Introduction

Obesity prevalence has risen dramatically over the last decades, and it is estimated that one third of the world’s population is overweight or obese [1]. In the USA, obesity is the second most preventable cause of mortality, and forecasts show that by 2030, 50% of the American population will be obese [2,3,4]. Obese patients undergoing bariatric surgery have a greater likelihood of remission of obesity-related comorbidities, as compared to those receiving medical treatment [5].

Currently, bariatric operations with intestinal bypasses are created using sutures or surgical staplers [6,7,8]. A third approach based on endoscopic magnetic compression anastomosis has also been described in both animal and human studies [9,10,11,12]. Compression anastomosis was first described in 1826 by Felix-Nicholas Denans, long before the advent of mechanical staplers [13]. Since then, multiple compression anastomosis devices, including most recently the compression anastomosis ring (CAR) [14, 15], have been used safely and effectively, but mostly for colorectal anastomoses, and have required surgical placement. Theoretical advantages of using magnets include its ease of delivery, greater strength, dynamic slow compression of tissue, lack of foreign material left within the anastomosis long term, and operator independency [16]. However, a significant historical disadvantage of compression anastomoses has been the size of the magnet limiting the size of the anastomosis, which ultimately jeopardizes its long-term patency [17, 18].

Our study group has developed a novel and minimally invasive method to create an intestinal anastomosis that is ideally suited for use in the duodenum. Self-assembling magnetic octagons using minimally invasive techniques (endoscopy and laparoscopy) were used to create a dual-path enteral diversion, side-to-side anastomosis in which both the diversion and “native” routes remain patent. The self-assembling magnetic device is a system of self-pairing magnets with endoscopic and laparoscopic delivery systems. After the magnetic octagons couple, they create a diversion over several days through a process called compression anastomosis whereby the body remodels in response to a constant compressive force across two approximated viscera. The tissues between the two segments of the intestine heal as the tissue within the magnet necroses. Once the anastomosis has formed, the mated magnet complex breaks free into the enteral flow and is excreted in the patient’s stool.

This system represents the first type of compression anastomosis device for use in the intestinal tract that is capable of forming a permanent and large-caliber (25-mm internal diameter) anastomosis using a device that can be delivered through the working channel of a flexible endoscope or through a 5-mm laparoscopic port.

The aim of this study was to determine technical feasibility, safety, and effectiveness of sutureless duodeno-ileal anastomosis in obese patients with type II diabetes using self-assembling magnets in a Sutureless Neodymium Anastomosis Procedure (SNAP).

Methods

Study Design and Patients

This was an open-label, prospective, and single-arm study conducted at one institution, approved by the Institutional Review Board and the National Health Regulatory Authority.

Eligible participants should fulfill the following inclusion criteria: age 18–65 years, body mass index (BMI) 30–50 kg/m2, and type II diabetes (diagnosis for more than 6 months and less than 10 years, taking at least one oral antidiabetic medication with fasting glucose < 200 mg/dl, and HbA1c 6.5–10% at the time of enrollment). In addition, any other comorbidity (if present) should be adequately controlled with medication, and smoking should be ceased during the study.

Key exclusion criteria included type I diabetes, titanium and/or nickel allergy, insulin requirement, pregnancy, current malignant disease, presence of a pacemaker or implantable cardioverter defibrillator, and coagulopathy.

Enrollment

Individuals who met the inclusion and exclusion criteria were screened by behavioral, nutritional, and medical evaluations. After all evaluations were completed, a consent form explaining the procedure, required follow-up, risks, and potential benefits was signed by all participants.

Procedural Technique

After induction of general endotracheal anesthesia, the patient was positioned supine in a low lithotomy position with the lower extremities extended on stirrups with pneumatic compression stockings, and knees flexed 20–30°. Both arms were left abducted and secured on a board with adequate padding. The Veress needle was placed at Palmer’s point, and pneumoperitoneum was established at 12 mmHg. Five ports were used for the procedure. The ileocecal valve (ICV) was identified, and an ileum loop 300 cm away from the ICV was grasped. A 5-mm ileotomy was created, and the magnet was delivered with a laparoscopic deployment tool under fluoroscopic guidance. The enterotomy was then closed with a single stitch of an absorbable suture (i.e., polyglactin 3.0) (Fig. 1).

Fig. 1
figure 1

Laparoscopic delivery of the magnet into the ileum (a enterotomy; b delivery of magnet into the ileum; c fluoroscopy during delivery of magnet into the ileum; and d closure of enterotomy)

Full size image

An upper endoscopy was then performed, delivering the magnet in the duodenum (2 cm distal to the pylorus) with an endoscopic deployment tool (Fig. 2). The ileum loop with the inserted magnet was then approximated towards the duodenum, and both magnets were coupled under laparoscopic and fluoroscopic guidance (Fig. 3). Instruments and trocars were then removed from the abdomen under direct vision.

