Obesity Surgery

, Volume 24, Issue 8, pp 1386–1390

Midterm Outcomes of Revisional Surgery for Gastric Pouch and Gastrojejunal Anastomotic Enlargement in Patients with Weight Regain After Gastric Bypass for Morbid Obesity

  • Abdulrahman Hamdi
  • Christopher Julien
  • Phillip Brown
  • Ian Woods
  • Anas Hamdi
  • Gezzer Ortega
  • Terrence Fullum
  • Daniel Tran
Original Contributions

DOI: 10.1007/s11695-014-1216-z

Cite this article as:
Hamdi, A., Julien, C., Brown, P. et al. OBES SURG (2014) 24: 1386. doi:10.1007/s11695-014-1216-z


Reoperative surgery for the morbidly obese has become increasingly common due to postoperative weight regain. There are limited studies evaluating the effectiveness of revisional surgery. This study evaluates the weight loss outcomes of revisional surgery over a 2-year period at our University Hospital, USA. Of the 412 patients who underwent laparoscopic bariatric surgery between June 2009 and June 2011, we identified 25 patients who had Roux-en-Y gastric bypass (RYGB) originally, who underwent laparoscopic revisional surgery for weight regain. Preoperative and postoperative data were reviewed. Statistical analysis was performed using paired t test. This study includes 0 male and 25 female patients with an average age of 42 (range min to max: 28–58), mean original body mass index (BMI) of 54.6 kg/m2 (r = 37.3–80.7), average lowest BMI achieved of 32.2 (r = 20.1–50.9), and average BMI at the time of revision of 41.0 kg/m2 (r = 29.5–60.7, standard deviation (SD) = 8.5). All laparoscopic revisions consisted of resizing the gastric pouch by resection and recreating the gastrojejunostomy. Average hospital length of stay was 1.28 days (r = 1–4). Perioperative morbidity was 8 %; one patient developed a trocar site hernia which required repair, and another suffered postoperative bleeding requiring transfusion. There was no mortality. Postoperative BMI averages at 3, 6, 9, 12, and 24 months were 35.0 (SD = 7.15), 34.7 (SD = 4.26), 36.2 (SD = 7.63), 33.0 (SD = 6.58), and 44.2 (SD = 12.87), respectively. Statistically significant weight loss was achieved at 3 [t (10) = 6.74, p < 0.05], 6 [t (7) = 4.69, p < 0.05], 9 [t (9) = 2.94, p < 0.05], and 12 [t (6) = 3.78, p < 0.05] months. However, there was no statistically significant weight loss at 24 months postoperatively [t (4) = −0.16, p > 0.05]. Laparoscopic revisional bariatric surgery can be performed with significant weight loss up to 1 year postoperatively. However, additional studies are required to evaluate longer-term success.


Bariatric surgery Bariatric revisional surgery Laparoscopic 



Body mass index


Excess weight loss






Roux-en-Y gastric bypass




Diabetes mellitus


Gastroesophageal reflux disease


Bariatric operations have become the fastest growing surgical procedures in the USA as a result of the continued rise in obesity and advances in laparoscopic techniques, with Roux-en-Y gastric bypass (RYGB) surgery leading the charge [1]. It has been a very successful procedure compared to other types of bariatric operations. In fact, studies have been published regarding conversion of other bariatric procedures to RYGB for various complications and most notably inadequate weight loss [2,3]. In the RYGB, the proximal part of the stomach is transected in order to create a small gastric pouch, measuring from 10 to 30 mL. The gastric pouch is anastomosed to a Roux-en-Y proximal jejunal segment, bypassing the remaining stomach, duodenum, and a portion of the jejunum [2]. In the postoperative period, the patient is instructed to consume smaller portions. In addition to the restriction, bypassing the proximal small bowel will induce a certain degree of malabsorption. Unfortunately, some patients regained weight over time. Weight regain has been reported to be as great as 25–35 % of initial excess weight loss and typically occurs within 2–7 years after RYGB surgery. It is usually secondary to multiple factors, but a substantial body of literature endorsed the correlation with gastric pouch and gastrojejunal anastomosis dilation to weight regain [4-6]. This loss of restrictive property is postulated to be the primary cause of weight regain. One type of revisional surgery to restore this restrictive property entails resection of the pouch to reduce pouch size and creating a new, smaller gastrojejunostomy (stoma). Currently, there are several studies advocating this procedure [7,8]. Other options for revision include injecting a sclerosant (sodium morrhuate) into the stoma to scar it down, endoscopic plication of the gastric pouch and stoma (StomaphyX and ROSE procedures), conversion to an adjustable gastric band, conversion from a proximal (conventional) Roux-en-Y gastric bypass to a distal Roux-en-Y gastric bypass by increasing the malabsorptive component (bypassing more of the small bowel), and conversion to a duodenal switch operation [9-13]. There are a growing number of patients who have significantly regained the weight and will require a revisional procedure. Our objective was to evaluate the effectiveness of gastric pouch and stoma size reduction over a 2-year period.


