assessment before and after operation made evidentiary uncertainty. In the future, we will design a more rigorous method to avoid different inspections for comparison. Third, our study population was from a single institution; thus, it cannot represent the whole Asian-Pacific population, and this factor may cause bias in the results. Third, our study was a retrospective study and only had a 1-year follow-up period. Further prospective studies and a long-term follow-up period are needed in the future to obtain a more convincing conclusion.

Conclusions
In conclusion, LSG is a safe and effective weight loss surgery. However, further prospective studies are needed to determine the risk factors associated with post-LSG GERD and overall quality of life.