Abstract
Background:
Failed restrictive procedures are usually managed with conversion to another bariatric procedure. Our aim was to evaluate one -anastomosis gastric bypass (OAGB) as a revisional option for failed restrictive procedures. In addition, we compare the outcomes of OAGB versus Roux-en-Y gastric bypass as a revisional bariatric procedures.
Material and Methods:
The current series is a prospective study, from May 2009 to December 2016. A total of 348 patients with failed restrictive bariatric operations underwent laparoscopic revisional gastric bypass. Revisional OAGB was performed in 243 patients and revisional Roux-en-Y gastric bypass in 105 patients. The demographic data and outcomes were studied by our multidisciplinary team.
Result:
By the end of the study, the mean age was 39.3 ± 10.3 years with body mass index of 37.5 ± 9.2 kg/m2. At 2-year follow-up, the overall intractable reflux (Symptom-Severity score questionnaire >4) was significantly higher after revisional OAGB (21.4%). The reflux with scoring ≥4 was significantly higher in the vertical band gastroplasty than laparoscopic adjustable gastric band and laparoscopic gastric sleeve (25.2%, 16.9%, and 14.3%, respectively).
Conclusion:
Although laparoscopic revisional OAGB is a feasible and safe option after failed restrictive procedures, it has a higher chance of reflux in long-term follow-up.
Introduction
Laparoscopic gastric banding and vertical banded gastroplasty (VBG) were popular worldwide during recent decades for the treatment of obesity. Both procedures were effective in weight loss and improvement of comorbidities with acceptable operative risk.1,2 Mechanical complications, weight regain, and poor quality of life in the eating manner were the main reasons for revising these procedures. The weight regain after VBG, ranged from 49.7% to 56%.3–5
The growth of the laparoscopic vertical sleeve gastrectomy (LVSG) in popularity is attributed to its perceived technical simplicity, excellent weight loss outcomes, and resolution of comorbidities.6–9 Although the excess weight loss (EWL) after 5 years is >50%, the long-term data reported a significant weight regain after LVSG.10–13
The increase in the number of unsatisfactory primary restrictive procedures will be a significant demand in the future.14,15 Therefore, it is necessary to have a safe and effective option that manages the restrictive eating pattern. Although the Roux-en-Y Gastric Bypass (RYGB) is the most common revisional procedure for failed restrictive surgeries with a satisfactory long-term weight loss, it is associated with a high rate of complications and metabolic side effects. 16 One anastomosis gastric bypass (OAGB) as a primary procedure has reported excellent results with low rates of postoperative complications and short operative duration.17–19
In this study, we aim to evaluate the OAGB surgery as a revisional procedure for the unsatisfactory or complicated primary restrictive bariatric operations. By extension, we compare the outcomes of OAGB versus RYGB for the revision of the different types of restrictive bariatric procedures.
Patients and Methods
This study was conducted in the Obesity Surgery Unit, at our University Hospital, a University Public Hospital reference center for metabolic surgery. The Clinical and Demographic data were prospectively collected from the preoperative evaluations. The presurgical and postsurgical management were performed by the bariatric surgery team. The inclusion criteria were essentially the recommendations of the National Institutes of Health Consensus Development Panel, made in 1991. 20
From May 2009 to December 2016, a total of 543 patients with failed restrictive bariatric operations underwent laparoscopic revisional bypass surgery in our center. The Assiut University Ethical Committee (AUEC) approved the study and written informed consent was obtained from all patients. Patients with large incisional hernias, severe debilitating nutritional deficiency, drug or alcohol addiction, history of personality disorder, advanced malignancy or not completing the 2-year follow-up were excluded from our study.
In the current series, 348 patients were included, all of them were available for 2-year follow-up. Failure of a prior restrictive operation included inadequate weight loss “<50% of excess weight lost (EWL), at 18 months post-operation,” weight regain “an increase of more than 25% EWL from nadir,” and reflux esophagitis or intolerance to the restriction (intractable side-effects or psychological) (Table 1). The preoperative diagnosis of reflux was diagnosed by Symptom-Severity (SS) score questionnaire, which was validated in 2000 by Allen et al. A severity score ≥4 is considered positive for gastroesophageal reflux disease (GERD). In our study, 203 patients presented following failed VBG, 95 following failed laparoscopic adjustable gastric band (LAGB), and 50 after failed LVSG.