A retrospective chart review and analysis of data were initiated. The study group consisted of 25 patients in a 2-year time frame from June 2009 to June 2011 with previous RYGB operation who underwent laparoscopic revisional bariatric surgery for weight regain and other symptoms thought to be associated with dilation of their pouch and/or stoma. Patients qualified for the revisional procedure if they meet the NIH criteria for weight loss surgery [14,15]. Patients were only considered if they were ≥3 years postoperatively and presented with a ≥10 point BMI increase from their nadir weight achieved after their original RYGB. Only those patients with a RYGB as their original operation were selected. Additional criteria for selection were demonstration of a dilated pouch and/or stoma (Gastrojejunostomy). We defined a dilated pouch as ≥5 cm. This is based on the difference in length measured from the incisor to the gastrojejunostomy and the incisor to the Z-line. A dilated gastrojejunostomy was defined as ≥2 cm. This measurement is based on an estimation with respect to the diameter of the endoscope which is 11 mm. Data collected consisted of patient demographics, medical history, bariatric procedural history, weight data before and after RYGB, weight data before and after revisions, and if any, prerevisional or postrevisional complications or morbidity were recorded and analyzed. Telephone follow-up was performed where needed. All patients received laparoscopic gastric pouch resection and creation of gastrojejunostomy. All patients underwent an upper endoscopy prior to the planned surgery. Analysis of data was carried out using the paired t test model.

Laparoscopic Revisional Surgery Techniques

Briefly, once all trocars are placed and pneumoperitoneum achieved, the gastric pouch and Roux limb are then freed from all associated adhesions. The Roux limb is generally divided at about 5 cm distal to the gastrojejunostomy using the laparoscopic linear cutter/stapler. The Roux limb mesentery supplying the proximal segment is then divided. The laparoscopic linear stapler is then used to resect the distal part of the gastric pouch. The pouch resection was generally performed 4–5 cm distal to the gastroesophageal junction to recreate the small gastric pouch. The specimen which includes the gastrojejunostomy is removed en bloc. The Roux limb is brought up to the gastric pouch, and an end-to-side anastomosis is performed using a linear stapler to create a 1- to 1.5-cm-diameter anastomosis. The otomy is closed with a running 2–0 Vicryl suture. No bougie is used. The size of the revised pouch is based on an ergonomic estimation by the operating surgeon based on his experience. The targeted revision pouch size was to be less than 5 cm in its greatest dimension. Generally, we aimed at achieving the smallest pouch size possible while still allowing an adequate gastrojejunostomy to be performed. We calibrate the size of the GJ using the millimeter markers on the linear stapler for a maximal allowance of a 15-mm opening. Furthermore, upon closing the otomy with intracorporeal suture, we estimated an additional reduction of 3–5 mm in the diameter of the anastomosis. The otomy closure is performed with an 18 F NG tube in place acting as a stent. This is only used as a reference guide to prevent accidental closure of the anastomosis from an inadvertent back wall stitch. The anastomosis is tested for leaks with methylene blue. The trocars are then removed and skin incisions are closed.


