Obesity Surgery

, Volume 21, Issue 3, pp 295–299

Symptoms of Gastroesophageal Reflux Following Laparoscopic Sleeve Gastrectomy Are Related to the Final Shape of the Sleeve as Depicted by Radiology


  • Olga Lazoura
    • Department of RadiologyUniversity Hospital of Larissa
  • Dimitris Zacharoulis
    • Department of SurgeryUniversity Hospital of Larissa
  • George Triantafyllidis
    • Department of RadiologyUniversity Hospital of Larissa
  • Michael Fanariotis
    • Department of RadiologyUniversity Hospital of Larissa
  • Eleni Sioka
    • Department of SurgeryUniversity Hospital of Larissa
  • Dimitris Papamargaritis
    • Department of SurgeryUniversity Hospital of Larissa
    • Department of SurgeryUniversity Hospital of Larissa
Clinical Research

DOI: 10.1007/s11695-010-0339-0

Cite this article as:
Lazoura, O., Zacharoulis, D., Triantafyllidis, G. et al. OBES SURG (2011) 21: 295. doi:10.1007/s11695-010-0339-0



Laparoscopic sleeve gastrectomy (LSG) is gaining popularity as a primary procedure in selected morbidly obese patients. Like most other bariatric procedures LSG results in alterations of the upper GI anatomy that might affect gastroesophageal reflux postoperatively. The study was conducted to assess the presence of reflux symptoms in patients before and after laparoscopic sleeve gastrectomy and any possible relation of these symptoms to the postoperative gastric anatomy as depicted by gastrografin swallow studies.


The study included 85 consecutive morbidly obese patients who underwent LSG as a primary bariatric procedure. Patients were evaluated for symptoms of gastroesophageal reflux (heartburn, regurgitation, and vomiting) preoperatively and at 1 and 6 months and 1 year postoperatively. To assess the postoperative gastric anatomy, the gastrografin studies that were routinely performed in all patients on the third postoperative day were retrospectively evaluated. Changes of each one of the reflux symptoms were assessed in relation to the radiological pattern of the gastric sleeve.


Three radiological patterns of the gastric sleeve were identified: (a) the tubular (65.9%), (b) the superior pouch (25.9%), and (c) the inferior pouch pattern (8.2%). Patients showed an overall tendency towards relief of heartburn and increase of regurgitation and vomiting postoperatively. However, only changes in regurgitation and vomiting were found to be statistically significant (p < 0.01); interestingly, those were observed in patients with the tubular gastric pattern.


The final shape of the gastric sleeve as depicted by radiological studies seems to have an impact on reflux symptoms after laparoscopic sleeve gastrectomy.


Sleeve gastrectomyRefluxGastrografin swallow


Obesity is a significant independent risk factor for gastroesophageal reflux symptoms [13]. Symptomatic reflux is also present in many patients following bariatric procedures, especially vertical banded gastroplasty and laparoscopic adjustable gastric banding [4]. In contrast, Roux-en-Y gastric bypass is essentially an antireflux procedure [5].

Recently, laparoscopic sleeve gastrectomy (LSG) has gained popularity as a primary operation for the treatment of morbid obesity [68]. During the procedure, the greater curvature of the stomach is removed from the antrum to the angle of His and a long tubular gastric sleeve is created [9]. These anatomical changes are expected to have an impact on upper gastrointestinal motility, and therefore, they might affect gastroesophageal reflux. Indeed, there is evidence that LSG induces reflux in the first postoperative year [1016] and that the symptoms improve over time [10, 15], although conflicting data have also been published [17, 18]. However, the exact relation of postoperative gastric anatomy to reflux symptoms has not been investigated. The anatomy of the gastric sleeve has been described in upper gastrointestinal gastrografin swallow studies (UGI), routinely performed between the first and fourth postoperative day to exclude early complications and permit gradual liquid intake [1922]. In fact, various anatomical patterns of the gastric sleeve that reflect variations or limitations of the surgical technique have been identified [21].

In this study, we assessed the presence of reflux symptoms in morbidly obese patients before and after LSG and any possible relation these symptoms bear to the anatomical patterns of the gastric sleeve as depicted by UGI studies.