Fig. 2
figure 2

Endoscopic delivery of the magnet in the duodenum, 2 cm distal to the pylorus (a magnetic octagon delivered through the biopsy channel of a flexible endoscope; b laparoscopic view of the magnet against the anterior duodenal wall; c magnet deployed inside the duodenum; d magnet attached to the cap)

Full size image
Fig. 3
figure 3

Both magnets are coupled under laparoscopic and fluoroscopic guidance (a schematic image showing how both magnets should be coupled; b the ileum loop containing the magnet is moved towards the duodenum; c laparoscopic view of the engagement of both magnets; d fluoroscopic confirmation that both magnets are coupled correctly)

Full size image

Postoperative Care

Patients were extubated immediately after completion of the procedure. Patients were fed the morning after the operation with clear liquids and usually discharged after 24 h, with instructions to continue with liquid/soft diet for the first 2 weeks, with no specific dietary restrictions thereafter. An abdominal X-ray was performed at 2 and 4 weeks after the procedure to determine the localization and/or expulsion of the magnets (in case the patient did not notice the magnets in the toilet). Follow-up endoscopies were performed at 2 and 12 months after device placement to confirm patency of the anastomosis.

Main Outcomes and Measures

The primary endpoints of the study were technical feasibility and safety: magnet deployment through the laparoscopic port and endoscope channel, engagement of the magnets, and device-related adverse events (30-day morbidity using Clavien-Dindo classification). The secondary endpoints include patency of the anastomosis, HbA1c reduction, and weight loss at 12 months after the procedure.

Results

A total of 8 patients were enrolled in the study; 4 (50%) were male patients, the median age was 51.5 years (range: 34–65) years, and median BMI was 38.8 kg/m2 (range: 35–47.9). Patient demographics are summarized in Table 1.

Table 1 Demographics of patients included in the study
Full size table

Laparoscopic and endoscopic delivery of the magnets into the desired segments of the bowel (ileum and duodenum) was successfully accomplished in all patients. Engagement of the magnets was also achieved in all patients. The mean procedural duration (trocar in–trocars out) was 63.5 min. The mean enterotomy creation to magnet coupling procedural duration was 52.3 min in the first 3 procedures and 36.3 min in the last 3 procedures. No intraoperative complications were recorded.

All patients were discharged on postoperative day (POD) 1; 30-day morbidity occurred in 6 (75%) patients (5 Clavien-Dindo grade I and 1 Clavien-Dindo grade IV). Two (25%) patients were readmitted at 3 weeks post-device implant: one for abdominal pain requiring intravenous analgesia for 24 h (abdominal CT without pathologic findings) which was identified as constipation and the other one with diabetic ketoacidosis due to poor medical therapy compliance. This patient required intravenous insulin therapy and was hospitalized for 3 days (Clavien-Dindo grade IV because the patient was admitted to the ICU for 24 h for strict blood glucose monitoring).

No anastomotic bleeding or leaks were observed in any of the patients. Magnets were expelled in the patients’ stool at a median of 29.5 days (range: 15–45) after the procedure. Mild abdominal pain was present in 3 (35.5%) and 1 (12.5%) patients at POD 30 and 60, respectively. Nausea was present in 4 (50%) and no patients at POD 30 and 60, respectively. Diarrhea affected 2 (25%) and 1 (12.5%) patients at POD 30 and 60, respectively (Table 2).

Table 2 Postoperative adverse events
Full size table

Upper endoscopy at 12 months confirmed patent anastomoses with healthy-appearing mucosa in all patients (Fig. 4). HbA1c (baseline: 7.4–9.6) was reduced below 7.0% in 6 out of 8 (75%) patients, and greater than 5% of total body weight loss was observed in 7 out of 8 (87.5%) patients at 12 months.

Fig. 4
figure 4

Upper endoscopy at 12 months confirming patency of the anastomosis and healthy-appearing mucosa

Full size image

Discussion

The primary aim of this study was to evaluate technical feasibility and safety of sutureless duodeno-ileal anastomosis with self-assembling magnets in obese patients with type II diabetes. We found that this novel metabolic procedure was feasible and safe, providing a durable intestinal diversion.

Our study group has previously demonstrated proof of concept of self-assembling magnets for entero-enteral anastomosis in both animal and human studies, showing critical features in the resulting anastomoses: absence of bleeding, absence of leaks, absence of residual foreign (i.e., sutures/staples) material, and long-term patency [19,20,21]. In these studies, however, a purely endoscopic procedure was performed with simultaneous enteroscopy and colonoscopy for jejunal-ileal diversion [19,20,21]. Although this endoluminal incisionless approach was indeed promising, we learned that the need for two highly trained endoscopists and the potential inadvertent capture of adjacent structures might limit the embracement of the procedure. Therefore, we decided to shift towards a laparo-endoscopic technique in the current trial focusing on a duodenal-ileal diversion, and we found that this approach is technically simpler, faster, and most likely safer. In all patients, the laparoscopic and endoscopic magnet delivery was uneventful, and the engagement of the magnets was rapidly accomplished under direct visualization.