Indications of Laparoscopic Revisional Surgeries for Restrictive Procedures
LAGB, laparoscopic adjustable gastric band; LVSG, laparoscopic vertical sleeve gastrectomy; VBG, vertical banded gastroplasty.
All patients were evaluated for surgical treatment by a multidisciplinary and integrated medical team. Patients with symptomatic GERD and evidence of Barret's esophagus on endoscopy or by histological examination were selected for RYGB. In addition, patients with intraoperative failure of creating a long gastric pouch were also selected for RYGB. All the other patients were planned for OAGB. R-OAGB was performed in 243 patients (group 1) and Revisional-RYGB in 105 patients (group 2), and the demographic data, surgical parameters, and outcomes were studied by our multidisciplinary team.
Informed consent was signed by all patients after providing them a full and clear explanation of benefits, risks, and long-term consequences of the conversion to bypass surgery.
All patients underwent revisional gastric bypass laparoscopically by five-port technique. For the prior VBG patients, as a first step, we began by separating the stomach wall from the liver and overlying omentum in an attempt to identify the site of the mesh to be removed, then the size of the created gastric pouch was decided. Based on the location of the mesh in primary VBG cases, we selected patients for OAGB versus RYGB. In 143 patients, the length of the pouch was long and OAGB was performed after bypassing 150–250 cm of small bowel from ligament of Treitz. However, in 60 patients, the length of the pouch was short, so we proceeded with RYGB where the biliopancreatic limb was 50–70 cm and the alimentary limb was 120–150 cm.
For the 95 patients who underwent revisional bypass after LAGB, the gastric band was first divided and removed followed by complete unwrapping of the stomach. Thereafter, OAGB was performed in 65 patients and RYGB in 30 patients. Regarding the prior LVSG patients, R-RYGB was performed in 15 patients and R-OAGB in 35 patients. In our study, Biliopancreatic limb (BPL) length was assigned independently to body mass index (BMI), age, or comorbidities. For patients who underwent R-OAGB, 250-cm BPL was used in patients with super obesity and when greater weight loss is needed.
The decision of patient transfer to the intensive care unit was taken by the intensivist while patients were being monitored in the recovery room. The early postoperative ambulation was strongly encouraged 2–4 h postoperatively. Water was allowed in the first 24 h, clear fluids for 10 days postoperatively followed by semisolid food. By the fourth postoperative week, solid food intake was feasible. Follow-up appointments were scheduled at 2 weeks postoperatively, then monthly for the first year and the follow-up was every 3 months. In our study we presented the postoperative surgical complications as minor or major complications, which were stratified according to the Clavien–Dindo scale.
Endpoints
The primary endpoint is the assessment of reflux symptoms. All patients were interviewed at 6, 12, 18, and 24 months postoperatively using SS score questionnaire, 21 which was validated by Allen et al. 22 A severity score ≥4 is considered positive for GERD. Weight loss was denoted as a percentage of EWL (%EWL). The copyrights registration for the questionnaire was performed and submitted to the ethics committee of the institute.
Statistical analysis
The paired t-test has been used for comparing baseline data of both groups and for the continuous data. Fisher's exact test was used to compare categorical measures between the two groups. Data were presented by mean ± standard deviation and percentages when appropriate. All statistical tests were carried out by using the PRISM version for Windows. Statistical significance was set at p < 0.05. Significance values with a confidence limit <1% are reported as p < 0.01. For all analyses with a confidence limit >1%, exact values are quoted.
Results
Between May 2009 and December 2016, 348 patients with failed restrictive bariatric procedures underwent laparoscopic revisional bypass surgery (339 [77.4%] women and 99 [22.6%] men); the mean age was 39.3 ± 10.3 years with an overall preoperative BMI of 37.5 ± 9.2 kg/m2. The baseline demographic data are given in Table 2. All the revisions to OAGB or RYGB were completed laparoscopically with an overall mean operative duration of 71.5 ± 50.2 min and mean hospital stay of 2.1 ± 0.8 days, with no cases converted to open surgery (laparotomy). The revisions were performed after a mean of 28.6 months.
Preoperative Demographics Presented by Mean ± Standard Deviation
Each p-value was calculated by paired t-test.
R-OAGB, Revisional-one anastomosis gastric bypass; R-RYGB, Revisional-Roux-en Y Gastric Bypass.
