Laparoscopic Roux-en-Y gastric bypass (LRYGB) has been proven to be an effective treatment of morbid obesity. It leads to substantial excess weight loss and reduction or even remission of metabolic comorbidities [1,2,3]. Unfortunately, insufficient weight loss or weight regain after LRYGB have been described as well [4]. Insufficient weight loss and weight regain may have multifactorial causes. It can either be patient related (e.g. dietary non-compliance, physical inactivity or hormonal/metabolic factors) or surgery related (e.g. enlarged pouch due to construction of a large pouch at the primary LRYGB, gastrogastric fistulas or dilation of the gastrojejunostomy) [4, 5].

Several studies have shown the importance of the size of the pouch in the primary surgery on weight loss [6,7,8]. In case of an anatomical cause of insufficient weight loss or weight regain, surgical treatment is challenging and controversial. Several techniques have been proposed for revisional bariatric surgery (BS) if insufficient weight loss or weight regain is caused by an enlarged pouch [9]. Laparoscopic resizing of the gastric pouch and gastrojejunostomy show good results on percentage excess weight loss (%EWL) with low reoperation rate and no mortalities [10, 11]. Two techniques for resizing of the gastric pouch and gastrojejunostomy are described in literature: sleeve resection of the gastrojejunostomy and gastric pouch (SGP) [11, 12], and resection of the gastrojejunostomy with creation of a smaller pouch and a new anastomosis (RGJ) [13, 14]. So far only short-term results in small groups have been described.

The aim of this study is to determine whether a sleeve resection of the gastrojejunostomy and gastric pouch or a revision of the gastrojejunostomy is the superior technique for additional weight loss after LRYGB.

Materials and methods

Patient selection

All patients who underwent revisional BS for insufficient weight loss or weight regain after LRYGB from April 2014 to June 2018 in two expert centers for BS were included in this observational cohort study. Patients after primary LRYGB as well as patients who underwent previous BS before LRYGB were included. Previous BS was either laparoscopic adjustable gastric banding (LAGB) or laparoscopic sleeve gastrectomy (LSG). The SGP technique was performed in one of the centers, and the RGJ technique was performed in the other center. The performed revisional technique was based on the preference of the bariatric surgeons. Both techniques are described below.

Before qualifying for revisional BS, all patients with insufficient weight loss or weight regain followed an obligatory and standardized trajectory led by a team of psychologists, dietitians and psychotherapists to optimize the patients motivation and compliance to an adjusted life style. Patients were only eligible for revisional BS after failing this standardized trajectory in terms of additional weight loss combined with no sensation of restriction and diagnostics tests showing a large pouch. Diagnostic tests to evaluate a large pouch were either barium swallow test (BST) in combination with gastroscopy or a three-dimensional gastric computed tomography (3D-GCT) scan. A pouch was defined as dilated if the pouch was > 5 cm on the BST in combination with gastroscopy, based on the difference in length measured from the gastrojejunostomy to the Z-line, or if the pouch volume was > 50 ml on the 3D-GCT. Examples of pouch enlargement signs on these tests are depicted in Fig. 1. Patients were excluded from this study if the insufficient weight loss or weight regain was caused by a gastrogastric fistula or if revision of the gastrojejunostomy was due to a marginal ulcer.

Fig. 1
figure 1

Examples of an enlarged pouch evaluated through barium swallow test (a), computed tomography (CT) scan (b) and gastroscopy (c)

Full size image

Data collection

Baseline characteristics (i.e. age, sex, presence of metabolic comorbidities) and surgical characteristics (i.e. type of primary operation, duration of surgery and early surgical complications) were collected. The initial weight and body mass index (BMI) before the primary BS, the initial weight response to the primary BS, and the weight before revisional BS were recorded. All patients signed a consent form before surgery, in which they agreed that their data could be used for retrospective studies. Therefore, an approval of the international review board was not needed.

During follow-up, outcomes were collected at 3, 6, 12, and 24 months. Additionally, patients with comorbidities were seen by the internal medicine specialist, who evaluated whether metabolic comorbidities persisted, improved or resolved. This evaluation was based on general use of medication. For type 2 diabetes mellitus specifically HbA1c levels, for hypertension specifically blood pressure and screening for concomitant organ damage, for hypercholesterolemia specifically cholesterol levels and lastly for obstructive sleep apnea an evaluation of the pulmonologist was included. The amount of weight loss was described as the percentage of total weight loss (%TWL) and was calculated as ((operative weight − follow-up weight)/operative weight) × 100%. BMI was calculated as weight (kg)/height (m)2.

Weight data were analyzed for patients undergoing primary LRYGB and undergoing LRYGB as conversion from LAGB or LSG separately. In addition, all weight results were presented for the entire group.

Surgical techniques

Laparoscopic sleeve resection of the gastrojejunostomy and gastric pouch (SGP)

Laparoscopic sleeve resections of the gastrojejunostomy, gastric pouch and blind afferent limb were performed using a 60 mm endoscopic stapler (Fig. 2). A 34-gauge gastric tube was used to calibrate the size of the revised gastric pouch that was transected longitudinally 6 cm from the diaphragm. The resected tissue was removed using an endoscopic bag.

