Introduction
Dumping syndrome (DS) and postbariatric surgery hypoglycemia (PBH) occur in up to 40% after bariatric surgery [
1,
2]. DS is triggered by osmotic pressure of nutrients and subsequent diffusion of fluid into the bowel lumen, while in PBH, consumed carbohydrates cause exaggerated insulin/GLP-1 release and subsequent hypoglycemia [
2,
3]. Most patients can control their symptoms by consumption of small meals with a low carbohydrate content [
4]. Approx. 2.5% of patients develop substantially impaired quality of life [
4]. Unless bypass anatomy as precondition for DS and PBH is reversed, efficient treatment for these patients is challenging [
1,
5,
6]. Treatment targets diminishing insulin response and slowing gastrointestinal transit [
7]. Dietary advice, combining small meals with a low-glycemic index, high fiber, and protein, with slow eating and well chewing is the primary approach. If dietary intervention is not successful, the addition of pharmacotherapy may lead to better symptom control. Scientific data with positive yet variable results are available for acarbose, diazoxide, somatostatin analogues, calcium-channel blockers, sitagliptin, GLP-1-receptor agonists (GLP-1RA), and SGLT2-inhibitors [
1,
3,
8]. Furthermore, continuous glucose monitoring (CGM) has been associated with reduced glycemic variability and allows for earlier intervention in patients with PBH [
9]. Endoscopic pouch outlet reduction led to resolution of DS in 57% 2 years following the procedure [
10]. A retrospective single-center series primarily analyzing revisionary surgery for weight regain after Roux-en-Y gastric bypass (RYGB) demonstrated a relief of dumping symptoms in 58% following silicone ring implantation [
11].
Given limited efficiency of medical and surgical approaches and largely missing prospective scientific data, we tested if silicone ring implantation and pouch resizing lead to clinical symptom improvement of DS/PBH in a prospective single-arm trial.
Materials and Methods
The study protocol was approved by the local ethics committee and conducted in accordance with the principles of the Declaration of Helsinki. All patients gave written informed consent.
Study Design
The study is a single-center, open-label, single-arm trial including patients suffering from DS or PBH following RYGB (DRKS00020293).
Participants
Patients \(\ge\) 18 years with relevant dumping symptoms following RYGB, who had undergone a failed conservative approach for dumping treatment, and were indicated for pouch resizing plus silicone ring placement were eligible to participate in this study. Surgical revision was indicated after discussion in a multidisciplinary team based on insufficient response to dietary modifications, clinically relevant dumping symptoms impairing quality of life or everyday functioning and failed or declined pharmacological dumping treatment. Patients of this study initially received the RYGB operation between May 2013 and August 2019. The majority (10/16) of patients was primarily operated at the author’s institution. During this time, 651 patients had received a primary RYGB. The first patient was included in May 2020, the last patient on October 2022. In this period, 242 patients were treated for dumping by the multidisciplinary team, whereas 185 received medical therapy and 7 an endoscopic transoral outlet reduction (TORe).
Formal study inclusion criteria were defined as Sigstad score > 7 and abnormal values during the mixed meal tolerance test (MMTT; increase in heart rate by 10%, drop in systolic blood pressure by 10% within the first 30 min or serum glucose value < 60 mg/dl or 3.3 mmol/l). The combination of a MMTT with clinical scores was the best-established characterization for a dumping syndrome at that time [
2]. Therefore, this combination was utilized not only for study inclusion but also for follow-up examinations despite the circumstance that a liquid glucose test may underestimate changes induced by pouch restriction.
Exclusion criteria were insulin-dependent type 2 diabetes (T2D), type 1 diabetes, pregnancy, untreated psychiatric illness, alcohol or drug abuse, liver cirrhosis, and Crohn’s disease.