By the end of the study, R-OAGB was performed in 243 patients (group 1) and R-RYGB in 105 patients (group 2). The patients who underwent R-OAGB showed a significantly higher weight loss, EWL 71.8% ± 5.9% versus 58.3% ± 6.6% for the R-RYGB patients (p = 0.032). In the current series, the overall mortality rate was 0.2% (1 case); in the patient who underwent R-OAGB, the death was owing to pulmonary embolism 5 days postoperatively. There was a nonsignificant difference regarding the minor or major complications between the two groups (p = 0.279 and 0.946, respectively). The incidence of leakage was higher in the R-RYGB group (2.8% vs. 0.9%, p = 0.031; Table 3).
Operative Parameters and Complications Between Revisional-One Anastomosis Gastric Bypass and Revisional-Roux-en Y Gastric Bypass
Data presented by mean ± standard deviation.
Significant difference (p < 0.005).
EWL, excess weight loss; Hb, hemoglobin.
There was a significant difference in the intractable reflux rates between both study groups; 21.4% (52 patients) for the R-OAGB group compared with 2.9% (3 patients) for the R-RYGB group (p < 0.001). The onset of reflux symptoms in patients who underwent R-OAGB is given in Figure 1. Regarding the SS score questionnaire, there was a significant difference for the score <4 (no reflux), 78.6% for R-OAGB versus 97.1% for R-RYGB as given in Table 4. Regarding the 52 patients complaining of intractable reflux after R-OAGB, 18 of them were managed medically by proton pump inhibitor, Braun anastomosis (jejunojejunostomy at 70–80 cm distal to the gastrojejunostomy) was performed in 4 patients and the other 30 patients underwent conversion to RYGB, 3 of them showed no improvement of the reflux symptoms.

Onset of reflux symptoms in patients who underwent R-OAGB, the vertical axis for number of patients and the horizontal axis for the time of first complaint of reflux. R-OAGB, Revisional-one anastomosis gastric bypass.
Comparison of the Gastroesophageal Reflux Disease Questionnaire Results Between Revisional-One Anastomosis Gastric Bypass Versus Revisional-Roux-en Y Gastric Bypass
A severity score ≥4 is considered positive for GERD.
Significant difference (p < 0.005).
GERD, gastroesophageal reflux disease.
Subgroup analysis was performed to assess the intractable reflux after R-OAGB as a revisional procedure for the different types of restrictive procedure. Among the 243 patients who underwent R-OAGB, 143 had VBG, 65 LAGB, and 35 LVSG. The onset of reflux symptoms in patients who underwent R-OAGB, based on the previous procedure is given in Figure 1. In the current series and with using the SS score, reflux with scoring ≥4 was significantly higher in the VBG group compared with LAGB and LVSG (25.2%, 16.9% and 14.3%, respectively; Table 5).
Gastroesophageal Reflux Disease questionnaire Uses a Grading of Symptoms of Heartburn and Regurgitation
A severity score ≥4 is considered positive for GERD. Data are presented as n (%).
Significant difference (p < 0.005).
Discussion
Since the VBG was endorsed by the NIH in 1991, it acted as a popular bariatric procedure during the past two decades because of its technical safety and simplicity. In the same manner, LAGB, which was introduced in 1994, produced similar results in terms of weight loss with low morbidity and mortality.20,23 In addition, laparoscopic procedures has more advantages in terms of cosmesis and comfort. 24
When you have a look at the long follow-up of the VBG in the literature, we found that the restrictive procedure was associated with unsatisfactory long-term follow-up. The Ortega et al. study reported a 50% failure rate after only 2 years of follow-up. 25 The 12-year follow-up of VBG by Van Gemert et al. reported 56% failure rate. 3 Regarding LAGB, Suter et al. study showed that each year delay in its reoperation increases the risk of major complication by 4%. 15 Thus, effective and safe rescue revisional procedures are required. The results of our study, which suggests that gastric bypass is a good revisional procedure, is comparable with what is published in the literature.25–28
In this study, no differences were found in both minor or major postoperative complications between the two groups with an overall mortality rate of 0.2%, which is comparable with what is published in the literature on revisional bariatric surgery. These complication rates are higher than primary bariatric procedures emphasizing the complexity of secondary bariatric procedures.29,30
In our study, not all patients had the additional weight loss following their revision surgery, but the mean of EWL was 71.8% ± 5.9% for the R-OAGB and 58.3% ± 6.6% for the R-RYGB patients, they are comparable with the previous reports on secondary procedures.31–33 However 24.1% of patients in the study underwent a revisional surgery because of reflux esophagitis or intolerance to the restriction (intractable side-effects or psychological).