Fig. 2
figure 2

Laparoscopic sleeve resection of the gastrojejunostomy. a Roux-en-Y gastric bypass with an enlarged gastric pouch. b Laparoscopic sleeve resection of the gastrojejunostomy, gastric pouch and blind afferent limb

Full size image

Laparoscopic revision of the gastrojejunostomy (RGJ)

In this technique, the gastric pouch was transected above the level of the anastomosis approximately 6 cm from the diaphragm and reduced by a 60-mm endoscopic stapler, using a 34-gauge gastric tube for calibration (Fig. 3). A new 30-mm linear gastrojejunostomy was created and the remaining defect was laparoscopically sutured. Subsequently, the jejunum was transected below the anastomosis. The resected tissue was then removed using an endoscopic bag.

Fig. 3
figure 3

Laparoscopic revision of the gastrojejunostomy. a Roux-en-Y gastric bypass with an enlarged gastric pouch. b Transection and reduction of the gastric pouch. c Creation of a new gastrojejunostomy. d Transection of the jejunum and removal of the old anastomosis

Full size image

All operations were performed laparoscopically with the use of one 12-mm vision port for the camera, two 12-mm working ports, one 5-mm working port and a liver retractor. The average use of surgical disposables was calculated and operation time were compared for both techniques.

Statistical analysis

Statistical analysis was performed with IBM SPSS Statistics, version 24 (SPSS, Chicago, IL). Continuous data are presented as median and interquartile range (IQR) and were compared by the Mann–Whitney U-test according to normality. The Chi-square test was used for categorical data analysis. A P value < 0.05 was considered significant. Missing data were addressed with a pair wise deletion in follow-up analysis.

Results

A total of 37 patients who underwent SGP and 21 patients who underwent RGJ as revisional BS were included in our study. The baseline characteristics are presented in Table 1.

Table 1 Baseline characteristics
Full size table

Weight loss after resizing the pouch

Weight data are presented in Table 2. Median BMI at primary BS was significantly different but no differences were found in BMI at revisional BS. There were no statistical differences in %TWL between the two techniques during follow-up, as presented in Fig. 3. %TWL was determined based on weight at primary BS (Fig. 4a), at LRYGB (Fig. 4b) and at revisional BS (Fig. 4c). Based on weight at revisional BS, the 24-month %TWL was 12.3% [5.8; 14.5] in the SGP cohort versus 10.8% [3.4; 22.1] in the RGJ cohort (P = 0.604). The total %TWL based on weight at primary BS was %TWL [IQR] was 22.0% [16.4; 30.3] in the SGP cohort and 22.2% [11.7; 32.1] in the RGJ cohort (P = 0.885).

Table 2 Prerevisional and postrevisional weight characteristics of patients who underwent revisional bariatric surgery after primary LRYGB as well as the entire study population
Full size table
Fig. 4
figure 4

Percentage total weight loss during follow-up. a %TWL based on weight at primary bariatric surgery. b %TWL based on weight at LRYGB. c %TWL based on weight at revisional bariatric surgery. *Significantly different (P < 0.005). SGP sleeve resection of the gastrojejunostomy and gastric pouch, RGJ revision of the gastrojejunostomy, rBS revisional bariatric surgery, %TWL percentage total weight loss, LRYGB laparoscopic Roux-en-Y gastric bypass

Full size image

Figure 5 shows the trend of overall %TWL based on weight at primary BS for patients who had undergone LYRGB as primary BS (Fig. 5a) and patients who had undergone LRYGB as a conversion from LAGB or SG to LRYGB (Fig. 5b) respectively.

Fig. 5
figure 5

Percentage total weight loss during follow-up. a %TWL based on weight at primary bariatric surgery in patients who underwent LRYGB as primary bariatric procedure. b %TWL based on weight at primary bariatric surgery in patients who underwent LRYGB as a revisional bariatric procedure after LAGB or LSG. *Significantly different (P < 0.005). SGP sleeve resection of the gastrojejunostomy and gastric pouch, RGJ revision of the gastrojejunostomy, rBS revisional bariatric surgery, %TWL percentage total weight loss, LRYGB laparoscopic Roux-en-Y gastric bypass, LAGB laparoscopic adjustable gastric banding, LSG laparoscopic sleeve gastrectomy

Full size image

Surgical factors, early postoperative complications and surgical costs

Operation time was significantly higher in the RGJ procedure (57.0 min SGP vs 74.0 min RGJ, P = 0.003). At the RGJ group, one patient underwent both revisional surgery, laparoscopic cholecystectomy and closure of the mesenteric defects within one procedure. After exclusion of this patient, operation time was still significantly higher in the RGJ cohort (73.5 min, P = 0.006).

