Our study showed that TORe was effective in patients with dumping syndrome after RYGB. Perioperative complication rate was low and a significant weight loss in some patients and dumping resolution could be achieved during follow-up. Even as revisional intervention after failed surgical procedure for secondary weight regain TORe can be effective.

The present analysis represents data from one of the largest obesity and bariatric surgery centers in Switzerland with 300 procedures performed at three different sites per year. All patients undergoing an endoscopic procedure for secondary weight regain or unresolvable dumping syndrome or both between January 2015 and December 2020 were included. Although some studies have evaluated the effects of endoscopic revisions with different endoscopic suture systems and different techniques, long-term follow-up data is still scarce. Callahan has conducted a retrospective analysis on 70 subjects with weight regain after bariatric surgery [20]. The same endoscopic suturing system and equipment as in our study was used, with the difference that both interrupted, and purse-string sutures were applied. The mucosa was circumferentially ablated by argon plasma coagulation. An anastomotic diameter of 5–9 mm was achieved. These technical details are in accordance with our study, except for the suturing technique. In contrast to the Callahan cohort, in our subjects exclusively interrupted sutures were used. In the aforementioned study a % EWL of 14.9% after 1 year, 12.2% after 2 years 8.7% after 3 years, 3.2% after 4 years and 7% after 5 years was described. These results were slightly higher, at least in the early post-interventional course, compared to our cohort. Our subjects had an EWL of 11.4% at 12 months and at 2 years the EWL data did not differ so much: 8.2% in our cohort and 12.2% in the Callahan study. Long-term data still shows a positive effect after 5 years, but in our study the nadir weight loss was achieved 3 and 6 months after the intervention. Nevertheless, further weight regain can be avoided up to 3 years post-intervention and the Kaplan–Meier curve showed no weight regain after endoscopic intervention in more than 75% of our subjects, even after 48 months follow-up, which is comparable to the Callahan study.

Concerning post-interventional complications, a total of four complications were mentioned: two subjects with gastrointestinal bleeding (one occurring intraoperatively, and one requiring readmission due to melena). One subject was readmitted with obstruction, which resolved after nasogastric tube decompression. Another one was readmitted with perforation 1 day after the intervention and the lesion was sutured laparoscopically. The complication rate seemed to be comparable to our study, with a total of three perforations in 71 patients. Concerning the severity of complications in both the Callahan and our study, there was only one reported Clavien-Dindo IIIb, as two of the perforations in our cohort were recognized during the initial intervention and could be closed endoscopically with clips. Nevertheless, it is striking, that in our cohort the occurrence of perforations seemed to be higher while in the compared study the rate of bleeding was more frequent. Most subjects in the study mentioned above were discharged at the day of revision (85%, 57/67), which is comparable to our results (88.7%, 63/71). Nowadays almost all patients are discharged at the day of the intervention.

In another study, Tsai et al. analyzed 107 subjects with a mean follow-up time of 9.2 months. Most subjects suffered from secondary weight regain (n = 81), whereas 13 of them presented with dumping syndrome and further 13 with both dumping syndrome and weight regain [21]. The mean BMI before the intervention was 32.9 kg/m2 which is comparable to our cohort (mean BMI 34.5 kg/m2). They could demonstrate a mean weight loss of 8 kg after 12 months compared to a mean weight loss of 3.3 kg after the same time period in our cohort. The complete absence of complications is remarkable. With a mean follow-up of 9 months, the follow-up period was much shorter compared to our study. The authors reported that in 26 subjects a second TORe procedure was necessary and in 13 others a laparoscopic pouch revision was required. Technical details seemed to be analogous to our interventions. Tsai et al. routinely used 1 or 2 sutures with 4 to 6 stitches and primed the anastomosis with argon plasma coagulation. Two sutures resulted in better weight loss. However, long-term data is missing.

The treatment strategy was identical to the algorithm at our center. After 1 or 2 endoscopic procedures a surgical revision is recommended. Depending on the preoperative examinations and the intraoperative findings, a pouch resizing, a revision of the anastomosis or an implantation of a silastic ring is performed.

Dhindsa analyzed 850 subjects from 13 independent cohorts in a meta-analysis [22]. He reported a short-term weight loss of 6.14 kg, 10.15 kg, and 7.14 kg at 3, 6, and 12 months respectively compared to 4.8 kg, 4.9 kg, 3.3 kg at the same time points in our subjects. Results seemed to be slightly better than in our study group. The TWL (%) was more pronounced in the present meta-analysis compared to our study where a TWL of 6.69% vs.4.8% at 3 months, 11.34% vs. 4.7% at 6 months and 8.55% vs. 30.2% at 12 months was seen. The overall frequency of severe adverse events in TORe was rather low with a reported rate of 0.57% ± 1.35%. This rate was clearly lower than in our cohort (hematemesis n = 12.2%, and perforation treated endoscopically n = 36.7%).

In the present analysis a TORe has been performed in 26 subjects to treat dumping syndrome. Twenty-two subjects (84.6%) showed an improvement of symptoms. Four subjects showed recurrent symptoms and in three of them a revisional surgery was performed (pouch revision and implantation of a silastic ring). In accordance to our results, Tsai reported an initial success rate of 77.5% in 40 subjects whereas nine suffered from recurrence of dumping syndrome and needed repeated TORe (n = 7) and laparoscopic revision (n = 2) [23].

In a multicenter study, Vargas et al. reported 115 patients with dumping syndrome non-responsive to medical therapy. After endoscopic narrowing of the gastrojejunal anastomosis a significant reduction of Sigstad`s score and a very low failure rate in nine subjects (3%) was noticed. Subjects with recurrent symptoms underwent either another TORe for gastrojejunal anastomosis dilation (n = 3 out of 9) or an insertion of a feeding tube (n = 3 out of 9) [24].

Petchers et al. analyzed 98 patients which were treated with a TORe procedure using OverStitch™ after gastric bypass surgery for obesity and dumping syndrome. They reported a resolution rate of 88% and positive results of 84% remained even after a long-term follow-up of 3, 5 years on average. The effect of the TORe procedure remained in almost all subjects even years after the initial procedures. Only 7% of all cases needed a second procedure for recurrence of symptoms and endoscopic evidence of recurrent anastomotic dilation 2–3 years after the initial intervention [25].

This study has some limitations. First, follow-up intervals are inconsistent. Although we established a protocol for medical appointments, some subjects could not entirely respect it. The study design is retrospective; data quality depends on correct documentation and completeness of intervention variables, documentation of weight trajectories, and relief of dumping symptoms. Furthermore, the endoscopic procedure is not compared to a control group (e.g. conservative intervention through medical nutrition therapy, health coaching, physical activity, or a surgical reduction of the gastrojejunal anastomosis). Unfortunately, Sigstad`s dumping score or Arts` questionnaire were not assessed systematically to distinguish between early and late dumping syndrome. Finally, the size of the gastric pouch after the procedure was not recorded in our patient collective.