Morbid obesity is defined by a body mass index (BMI) greater than 40 kg/m2, as outlined by the National Institutes of Health (NIH) consensus conference in 1995 [1]. Morbid obesity has been associated with increased mortality when compared to nonobese individuals [2]. It is associated with serious comorbidities, including hypertension, diabetes mellitus, dyslipidemia, obstructive sleep apnea, and osteoarthritis. Medical treatment for obesity has had disappointing results, particularly in morbidly obese individuals. Currently the Roux-en-Y gastric bypass (RYGB) is recognized as the standard surgical procedure to treat morbid obesity [2, 3]. The number of bariatric surgeries performed every year is growing because of the increased number of morbidly obese individuals that fail medical treatment. LRYGB has a shorter recovery time than the conventional open approach, yet is thought to have the same long-term results [4–7].A relatively common complication of LRYGB is the gastrojejunal (GJ) stricture, which has been reported in 3–27% of patients. Strictures can lead to refractory vomiting and nutritional deficiencies. The exact mechanism responsible for stricture formation is unknown. Detailed reports of endoscopic management of GJ strictures are limited. EGD with through-the-scope (TTS) balloon dilation has been advocated as an effective therapy for this complication. Few studies have been published regarding the efficacy and safety of endoscopic balloon dilation for the management of GJ anastomotic stenosis. We report our center’s experience in the management of GJ strictures after LRYGB, determining its incidence, safety, and complications of balloon dilatation. To our knowledge this is the largest series reported in the literature.

Materials and methods

Institutional review board approval was obtained for a retrospective review. One thousand and twelve patients underwent a LRYGB from January 2001 to May 2004 at our institution. Sixty-one patients were found to have GJ strictures by initial UGI series or EGD. Some patients had repeat dilations if the stricture was very tight on initial endoscopy or if symptoms recurred; patients would return after an interval of 2–4 weeks for repeat dilation. Demographic and clinical information was collected from each of these patients.

The LRYGB operations were performed by two laparoscopic surgeons using a modified technique initially described by Wittgrove and colleagues [6]. In summary, a 15–30 ml gastric pouch was created with a 3.5-mm stapler. The small bowel was transected 45 cm below the ligament of Treitz (biliopancreatic limb) and the alimentary Roux-en-Y limb is brought up to create the gastrojejunostomy with a 45-mm linear cutter and a 2.5-mm stapler, reinforced with hand sewn sutures. The gastrojejunal (GJ) anastomosis was tested intraoperatively for leakage by using air insufflation and methylene blue test. The anastomosis was created to be tight enough to allow proper weight loss with a maximum diameter of 12 mm. The length of the alimentary limb depended on the patient’s BMI. For the patients with a BMI between 35 and 50 kg/m2, a 100-cm alimentary limb was performed; for individuals with a BMI between 51 and 60 kg/m2, a 150-cm limb was created; and for the those with a BMI greater than 60 kg/m2, a 200-cm limb was necessary. The jejunojejunal anastomosis was performed with a 45-mm linear cuter and 2.5-mm linear stapler. Discharge was planned for the first 48 h postoperatively, if there were no complications observed.

All patients suspected of having anastomotic strictures underwent UGI series or EGD after informed consent was obtained. Under intravenous sedation with Propofol, an adult upper-endoscope (Pentax) was used to inspect the size and integrity of the gastric pouch. The gastrojejunal anastomosis was visualized routinely and the opening size was estimated and recorded in each case by the endoscopist. A clinically significant stricture was recorded if the scope could not pass through the anastomosis or if the diameter was less than 10 mm.

Once the stricture was diagnosed, it was treated during the same procedure by passing a TTS balloon dilator with or without guidewire. The balloon catheters were selected from either Cook CRE (Winston-Salem, NC) or Boston Scientific (Watertown, MA). During the same procedure the strictures were dilated with progressively larger size balloons, from 6 to 18 mm, if necessary. The balloon was inflated for 30 s each time, followed by inflation for 1 min with the final balloon size. If the strictures were not successfully dilated in the first attempt or if symptoms recurred, the patient would come back in 2–4 weeks for another EGD with possible dilation.

After adequate dilation, the scope is advanced to explore the distal anatomy and exclude anastomotic ulcers. All patients received acid suppression with proton pump inhibitor after the procedure for 2 months or longer (4–6 months if uclers were present) if ulcers were present.

Results

Of the 1012 LRYGB patients, 61 patients (46 females, 15 males) were found to have GJ strictures. The mean age of patients was 44 years (range 20–69 years, see Table 1). The incidence of strictures was 6%. Other findings were observed as follows: 13% had anastomotic ulcers, 6.5% had pouchitis, 4.9% had esophagitis and 3.3% had bezoar (Table 2). The most common comorbidities in our patients were: obstructive sleep apnea (59%), osteoarthritis (57%), hypertension (47%), and diabetes (21%).