A new intestinal bypass procedure was introduced in 2007; the single anastomosis duodenal-ileal (SADI) bypass procedure which provides good weight loss outcomes has less metabolic complications as the biliopancreatic diversion with duodenal switch or Roux-en-Y gastric bypass and is rarely associated with “dumping syndrome” because the procedure preserves the pylorus [22,23,24,25]. However, the SADI procedure is technically challenging using a conventional laparoscopic hand-sewn anastomosis technique because the duodenum is a fixed retroperitoneal structure and the duodenal dissection for a SADI involves close proximity to the portal triad. In addition, this is also a difficult area in which we use surgical staplers to create an anastomosis. In fact, anastomosis leaks, strictures, abscess formation, and postoperative bleeding have been observed after SADI [26, 27]. These technical considerations and concerns over long-term malabsorptive complications are likely the rationale behind the limited number of surgeons adopting the SADI procedure as something that they routinely offer their patients. The proposed SNAP procedure offers an alternative to both concerns. Although our entero-enteral dual-path diversion SNAP procedure using self-assembling magnets partially resembles the SADI procedure, it differs in that there is no division of the duodenum, and some enteral flow through the duodenum is maintained. Therefore, the SNAP configuration resembles more the so-called transit bipartition operation [28, 29] (Fig. 5). As such, this procedure would offer less malabsorption and would likely be less effective than the SADI procedure. However, it also has significant potential advantages over the SADI including being much less technically challenging and potentially offering a lower risk of anastomotic leak or bleeding. The maintenance of duodenal enteral flow also potentially may lessen vitamin and mineral deficiencies while establishing new stimulation of distal K cells in the small intestine. In addition, such a procedure maintains access to the ampulla of the Vater biliary system for interventional gastrointestinal procedures if needed. Clearly, further clinical data are needed to determine the metabolic and relative complications of this procedure.

Fig. 5
figure 5

a SADI-S procedure; b transit bipartition procedure; c Sutureless Neodymium Anastomosis Procedure (SNAP); and d magnified view of duodeno-ileal side-to-side anastomosis with self-assembling magnets

Full size image

Some precautions should be followed due to the use of magnets. Intraoperatively, efforts should be made to minimize contact between the magnets and stainless-steel laparoscopic instruments, which may lead to procedure time delays. Postoperatively, patients cannot undergo magnetic resonance imaging until the expulsion of the magnets is confirmed. In our cohort, we had no intraoperative complications and no major postoperative morbidity. Although one patient was readmitted to the ICU due to diabetic ketoacidosis, this complication was most likely associated with poor patient compliance with prescribed medical therapy. In addition, the severity of the gastrointestinal adverse events observed in our trial was relatively mild and consistent with the altered anatomy. Most symptoms resolved upon diet correction, and only one patient persisted with diarrhea at the 2-month follow-up control.

Long-term patency of magnetic compression anastomosis is a common concern because small-diameter magnets are often passed endoscopically over a guidewire [30, 31]. For instance, a recent study analyzed the outcomes of 14 patients undergoing magnetic compression anastomosis for gastrointestinal obstruction. Anastomotic stenosis occurred in two (14%) patients, and one (7%) patient had an anastomotic perforation due to balloon dilation to prevent stenosis (perforation with generalized peritonitis after the 8th balloon dilation) [31]. In our cohort of patients, endoscopic evaluation of the duodenal bulb 12 months after the procedure showed a widely patent dual-path diversion. Our “flexible” magnets progress through a linear configuration to the final octagonal configuration, allowing for the creation of a large-caliber anastomosis. The absence of residual foreign material (i.e., staples or sutures) in the anastomosis avoids ongoing inflammation and fibrosis, which may potentially lead to anastomotic strictures. Additionally, the 360° compression force of the self-assembling magnets creates a complete hemostatic effect theoretically eliminating bleeding at the anastomosis site. All the patients included in this study had no prior bariatric procedure for weight loss, but the study group believes there may be a valuable use in patients that are regaining weight several years after a sleeve gastrectomy.

Metabolic surgery is evolving rapidly, and alternative endoscopic gastrointestinal bypass techniques might be also available in the near future. For instance, an anastomosis between the stomach and the intestinal limb using a transgastric approach with a two-channel endoscope with a novel stent was described in a porcine model [32]. Recently, an endoscopic ultrasonography-guided gastrojejunal anastomosis using a lumen-apposing metal stent was also reported in a patient with a duodenal stricture [33]. Overall, we believe that novel anastomotic techniques should be further explored and may become soon part of the armamentarium of metabolic surgery.

This study has several limitations. First, a small number of patients were enrolled in the trial. Second, all procedures were performed by the same surgical team, which limits the reproducibility of our results. Third, there was a lack of a control group. Finally, longer follow-up will be needed to demonstrate a more durable patency of the anastomosis and metabolic benefits.

Conclusions

Sutureless duodeno-ileal anastomosis using self-assembling magnets is feasible and safe in obese patients with type II diabetes, and a dual-path enteral diversion with large-caliber and durable anastomosis can be achieved. Potential additional applications may include patients that had previous sleeve gastrectomy with inadequate weight loss. Further studies with longer follow-up are needed to determine if this procedure can serve as an effective treatment modality for obese patients with type II diabetes.