Intervention
Patients received a complete adhesiolysis of the pouch and gastrojejunostomy (GJ). Pouch dilatation was assessed by air insufflation. Pouch resizing was conducted alongside a 35 French bougie using 2–3 loads of a linear Endo-GIA device (Medtronic, purple cartridge). The ring (MiniMizer, Bariatric Solutions) was implanted in a perigastric technique, positioned 3–4 cm below the esophago-gastric junction and a minimum of 2 cm above the GJ. As described previously, constant pressure on the gastric wall should be avoided in a ring-augmented procedure [
12]. Therefore, the ring was closed to leave a circumference of 7.5 cm. Including bougie size and stomach wall thickness, this enabled a loose ring placement.
Outcome Measurements
Primary outcome was defined as Sigstad score after 12 months. The score value was determined at inclusion, 3 and 12 months after revisional surgery.
Secondary Outcomes
Clinical outcome measurements including Sigstad score and Arts dumping severity scale were determined at inclusion, 3 and 12 months postoperatively. MMTT, barium swallow, and gastroscopy were performed at 0 and 12 months.
Carbohydrate tolerance was recorded semi-quantitively as carbohydrate intake as (a) avoided completely, (b) partially possible, and (c) possible without limitation. Additional medical therapy of DS/PBH, number of antihypertensive agents, and nicotine use were recorded.
For the MMTT, blood samples were taken at the beginning and every 30 min up to 240 min following ingestion of 200 ml of Fresubin energy drink (Fresubin® energy fiber drink). Items recorded included insulin, glucose, hematocrit, heart rate, and blood pressure. During gastroscopy, the presence of hiatal hernia and reflux esophagitis, ring position, and pouch length were assessed. In addition, contrast reflux was assessed during barium swallow.
T2D was evaluated recording HbA1c and use of anti-diabetic medication. Quality of life was evaluated using the Bariatric Analysis and Reporting Outcome System (BAROS). Reflux symptoms and regurgitation were documented semi-quantitatively as either not present, incidence of \(\ge\) 1/week, or \(\ge\) 1/month. Additionally, a reflux questionnaire was used (Reflux Symptom Index (RSI)).
Postoperative complications were defined as early/late or major/minor by the guidelines for outcome reporting of the American Society for Metabolic and Bariatric Surgery [
13].
Statistical Analysis
Sample size was calculated to detect a difference in Sigstad score 12 months after revision compared to baseline applying a sign-rank test. It was assumed that ring implantation would lead to score improvement in 90% of patients. Based on a two-sided significance level of 0.05, 12 patients were required. Due to a high uncertainty based on limited data availability and to account for drop-outs, sample size was adjusted to 16 [
14].
Continuous variables were characterized using mean and confidence interval (CI). Categorical variables were summarized using frequencies or percent of patients in each category. For MMTT evaluation, area under the curve was calculated for glucose and insulin. A Wilcoxon signed rank test was used to calculate median treatment differences of continuous variables.
To obtain dichotomous outcomes, reflux symptoms, regurgitation, and dysphagia were grouped as either not present or present at any degree.
All statistical tests were performed at the two-sided 0.05 significance level. Prism 8.3 for macOS (GraphPad Software, LLC) was used for statistical analysis. A P-value < 0.05 was considered significant.
Discussion
This is the first prospective trial analyzing symptoms of DS and PBH following silicone ring augmentation. This intervention led to a sustained Sigstad score improvement by 50% or more at 12 months in 56% of patients. The improvement in Arts dumping score suggested a similar effect on symptoms of DS and PBH. Full dumping symptom resolution was achieved in 25% of patients. In contrast, 1-year retrospective data on silicone ring implantation without pouch resizing in 24 patients demonstrated a resolution of 58%. However, dumping symptoms were measured only semi-quantitatively in this trial [
11]. For TORe, resolution of dumping was described in 67% at 1 year in a set of patients with a considerably lower initial Sigstad score [
10]. This may have altered the rate of complete dumping resolution observed in these studies. Furthermore, TORe may not be ideal in patients with dilated gastric pouches, as outlet reduction in these patients potentially leads to regurgitation. Acarbose, considered to be first-line medical treatment in patients with PBH, led to lower glucose values in a MMTT performed in 11 patients after RYGB and documented PBH, demonstrating the principal effect of this medication. However, acarbose could not decrease the number of hypoglycemic episodes in an outpatient setting in the same trial [
7]. Data from a multicenter case series including 22 patients with DS or PBH demonstrated at least a partial response in 3 of 6 patients for diazoxide and in 8 of 13 patients for a somatostatin analogue [
15]. Lately, GLP-1RA are offered to patients with PBH targeted at reduction of gastrointestinal motility, yet promising data are still limited to a small body of studies [
3]. Clearly, these studies revealing a success rate in often less than two out of three patients demonstrate the challenges of dumping therapy following RYGB. In this constellation, pouch resizing and ring augmentation may be positioned as alternatives to eTOR in second-line treatment after failed medical dumping therapy or for patients unwilling to take lifelong medication. When eTOR has a lower rate of severe complications such as staple line leakage, pouch resizing and silicone ring augmentation could be favorable in patients with dilated gastric pouches [
10,
16].
