INTRODUCTION

The World Health Organization (WHO) defines being overweight and obesity as an abnormal or excessive accumulation of fat that can be harmful to health, with a considerable associated morbidity and mortality ¹. The rising prevalence of obesity and its associated metabolic diseases have led to a marked increase in the number and type of bariatric procedures. Today, bariatric surgery remains the most effective long-term therapeutic option for patients with morbid obesity. Furthermore, RYGB remains the predominant bariatric surgical procedure, which results in an average excess weight loss (% EWL) in the range of 65-80% after 2-3 years ^2,3. However, after the initial weight loss many patients can regain weight, either partial or total. One study reported an average weight regain of 18 kg after two years ⁴; at ten years post-RYGB, the average weight regain was reported to be over 30% of the weight initially lost and over 25% of patients regained almost all the weight lost ^5,6.

Etiologies and mechanisms for weight regain are not fully understood and the process is likely multifactorial. This has been related to preoperative factors (previous body mass index [BMI] and genetic factors), endocrine-metabolic factors (patient dietary adherence or neuroendocrine-metabolic dysregulation) or mechanic factors (type of surgery, gastro-gastric fistula or enlargement of the gastric pouch) ^7,8. Recently, a positive linear association between the size of the gastro-jejunal anastomosis (GJA) and weight regain has been reported. A GJA diameter > 15 mm was proposed as a cut-off for dilated GJA ⁹.

Weight regain treatment may be surgical or endoscopic. Surgical revisions include the recreation of the gastro-jejunostomy, reduction of a gastric pouch, placement of an adjustable gastric band, taking down the Roux limb and distal gastric bypass. Although effective, surgical techniques have a high complication rate of up to 50%. The mortality rates are 2%, which is more than double that of the original procedure. Furthermore, there are high costs and many patients do not wish to undergo surgery ^10,11. Therefore, less-invasive, endoscopic revision methods can have an enormous potential in these patients. To date, sclerotherapy techniques and different experimental models of gastric sutures have been used ^12,13,14,15.

This is the first Spanish series published using the full-thickness endoscopic suture system, OverStitch(r) (Apollo Endosurgery, Austin, Texas, USA), for this condition. The study includes a retrospective descriptive review of our initial experience on the technical feasibility, safety and early efficacy of the endoscopic transoral outlet reduction (TORe) procedure in patients with weight regain after a RYGB and a short-term follow-up. Plications were applied around the GJA to reduce the aperture and also in the gastric pouch to provide further size reductions. The patients were followed-up for six months.

PATIENTS AND METHODS

Consecutive patients submitted to a TORe evaluation from June 2014 to June 2016 with weight regain after RYGB were included in the study. The time interval between RYGB and endoscopic re-intervention was eleven years (range 3-16). All gastro-jejunal anastomosis (GJA) were performed with mechanical sutures; they were dilated with a diameter of > 15 mm. Patients with a GJA of < 15 mm and cases that underwent suturing procedures for a gastro-gastric fistula or marginal ulceration were excluded from the study. The procedures were performed by three endoscopists at the Dexeus University Hospital in Barcelona and at the Hospital Universitario Sanchinarro in Madrid, Spain.

TORe was performed using the OverStitch(r) Endoscopic Suturing System (Apollo Endosurgery, Austin, Texas, USA), which attaches to a double-channel Olympus gastroscope (GIF-2T160, Olympus America, Central Valley, Penn, USA) that is passed through a specific overtube, 25 cm in length. TORe was performed under general anesthesia with endotracheal intubation controlled by an anesthesiologist, with antibiotics and proton pump inhibitors (PPIs) prophylaxis. The procedure was performed by three endoscopists with prior experience in advanced suturing methods (e.g., endoscopic sleeve gastroplasty) using the same endoluminal suturing system. However, they had no previous experience in stoma reduction; a standard gastroscopy was first completed to discard contraindications and to measure the diameter of the anastomosis and the pouch. These were measured using a polipectomy handle or a balloon (anastomosis) and distance from the incisors to the anastomosis and Z-line (pouch). Tissue at the rim of the anastomosis was ablated circumferentially using argon-plasma coagulation at a setting of 60 W and a flow rate of 1.6 l/min, in order to improve the narrowing of the lumen by healing and strengthening the fusion of the sutured edges. Interrupted transmural stitches were placed in the anastomosis from right to left until reaching a diameter < 12 mm. Patients with a dilated pouch (greater than 5 cm in length) also underwent a suture and reduction, beginning at the distal region until reaching a size of < 5 cm.

After the procedure, the patients remained in observation for a few hours and were discharged on the same day if there were no clinically apparent adverse events. Each patient was managed by the endoscopist for the first week. Patients were then followed by a nutritionist, who initially established a regimen based on a liquid diet during the first two weeks, followed by soft intake for another two weeks. From day 30 to six months, a standard hypocaloric diet was followed, which is used for other obese patients. Furthermore, the nutritionist used the same dietary guidelines as those established before TORe. Therefore, the endoscopic procedure could be considered as the only variant and modification related to weight changes. PPIs were maintained for one month and clinical follow-up was performed at one, three and six months.