Materials and Methods

The study included 85 consecutive morbidly obese patients (71 women, 83.5%), aged 18–62 years, who underwent LSG as a primary and potentially definite bariatric procedure between August 2006 and September 2009 at our center. Inclusion criteria for surgery were according to the 1991 NIH consensus criteria for bariatric surgery [23]. Exclusion criteria were (a) heavy sweeters and (b) patients with suspected gastroesophageal reflux disease (GERD) as suggested by severe symptoms and endoscopic findings. The procedure was performed by the same team consisting of two surgeons experienced in advanced laparoscopic surgery. The demographic characteristics of the patients are listed in Table 1. The surgical technique used has been previously described [24].
Table 1

Demographic characteristics of patients who underwent laparoscopic sleeve gastrectomy




n (%)

15 (17.6)

70 (82.4)

Mean age: years (range)

39.7 (25–58)

39.8 (18–62)

Mean BMI: kg/m2 (range)

43.8 (37–61)

42.3 (35–58)

Evaluation of Reflux Symptoms

Upper GI endoscopy was routinely performed in all patients preoperatively, in order to detect esophagitis or any other abnormal finding precluding those patients from the procedure. As patients with suspicion of gastroesophageal reflux disease—on the basis of their symptoms and endoscopic findings—were excluded from proceeding to the procedure and, thus, are not included in this series, no pH-metry studies were performed preoperatively. A detailed questionnaire regarding symptoms and medications for reflux at preoperative evaluation and postoperative visits at 1 and 6 months and 1 year postoperatively was completed for all patients. Based on their response, the patients were classified according to the score for each of the three symptoms of heartburn, regurgitation, and vomiting (Table 2).
Table 2

Scoring system for reflux symptoms evaluation


0: no symptoms

1: <2 episodes per week

2: >2 episodes + PPIs (proton pump inhibitors)

3: esophagitis grade > II (LA classification)a

4: complications (stenosis, ulcer, Barrett’s)a


0: no symptoms

1: <2 episodes per week

2: >2 episodes per week, no medication

3: >2 episodes per week + medications (i.e., prokinetics)


0: no symptoms

1: <2 episodes per week

2: > episodes per week

aPatients with a preoperative heartburn score of 3 or 4 are not included in this series as described in the “Materials and Methods” section

Evaluation of Gastric Sleeve Anatomy

A UGI study was routinely performed in each patient on the third postoperative day to exclude leaks and strictures. We retrospectively evaluated these studies to detect the presence of one of the five radiological patterns of the gastric sleeve that have been described by Werquin et al. [21]. These patterns include the tubular, the superior pouch, the inferior pouch, the inferior–superior pouch, and the pseudodiverticular pattern. The tubular pattern characterizes an almost tubular gastric sleeve with homogeneous opacification after ingestion of contrast medium. The superior pouch pattern describes a wider superior compartment of the sleeve near the gastroesophageal junction, which is gradually filled with a contrast agent creating an air fluid level. The inferior pouch pattern represents a wider compartment in the antral region. The superior–inferior pouch pattern is characterized by the presence of two pouches in the fundus and the antrum. Finally, the pseudodiverticular pattern describes a diverticular dilatation of the lesser curvature of the gastric body.

Statistical Analysis

The Friedman test was used to assess changes of symptoms for each pattern in the first year following surgery. The Wilcoxon signed-rank test was used to determine differences between pairs where the Friedman test was significant. Bonferroni correction was set at 0.01.


Three distinct patterns of the gastric sleeve were identified at UGI in this series: (a) the tubular, (b) the superior pouch, and (c) the inferior pouch pattern. The tubular pattern was observed in 56 patients (65.9%), the superior pouch pattern in 22 patients (25.9%), and the inferior pouch pattern in 7 patients (8.2%).
Fig. 1

Alterations in heartburn

Changes in heartburn during the first postoperative year are shown in Fig. 1. Although there seems to be a tendency towards heartburn relief in all patients, the changes were not significant (p > 0.01) for any of the patterns.

Alterations in regurgitation are shown in Fig. 2. A tendency towards symptom deterioration is noted in patients with the tubular and superior pouch patterns. However, the changes are significant (p = 0.008) only for patients with the tubular pattern.
Fig. 2

Changes in regurgitation

Changes in vomiting are presented in Fig. 3. A tendency towards vomiting increase in the first postoperative month and decrease afterwards is noted in patients with the tubular and superior pouch patterns. However, significant changes (p < 0.001) are only seen in patients with the tubular pattern. These patients experience a significant increase in vomiting during the first month, which significantly decreases again by completion of the first year. Nevertheless, vomiting is still significantly higher at 1 year postoperatively compared to the preoperative state.
Fig. 3

Changes in vomiting


From the physiology and motility point of view, the fundus and body of the stomach are responsible for the receptive relaxation mechanism and food accommodation; the antrum is mainly responsible for trituration and emptying. Sleeve gastrectomy interferes with both mechanisms, and therefore, it is expected to have a significant impact on upper gastrointestinal motility.