In the current series, the wide range of the operative time may be attributed to fact that all patients underwent a revisional surgery with wide variations in the intraoperative findings. In our study, the weight loss was significantly higher after R-OAGB (EWL was 71.8% ± 5.9% for R-OAGB vs. 58.3% ± 6.6% for the R-RYGB, p = 0.032). The better weight loss after R-OAGB mainly attributed to the longer bypassed part of the jejunum, which routinely bypass 150–300 cm as a biliopancreatic limb.19,34–36 However, the biliopancreatic limb length in R-RYGB varied from 50 to 100 cm.37,38 Although the longer bypassed limb of the intestine was associated with higher weight loss,39,40 it is associated with more nutrition deficiency owing to the higher malabsorptive component, which is evident by the lower hemoglobin (Hb) level after R-OAGB.36,41 This could explain the higher incidence of iron deficiency and the significantly lower mean Hb level after R-OAGB in our study (8.2 ± 0.5 g/dL vs. 12.8 ± 3.2 g/dL, p = 0.030).
The gastrojejunal anastomosis in OAGB is technically an end-to-side, which in contrast to RYGB, allows bile and intestinal juice to freely reflux into the gastric pouch. This may explain the higher incidence of intractable reflux and upper gastrointestinal discomfort after R-OAGB in our study (21.4% for the R-OAGB group vs. 2.9% for the R-RYGB group). Although Manson's loop gastric bypass reported 70% bilious vomiting and gastritis after surgery, 42 the intractable reflux rate after OAGB at the Osama et al. study was 1.2% (18 patients). 19 This reflux rate is similar to most of the published studies of primary OAGB.43–46 In addition, the studies of Noun et al. and Tolone et al. reported no cases of reflux after OAGB. However, those two studies had small number of patients and were therefore subjected to type 2 error.47,48
The literature data reported that the restrictive procedures worsen the upper gastrointestinal symptoms and probably increase the incidence of the gastrointestinal reflux49–51 or pseudo-achalasia symptoms.52–54 The mechanism is unknown, but some authors attributed it to the fact that high-outflow resistance with chronic high pressure after the LAGB, VBG, or LVSG can lead to progressive weakening of the lower esophageal sphincter and the esophageal musculature. 55 These may explain the high intractable reflux and the upper gastrointestinal discomfort in our study.
Based on the previous theory, the high-outflow resistance with chronic high pressure in restrictive procedures is the main reason for reflux after revisional OAGB.52–54 As an example, LVSG is associated with the least outflow resistance (pylorus) with no adhesions55,56 followed by LAGB with limited adhesions located mainly just distal to the esophagogastric junction. On the contrary, VBG is associated with the highest outflow resistance owing to the intense adhesions and inflammatory reaction surrounding the mesh.26,56 Although further studies are required, this may explain the subgroup findings in our study; reflux scores were significantly higher in the VBG group compared with the LAGB and LVSG groups (25.2%, 16.9%, and 14.3%, respectively).
The presence of a distended upper part and narrowing middle part of the stomach for a long time after the restrictive procedures may be lead to irreversible gastric motility disorder; the mechanism behind this theory is not clear, but it was reported in multiple series.57–60 In the same manner, Baumann et al. 61 reported absent or abnormal peristalsis with delayed emptying after LVSG, which leads to proximal gastric distension and precipitating lower esophageal relaxations. This may explain the high reflux rate in our series and explain the nonimprovement of reflux symptoms in three patients converted from R-OAGB to RYGB.
The main limitations in our study were the dropout of patients at follow-up assessment and a predictable loss of some data. Another limitation is the short follow-up period. So, we are planning for upper endoscopy, esophageal manometry, and PH-metry studies with longer follow-up periods. In the same manner, more randomized comparative studies are mandatory to validate these results. The main message from the current series is to encourage and ask for more studies about OAGB as a revisional procedure.
Conclusion
The R-OAGB is simple, safe, effective, and easy to learn with a better weight loss than R-RYGB, but it has a higher chance of reflux and anemia in long-term follow-up. R-OAGB has acceptable reflux rates after LAGB and LVSG, but not recommended after LVBG.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
Funding Information
The study didn't receive any funds.