The study was approved by the Ethics Committee of the Hospital, all details of the technique were explained to all patients and they gave their written informed consent for the procedure and follow-up.

RESULTS

A total of 13 patients (8 females) were included in the study. The mean patient age was 50 years (range 39-66). The average interval between RYGB and TORe was eleven years (range 3-16), with a mean maximum weight loss post-RYGB of 37.69 kg and a subsequent mean weight regain of 21.62 kg (Table 1). At the time of the TORe, the patients had an average weight of 98.48 kg (range 80.00-131.80), which was equivalent to a mean excess weight of 35.03 kg (55.69% of excess weight) and a mean BMI of 37.33 kg/m² (range 30.12-45.60) (Table 2).

Table 1 Clinical characteristics related to RYGB 

M: male; F: female; HT: arterial hypertension; DLP: dyslipidemia.

Table 2 Efficacy of TORe 

M: male; F: female; W: weight; WL: weight loss; EWL: excess weight loss; BMIL: BMI loss; No. SUT-GJA: number of sutures in the gastro-jejunal anastomosis; WL-RYGB: weight loss post-RYGB; WR-RYGB: Weight regained post-RYGB; %WRL-RYGB: % of weight lost regained post-RYGB.

The average diameter of the GJA was 36 mm (range 20-45). After preparing the anastomosis with argon plasma coagulation, an average of 2.5 sutures were used (range 2-3). This reduced the anastomosis to a mean diameter of 9 mm (range 5-12), which resulted in an average reduction of 27 mm (75% reduction) (Table 3). A final diameter of less than 12 mm was obtained in all cases. There was one case of weight gain with a re-dilation of the stoma (from 10 to 20 mm after 12 months), which was successfully reduced to 9 mm without any incidents. The average size of the pouch was 7.2 cm (range 2-10) and was dilated in 11/13 patients. An average of 2.7 plications were created (range 0-4), reducing the mean length to 4.7cm (range 4-5). This resulted in an average decrease of 2.5 cm (34.72% reduction) (Table 3).

Table 3 Main technical aspects of TORe 

Ø: diameter; GJA: gastro-jejunal anastomosis.

The average procedure time was 50 minutes (range 120-25 minutes). The most durable were the first cases and those that also required a joint stoma and pouch suture. The shorter ones required only the stoma suture. There were no complications related with the procedure and TORe was well tolerated in all cases. One patient had an intra-procedural self-limited bleed from a needle puncture site. Epigastric pain was common for the first 48 hours. Nausea was experienced in two patients during the first 24 hours. All were treated conservatively and were discharged on the same day of the intervention; there were no delayed adverse events during follow-up.

Follow-up data were available for 12/13 patients at one, three and six months. There was an average weight loss of 6.19, 10.06 and 12.29 kg, respectively, which represents a EWL of 16.37, 26.17 and 28.41% and an average BMI loss percentage of 6.96, 10.43 and 12.40%, respectively. These results demonstrate a loss of 28.63, 46.53 and 56.85% of post-RYGB weight regained (Table 2). The weight loss at six months in the eleven patients treated simultaneously for a GJA and gastric pouch (12.8 kg, 29.52% EWL) was somewhat higher compared to the two cases treated only for a stoma (11.3 kg, 26.60% EWL). However, the difference was not statistically significant.

DISCUSSION

RYGB represents the most effective treatment for morbid obesity and its co-morbidities ^2,3. Nevertheless, many patients tend to regain weight over time ^4,5,6. Surgical revision is one option, although it has a higher rate of morbidity and mortality than the primary surgery ^10,11. For this reason, there is an increasing interest in less invasive techniques, especially endoluminal ^12,13,14,15.

In this study, we describe our initial experience of the technical feasibility and weight changes that resulted from the transoral outlet reduction (TORe) in patients with weight regain after RYGB and with a dilated GJA. Our patient cohort had an average weight regain after RYGB of 21.62 kg, and the data in the literature varies and oscillates in some studies between 13-23 kg ^16,17 or 52.6% of the lost weight ¹⁸. Dilated GJA was defined as an anastomosis diameter of > 15 mm ⁹. The success of the technique was defined as its reduction to < 12 mm, which was achieved in all cases. These results are similar to those described previously. A rate of 94.9% was described by Changella et al. ¹⁶ and one of 100% by Jirapinyo et al. ¹⁸. In addition, when the gastric pouch was dilated with a measured length of > 5 cm, this was thought to require a reduction as an adjunct to the anastomosis reduction in the same endoscopic procedure. Combined stoma and pouch treatment is scarcely documented in the literature ^15,19.

Fig. 1 The TORe procedure. A. Dilated stoma. B. Circular treatment with argon. C. Reduction of the diameter with an OverStitch of Apollo. D. Reduced diameter, the final result. 