Thus far, data published on the effect of LSG on reflux symptoms are controversial. Most studies have reported an increase in the incidence of reflux symptoms during the first year following LSG [1016] and a gradual decrease afterwards up to the third postoperative year [10, 15]. This has been mainly attributed to anatomical changes at the esophagogastric junction that have been considered to impair the antireflux mechanism of the cardia [13, 25, 26]. The sling fibers of the lower esophageal sphincter (LES) are often resected during the procedure, as found in pathology specimens after sleeve gastrectomy, and this has been associated to a hypotensive LES [27, 28]. Indeed, in a recent manometric study, an important decrease in LES pressure following LSG was detected and attributed to the partial resection of these fibers [29]. Lack of gastric compliance, severely restricted gastric capacity with an intact pylorus, and impaired gastric emptying have also been suggested to predispose patients to reflux during the first postoperative period [10, 30].

On the other hand, few authors believe that sleeve gastrectomy (SG) has a beneficial effect on reflux. Moon Han et al. [17] reported a decrease in the incidence of reflux symptoms and esophagitis after SG in 70% of cases, partly attributed to a decrease in intra-abdominal pressure when reduction in body weight was achieved. Santoro [18] suggested that SG may treat reflux by reducing both acid production and tension on the gastric wall below the cardia by Laplace’s law. Melissas et al. and Braghetto et al. have shown an accelerated gastric emptying at scintigraphy obtained 3 months postoperatively that may improve reflux symptoms [22, 31, 32], but this is debatable and it probably has to do with either methodology differences between trials or with the amount of antrum preserved. In this series, an overall tendency towards relief of heartburn and increase of regurgitation and vomiting is noted during the first year following LSG, and those findings especially characterize the tubular pattern. Heartburn relief could be attributed to decreased gastric acid production as a result of the removal of the mass of the gastric parietal cells. On the other hand, impairment of the antireflux mechanism of the cardia after sleeve gastrectomy may lead to increased regurgitation of content that is not acid. In other words, there might be—at least initially—increased gastroesophageal reflux of non-acid contents, which explains the increased regurgitation and vomiting without heartburn.

In our patients, the three distinct patterns of the gastric sleeve depicted at UGI after LSG reflect the technical limitations of the surgical technique. The tubular gastric sleeve is the desired pattern of this restrictive procedure. However, incomplete dissection of the posterior fundus, in an effort to avoid injury of the esophagogastric junction, or an excessive redundant fundus has led to preservation of a part of the gastric fundus, which is depicted as a superior pouch. On the other hand, misplacement of the bougie or misidentification of the pylorus has led to the use of the wrong landmarks for the beginning of the gastric division, which resulted in antral preservation, depicted as an inferior pouch.

These variations of the postoperative gastric sleeve anatomy seem to affect symptoms of reflux. It is not known if the size of the sleeve as a result of the size of the bougie used might have played a role on these findings, and certainly, no definite answer could be given unless someone specifically looks at the possible differences of using different sizes of bougie. However, all patients in this series had the same bougie size used—36F, which is in the range between 32F and 40F most commonly reported in the literature—but not all had the same symptoms postoperatively. The parameter that seemed to affect postoperative symptoms was the shape of the sleeve. The observation that both regurgitation and vomiting are significantly higher in the tubular pattern as compared to the superior and inferior pouch patterns during the first postoperative year needs further investigation. It is probably associated with the expected increased intragastric pressure after sleeve gastrectomy due to the loss of the fundus and body of the stomach and the impaired receptive relaxation mechanism mentioned above. It seems that the presence of a pouch may increase the ability of the stomach to distend and accommodate food, so that less gastric content is “available” for reflux. This could potentially lead to reduction in regurgitant volume in patients with pouches as compared to tubular gastric sleeves and, therefore, less severe regurgitation and vomiting. In fact, Nocca et al. [33] suggest preservation of the antrum in order that symptoms of reflux are decreased, since they noted a low rate (11.8%) of symptoms when the dissection of the great curvature began 10 cm from the pylorus.

Patients with suspicion of GERD were excluded from the study, based on the existing data in the literature that sleeve gastrectomy induces reflux symptoms, at least in the early postoperative period. As such, pH studies and esophageal manometry were not part of our routine evaluation of these patients. However, based on our described observations regarding this possible correlation of reflux symptoms with the sleeve shape, we recently have included for research purposes these studies in our evaluation protocol, in both the preoperative and postoperative setting. These studies would be helpful in order to fully elucidate the functional changes resulting from the alterations in the normal anatomy this procedure involves.

In conclusion, it seems that LSG affects the behavior of the esophagogastric junction in a way that is not clear and predictable, which may have to do with variations in the final shape of the sleeve as depicted by radiography. Further studies are needed in order that we could assess the functional results in relation to the anatomy of the gastric sleeve.

Conflict of Interest

The authors declare that they have no conflict of interest.

Copyright information

© Springer Science + Business Media, LLC 2010