A total of 25 patients underwent revisional surgery. All original RYGBs were performed laproscopically except in two patients (8 %), who had an open procedure. The mean excess weight at the time of the initial RYGB was 152 % of ideal body weight (IBW). All patients had their original surgery performed at outside institutions. All patients in our study were female. Twelve (48 %) patients had reached the 2-year time interval at the time these data were collected. Table 1 shows the mean age at the time of revision of 42 (28–56) years. Mean weight was 115 (84–157) kg, and mean BMI was 41 (29.5–60.7). The lowest mean BMI and mean excess weight loss percent (EWL%) recorded after primary surgery were 32.2 and 69.2 %, respectively. The mean time between RYGB and achieving the lowest BMI was 22 months (4–120). The mean time between the original RYGB and the revision surgery was 7 years and 3 months. At the time of reoperation, weight regain had occurred such that the mean EWL percent was 39.8 %. Although the majority of patients were morbidly obese, there were six patients (24 %) who had a BMI of less than 35 at the time of revision.
Table 1

Demographics and anthropometric measurements of study population (n = 25)



At nadir after RYGB

At GJ revision

Age in years

35 (22–51)

37 (22–53)

42 (28–56)

Weight in kg

153 (92.5–226.7)

90.7 (49.8–156.4)

114.7 (83.9–156.9)

% Excess weight loss


69.3 (35–103.9)

39.8 (−11.9 to 72.7)


54.6 (37.3–80.7)

32.2 (20.1–50.9)

41.0 (29.5–60.7)

Data presented in numbers are mean values with ranges in parenthesis

BMI Body mass index, Kg kilograms, GJ gastrojejunostomy, RYGB Roux-en-Y gastric bypass

Endoscopic Findings

All measurements taken were based on edoscopic markings measured from the incisors. The mean pouch size measured at the longest dimension was 6.8 cm (±2.6, n = 24). The mean stoma diameter was 2.5 cm (±1, n = 24). Of the 24 patients, 5 patients (20.8 %) had evidence of gastritis, 3 patients (12.5 %) had benign gastric polyps, 1 patient (4 %) had a gastrogastric fistula between the pouch and remnant stomach, and 1 patient had a stoma stricture <0.5 cm which required endoscopic dilation.

The most common comorbidity was gastroesophageal reflux disease (GERD, 92 %). All patients who suffered from dumping also had concomitant GERD symptoms (48 %). Dumping symptoms were resolved or significantly improved in nearly all patients postrevision even up to the 2-year follow-up. GERD symptoms resolved or significantly improved in nearly half of those patients seen postoperatively at each of the follow-up intervals. The remainder of the patients reported no changes in their reflux symptoms postrevision. There were complications in two patients (8 %). One involved a trocar site hernia that required reoperation, and a second patient had a postoperative bleeding requiring transfusion of 1 unit of packed red blood cells. There were no mortalities. There were no intraoperative complications. All patients in our series received an upper GI swallow study with small bowel follow though using Gastrografin contrast on postoperative day 1. This was done to delineate patency and check for leaks. The average hospital stay after revisional surgery was 1.3 days (1–4).

Figure 1 shows the average BMI for patients from pre-RYGB to 2 years postrevision. Five patients were lost to follow-up after the revision. The highest weight loss after revision was noted in seven patients at 1-year follow-up, 14.9 kg (p < 0.001), BMI 33.0. The second highest weight loss was noted at the 6-month follow-up which had an average BMI of 34.7 (p < 0.001). Significant weight loss was also noted at 3 and 9 months postrevision with average BMIs of 35 (p < 0.001) and 36.2 (p < 0.008), respectively. There was no statistically significant weight loss at 2 years postrevisions when compared to prerevision BMI of 44.2 (p = 0.4).
Fig. 1

Bar chart showing average BMIs along a time interval. Numbers above bars indicte mean values. Error bars presented in standard error. BMI: Body mass index, RYGB: Roux-en-Y gastric bypass

Figure 2 shows the EWL percent, calculated relative to the pre-RYGB excess weight, at different intervals from post-RYGB nadir to 2 years postrevision. The highest EWL postrevision was noted in the 1-year group at 64.1 % (p < 0.001).
Fig. 2

Change in %EWL over time from original RYGB nadir to 2 years post revision of Gastro-jejunostomy . Numbers on graft represent mean values. EWL: Excess weight loss

The EWL at 1-year post revision surgery (64.1 %), using the weight at RYBG, was not statistically different from the EWL achieved at nadir post-RYGB (74.5 %, note that here, the average EWL at nadir is different from the average indicated on the graft (69.3) as it represents that of the same seven patients indicated at the 1-year follow-up.) (p = 0.12, n = 7).

When calculating the EWL from the revision only, again, the highest percentage is noted in the 1-year group at 41.5 %. EWL at 3, 6, and 9 months and 2 years is 28, 29.7, 27.6, and 16.7 %, respectively.