Patients in this study reported a drastic reduction in almost every item of the Arts dumping score. Especially for characteristic DS items such as diarrhea, palpitations, and dizziness, or items related to PBH such as postprandial hunger, score rating went from “relevant” to “mild.” None of the patients reported drowsiness following the trial intervention. Despite these findings suggesting a retraceable pathophysiologic change, the standardized liquid MMTT revealed largely unchanged glycemic values and vital signs. In light of the aforementioned changes, we would have expected a slower rise in glucose values and vital signs, a later peak, and a smaller AUC as indirect evidence for delayed gastric emptying following silicone ring implantation and pouch resizing. Possibly, a liquid MMTT lacks the precision to detect this change as delayed gastric emptying after silicone ring implantation had previously been demonstrated by a primarily solid test meal [
17].
On the other hand, it is plausible to assume that ring implantation led to consumption of smaller meals. This is a cardinal goal in dumping treatment and may explain the observed benefit. Meal size was not explicitly recorded in this trial.
A recently published meta-analysis on revisional treatments for impaired weight loss after RYGB reported a TWL of 17.2% after pouch resizing and 13.6% for silicone ring implantation [
16]. Weight loss in the current trial was considerably lower. However, the indication for surgery in the current trial was focused on dumping, with only half of the patients having experienced insufficient weight loss or weight regain. Comparing good and poor responders, Boerboom identified patients with weight regain after previous good weight loss to benefit most from secondary silicone ring implantation [
18]. The correlation of weight regained and weight lost after revision in the current trial identifies a similar patient subgroup for a good response after revision in this respect.
This study has several limitations. Diagnosis of dumping syndrome varies largely on the test used, and neither clinical nor functional tests have a high specificity for the diagnosis of DS or PBH [
19]. For example, the MMTT cut-off used for PBH in the current trial was 3.3 mmol/l. Possibly, a stricter cut-off of 3 mmol/l or 2.8 mmol/l would have led to a more rigorous patient selection [
2,
20]. Nonetheless, patients required pathologic findings in both MMTT and Sigstad score for study inclusion. The study design, with repeated measures in the same patient, limited inter-patient variability.
Functional testing in the current trial was confined to a MMTT with a chemically defined meal. Although considered standard of care, this may not be ideal to evaluate treatment effects in patients with DS or PBH [
2]. Despite growing experience with GLP-1 receptor agonists as an effective treatment option for the aforementioned patient group, liraglutide had no effect on glucose or insulin response to a MMTT but led to a lower frequency of hypoglycemic episodes in continuous glucose monitoring of the same patients [
7]. Patients of the current trial did not receive continuous glucose monitoring to assess functional results in a real-world setting. Although all patients had received dietary counselling prior to study inclusion, postoperative diet was not formally standardized. Nonetheless, carbohydrate consumption is relevant for the assessment of PBH. In relation to trial initiation, patients’ carbohydrate tolerance was decreased at 12-month follow-up. This, despite dietary counselling before revision, may be due to an increased awareness of symptom triggers throughout patient treatment in this trial. This may, in part, explain the patients’ clinical benefits independently of ring implantation. Furthermore, the study intervention was a combined intervention of pouch resizing and silicone ring implantation. This design does not allow to discriminate which part of the intervention contributed to the effects reported.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.