We were able to decrease the mean diameter of the GJA from 36 to 9 mm (range 5-12), which represents a 75% reduction of the initial diameter. The final result can be compared to two published studies by Jirapinyo et al. and Kumar et al. ^20,21, where the mean diameter was reduced to 6 and 6.9 mm. The review by Changela et al. ¹⁶ reported an average GJA reduction from 24.85 to 8 mm. One patient in our cohort required a second TORe due to a partial failure of the first procedure, due to weight regain and the GJA re-dilated after 12 months. We were able to successfully reduce the stoma in the second TORe, which supports the possibility of sequential treatments with this system. In this sense, Kumar et al. have estimated that the number of sessions required to stop weight regain was 1.0 at six months, 1.1 at one year and 1.2 at two and three years. The number of sessions required to maintain a weight loss of > 5 kg from TORe was 1.2 at six months, 1.5 at one year, 1.9 at two years and 2.0 at three years ²¹.

No major adverse events were reported in our study; three intra-procedural events were described in 25 patients in the first series published using a first generation suturing device ²². These included one small esophageal abrasion from the overtube treated with fibrin glue and two patients with arterial bleeding that was stopped by over-sewing. There were also six post-procedural events; two hemorrhages and four cases of major vomiting, two with torn sutures and a GJA stenosis in another case that required endoscopic balloon dilation. In subsequent series, the procedure was a safe and reliable intervention with few complications, and there were only some isolated cases of bleeding and stenosis ^{16,17,18,19,20,21}.

In our study, all procedures were performed with general anesthesia with endotracheal intubation. Prophylaxis with PPIs and intra-procedural antibiotics were systematically provided due to the placement of transmural sutures. The safety, good tolerance and short duration of the procedure (less than other series; an average of 74 minutes was reported by Changela et al. ¹⁶⁾⁾ allowed us to perform the procedures in the endoscopic room with a same day discharge (outpatient procedure).

The patients were under the care of the endoscopist the first week after the procedure and subsequently, of a dietician. In our opinion, the post-procedure management plan used in this study was effective. From 30 days after TORe and up to six months later, the dietary guidelines used by the dietician were similar to those established before TORe and also standardized for all obese patients. No specific diet was established for these patients, thus the endoscopic procedure was the only factor that could affect weight changes at three and six months of follow-up. In the same way, a randomized multicenter clinical trial also concluded that the weight loss efficacy of patients that underwent TORe was greater than that of the control group ²².

With regard to efficacy, an average weight loss of 10.06 kg at three months was obtained. These results are similar to four studies and a review of 59 patients which obtained a 10.1 kg weight loss ¹⁶. At six months, the weight loss was 12.29 kg (28.41% EWL); this data is similar to that published by Jirapinyo et al. (11.7 kg) ²⁰ and somewhat better than the study by Gitelis et al. (5.6 kg, 12.4% EWL) ¹⁷. This may be due to our follow-up care program, a somewhat lower initial BMI or a gastric pouch reduction in most of our patients. Kumar et al. ²¹ performed the longest series published to date and reported an average weight loss of 10.5-9.0-9.5 kg (24.9-20.0-19.2% EWL) at one, two and three years, respectively. Jirapinyo et al. ¹⁸ observed a total weight loss of 9.6% and 8.4% at six and 12 months, respectively. In addition, the technique also seems to improve metabolic diseases ^18,22, although this was not evaluated in our study. In our series, two patients had no weight loss after six months of follow-up (patients 10 and 13) (Table 2). Both patients did not adhere to the dietary instructions and neither made any additional lifestyle changes. The patient that was lost to follow-up had changed their care to another hospital (patient 7).

With regard to the eleven patients treated simultaneously with a GJA and pouch, the weight loss was somewhat higher in the two cases treated only for the stoma. However, the difference was not statistically significant. Schulman et al. ¹⁹ compared the TORe (traditional pattern of simple sutures) with the purse-string, and no significant differences in weight loss after six months were identified. However, they concluded that although TORe is effective, the purse-string technique results in a greater weight loss than the traditional interrupted suture pattern 12 months after the procedure.

The study presents several limitations, such as the retrospective descriptive design, the absence of a control group (only the weight changes before and after the procedure are described), the small sample size of 13 patients, the short follow-up time period of six months, the lack of prior experience with this technique and the fact that only two centers participated in the study. We did not assess the effect of the procedure on the metabolic syndrome. Endoscopic follow-up was not systematically performed to confirm the diameter of the GJA or the pouch, since the study was not a sponsored clinical trial.

Although the results are quite similar to the few published reports in the literature, future randomized interventional studies with a greater number of patients and longer follow-up will be necessary to confirm our results. From a technical point of view, it will be necessary to define improvements in the device, the endoscopic technique, the suture pattern and the location of the stitches. Based on our short experience, we feel that there is a need to complement the reduction of the diameter of the GJA with the reduction of the gastric pouch when it is dilated.

In summary, the results of this preliminary descriptive case series suggest that TORe with the OverStitch Endoscopic Suturing System could be a feasible and safe procedure for the management of weight regain after RYGB. The reduction of the GJA seems to offer good weight loss results in the short term, especially if it is complemented with a gastric pouch reduction and supported with a regimented dietary and nutritional control. These initial results of feasibility, efficacy and safety should be confirmed in future randomized clinical trials.