Weight regain after bypass surgery has paralleled the procedures’ growing popularity [16]. Although the Roux en-Y gastric bypass is emerging as the most commonly performed procedure for weight loss, it has some limitations. Up to 20 % of the patients with RYGB will experience significant long-term weight recidivism [3,17]. Although scant, the data regarding revisional surgery after RYGB are accumulating. Surgical options described ranges from conversion of RYGB to duodenal switch, distal bypass, adjustable gastric banding, and various endoscopic procedures, each with its unique set of advantages and disadvantages. One of the earliest RYGB revision methods described was the revision of the stoma and/or the gastric pouch, a redo gastrojejunostomy [18].

Our study has shown that laparoscopic gastric pouch and gatrojejunostomy revision can produce EWL (64 %) comparable to that at the patients’ nadir weight after primary RYGB. Although still less than nadir EWL (74 %) after Roux-en-Y bypass, there was no statistically significant difference between the two groups (p = 0.12, n = 7). Patients lost an average of 8 BMI points over 1 year. According to our data, this 1-year mark represents the excess weight lost nadir postrevision.

The patient’s progress at the 2-year interval is notably different. Not only was there a significant increase in average BMI from 1 to 2 years, 33 to 42, respectively, but there was also no statistically significant weight loss noted at the 2-year interval compared to the prerevisional BMI, 44.2 and 43.9, respectively (p = 0.43, n = 5).

An early report by Schwartz R et al. in 1988 of 42 patients on revising the gastric pouch and gastrojejunostomy showed promising weight loss results but with a 50 % complication rate [19] Although eight patients did not match the current indications for surgery, the outcomes were correlated. In their series, patients were able to achieve a peak EWL that exceeded that of the original operation. In addition, EWL at the 2-year follow-up interval was comparable to EWL at nadir post-RYGB. Albeit, overall, their group lost less EW than our study at every interval; their initial EW was substantially lower than that of our patient population (129 % vs 152 % of IBW, respectively). This is consistent with the current trend in the obesity epidemic as the average BMI continues to increase. It is also important to note the drastic reduction in complication rates probably owed to improved surgical technology, instrumentation, surgical techniques, and improved postoperative care that have occurred over the years.

Other studies have shown more moderate results with smaller cohorts. Mason E et al. reported a 10-kg median weight loss 1 year postrevision of 15 patients [20]. Muller M et al. described a 4.6 point decrease in median BMI of five patients over a 1-year follow-up [21].

Although there is a positive correlation between a larger gastric pouch and stoma size with weight regain following bypass surgery, the extent of this is still being investigated [22]. We believe that revision of pouch and gastrojejunostomy can be successfully and safely performed in the correct settings. Proper selection of patients is the key. Those with dilated gastric pouch and/or stoma with associated symptoms such as GERD and dumping might benefit the most from this type of revision. Laparoscopic adjustable gastric banding (LAGB) has shown some promising early results in one study [23]. We have shown that transoral endoscopic gastric plication also resulted in early modest weight loss but unfortunately, at 1 year, all patients had regained the weight [5]. Further studies will be required to evaluate the long-term efficacy of the laparoscopic gastric pouch revision and to compare the results to other methods of revision including conversion from RYGB to a duodenal switch or a distal gastric bypass.

Although there is a significant lack of data in the literature addressing outcomes of revisional surgery in general, our paper addressed the midterm outcomes specific to revision of failed RYGB. Other strengths include a standardized revision technique from a single institution minimizing operative variability. Limitations of this study include a retrospective study design, small patient series, and significant loss to follow-up, especially at the 2-year interval. Patients lost to long-term follow-up is a major limiting factor in the power of many revisional studies including ours. Aggressive follow-up protocols with improved data collection may help provide a more concise algorithm for the indication for each of these revisional procedures.


Laparoscopic gastric pouch and gastrojejunostomy revision can be performed safely with significant weight loss up to 1 year postoperatively. However, additional studies are required to evaluate longer-term success.

Conflict of Interest

The authors have no conflicts of interests or sources of financial support to declare.

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Abdulrahman Hamdi
    • 1
  • Christopher Julien
    • 1
  • Phillip Brown
    • 1
  • Ian Woods
    • 1
  • Anas Hamdi
    • 1
  • Gezzer Ortega
    • 1
  • Terrence Fullum
    • 1
  • Daniel Tran
    • 1
  1. 1.Howard University HospitalWashingtonUSA

Personalised recommendations