Table 1 Demographic and clinical data of studied patients
Full size table
Table 2 Other endoscopic findings in patients requiring dilation
Full size table

In total, 128 dilatations with TTS balloons were performed. Seventeen patients (28%) had symptoms relieve with single dilation, and did not require further intervention. Forty-three patients (72%) required repeat dilations, with one patient requiring five dilations (Table 3). The mean time from the surgery to the onset of symptoms was 2 months (range 1–6 months). The size of TTS balloons used ranged from 6 to 18 mm for all dilations, depending on the diameter of the GJ stricture, which ranged from 2 to 10 mm. Two patients developed severe abdominal pain after balloon dilation. Both were hospitalized for observation and discharged in 24 h after normal UGI study and symptom relief.

Table 3 Number of dilations per patient
Full size table

Three out of 61 patients had classic symptoms and radiological findings of bowel perforation after dilation, with 2.2% incidence of perforation in our series. All these dilations were done with a maximum balloon size of 13.5 mm (Table 4). All the patients were young females and supermorbidly obese. Two perforations occured after first endoscopic dilation and a third after three previous dilations. All three patients underwent laparoscopic surgical exploration, without localizing the site of perforation by water test. It was felt that patients had sealed microperforations. All cases were managed with supportive treatment postoperatively (bowel rest, antibiotics intravenously for 5–7 days, and drains placement). None of the patients required surgical revision of the gastrojejunal anastomosis.

Table 4 Characteristics of the patients with perforation after dilation
Full size table

Discussion

LRYGB has shown a number of advantages over an open RYGB including shorter procedure time and less complications [6]. GJ strictures have been described with both the open and laparoscopic procedures, and there has been a concern that this problem may be more commonly after laparoscopic technique. Symptomatic GJ anastomotic stricture is a relatively common finding following gastric bypass and needs to be aggressively investigated and appropriately treated [7]. A handsewn technique (such as the one used in our series) using linear staplers instead of circular staplers, has been reported to have a lower rate of GJ strictures [8]. The cause of the stricture is not well understood, but ischemia caused by the stapler, tension on the anastomosis, edema, or a foreign-body reaction are recognized factors [9]. Patients with GJ strictures present with nausea, vomiting, and/or abdominal pain.

Several techniques have been used to treat GJ strictures. Currently endoscopic balloon dilations are used as a standard technique. The early endoscopic approach used pneumatic balloon catheters, replaced by the through-the-scope ones.

Some patients may require multiple dilations. It has been our practice to schedule subsequent EGD when, after the initial dilation, the anastomosis was very narrow or the endoscopist was concerned about perforation risk. If the initial stricture was dilated up to 10 mm, then it was recommended to perform another procedure after 2–3 weeks, with subsequent escalating dilations until the scope is able to advance or a minimum of a 12-mm stoma is achieved. In our series we followed a minimum of 2-week interval between dilations, including the patient with perforation. The maximal dilation size still needs to be determined. In our series one patient had 18-mm balloon dilation without complications. Unremarkable dilations with 15–16 mm balloon size were performed in several symptomatic patients, so the balloon size alone does not seem to be a risk factor for perforation. Other similar features shared by these patients were the fact that they were supermorbidly obese, and diabetes mellitus and hypertension were observed in two of them.

In light of the fact that none of these patients required surgical revision of the GJ anastomosis, despite exploration, it remains to be seen whether the optimal treatment of perforation should be either conservative medical management or operative intervention.

Previous authors have described their outcome with strictures after LRYGB (Table 5), emphasizing the surgical technique; only a few reports have published specific outcomes regarding management of strictures [19–22]. The incidence of strictures ranged tremendously, from 3% to 27%, likely related to different surgical techniques and experience.

Table 5 Incidence of GJ anastomotic stricture in selected series of LRYGB
Full size table

The role of steroid injection into the anastomosis has to be explored. From our personal experience the use of steroid injection added no benefit. Most of the patients in these series responded well to balloon dilation.

Conclusion

In summary, we believe endoscopic balloon dilation is a highly effective treatment for GJ strictures after LRYGB and should be considered as a primary intervention prior to proceeding with a surgical revision. We describe one of the largest series evaluating endoscopic management of these strictures. The incidence of strictures in our series was relatively low, which was similar to most previously published reports, but multiple dilations were still necessary. We had 2.2% incidence of perforation and the only common factors among these patients were diabetes, hypertension, and the fact that they were all supermorbidly obese. The size of the balloon used and the number of dilations do not seem to play a significant role in the risk of perforation.