Three early postoperative complications occurred in the SGP group (8.1%) as compared to two complications in the RGJ group (9.5%), as demonstrated in Table 3. According to the Clavien–Dindo classification of surgical complications [15], in the SGP group two grade I and one grade IIIb complications occurred as compared to one grade I complication and one grade IIIB complication in the RGJ group. In both study groups, one patient had an anastomotic leakage (2.7% in the SGP group and 4.8% in the RGJ group). In the SGP group, a jejunal tube was placed endoscopically for enteral feeding to treat the anastomotic leakage. In the RGJ group, a relaparoscopy was performed to close the defect of the anastomotic leakage. Both patients recovered without any negative residual effects. The average surgical costs for the SGP technique were approximately €337 lower than for the SGJ technique (Table 4).

Table 3 Early postoperative complications
Full size table
Table 4 Surgical disposables
Full size table

Obesity-related comorbidities

In both groups, all patients achieved either improvement or remission of DM2 after revisional BS (Table 5). Hypertension improved or even resolved in 60% of the patients in the SGP group as compared to 66.7% in the RGJ groups. Two patients (50%) achieved remission of hypercholesterolemia in the SGP groups as compared to one patient (50%) in the RGJ group. There was no remission or improvement of OSAS achieved in both study groups.

Table 5 Prerevisional presence and postrevisional improvement or remission of comorbidities
Full size table

Discussion

Resizing a large gastric pouch after LRYGB leads to additional weight loss and has a positive effect on obesity-related comorbidities. The two techniques, SGP and RGJ, were equally effective in terms of weight loss. However, the SGP technique did result in less usage of disposables and a shorter operation time compared to the RGJ technique.

Based on weight prior to resizing the gastric pouch the median %TWL at 24 months after revisional BS was 12.3% [5.8;14.5] in the SGP cohort versus 10.8% [3.4;22.1] in the RGJ cohort (P = 0.604). When we compare the %TWL based on weight prior to the first bariatric procedure, the %TWL is 22.0% in the SGP cohort and 22.2% in the RGJ cohort (P = 0.885), which is comparable to previous studies [16,17,18,19]. Considering these results it should be noted that the additional effect on %TWL of revisional surgery was relatively small. This study, however, shows that the additional effect may be well worth the effort as all patients now achieved %TWL of more than 20% and additional resolution of comorbidities.

Previous studies have shown that the clinical effect on comorbidities in revisional BS is similar to primary BS [20, 21]. In this study, there seems to be an improvement of comorbidities in both groups. As the sample size was small, this study could not demonstrate a statistically significant reduction or improvement of comorbidities between the two techniques. Nevertheless, it can be concluded from this study that revision of a large pouch can exert a positive effect on obesity-related comorbidities. Therefore the continuous presence of obesity-related comorbidities should be considered as an indication for revisional BS.

Although this study did not intend to perform a full cost effectiveness analysis, we did find lower surgical costs of the SGP technique as compared to the RGJ technique. This difference was a result of a shorter operating time and less use of disposables in the SGP technique.

The effect of pouch size on the achieved weight loss after LRYGB remains controversial. Even though some studies have shown that a small pouch size results in higher achieved weight loss [6,7,8], others could not demonstrate a correlation [22, 23]. In this study, no calibration for pouch size was used at the primary LRYGB. However, revisional BS for a large pouch can lead to additional weight loss. Thus we might suggest the use of a calibration tube for LRYGB to prevent insufficient weight loss or weight regain due to a large pouch.

In this study, there was no consistent diagnostic technique used for pouch volume measurement, as no standardization for pouch volume measurement is defined in literature yet. The barium swallow test (BST) with upper gastrointestinal series has been used to measure pouch volume after LRYGB [6, 14]. However, it is challenging to calculate a three-dimensional pouch volume from two-dimensional radiological imaging. Therefore two other suggested techniques are 3D-GCT and upper endoscopy. The volumes of the gastric pouch and the diameter of the gastrojejunal anastomosis can be measured exactly in these two techniques [24, 25]. Unfortunately, the exact pouch size was not measured according to a standardized protocol preoperatively in this retrospective cohort study. However, all patients had demonstrated a large pouch, either diagnosed by BST, upper endoscopy combined with BST or 3D-GCT. Patients were excluded from this study if the diagnostic technique showed other causes for insufficient weight loss or weight regain, such as gastrogastric fistula.

The revisional BS technique was based upon the preference of the surgeon. Because of this, two bariatric surgeons performed the RGJ and three bariatric surgeons performed the SGP technique. There was no crossover in treatments between the surgeons. As a consequence, there were significant differences at baseline: gender, BMI before LRYGB, BMI before primary BS and best %TWL after primary BS and BMI before primary gastric bypass. In multivariate analysis, only best %TWL after primary BS was a positive predictor of %TWL 12 months postrevisional. However, as standardization of revisional BS is needed, we analyzed the results to assess whether one technique is preferable. Further research in a randomized controlled trial is recommended to prevent selection bias.

In conclusion, both SGP and RGJ techniques are feasible to perform and achieve adequate weight loss after revisional BS for insufficient weight loss or weight regain as a consequence of a large pouch after LRYGB. There was no statistical difference in %TWL between either procedures during follow-up, and both techniques showed improvement of obesity-related comorbidities. However, the average surgical costs of the SGP technique were lower and may therefore be the preferred revisional bariatric technique.