Endoscopic Gastroenteric Anastomosis Using Magnets

• Introduction
• Patients and Methods
• Placement of Magnets
• Removal of Magnets and Placement of Stent
• Results
• Discussion
• Acknowledgment
• References

Background: Current management of malignant obstruction of the upper digestive tract includes surgical gastrointestinal bypass or endoscopic insertion of self-expandable metal stents. The safety, efficacy, and long-term patency rates of anastomoses created using the novel technique of endoscopic gastroenteric anastomosis using magnets (EGAM) are evaluated in this study.
Patients and Methods: 15 patients (13 men, 2 women; mean age 64.5 years) with malignant obstruction, who underwent EGAM and had monthly follow-up between December 2001 and May 2003, were included in this study.Results: The procedure was successful in 13 patients (88.66 %). The mean survival was 5.23 months. There were four minor complications (30.76 %) during the follow-up period.
Conclusion: Our results demonstrate the feasibility, safety. and efficacy of this technique for creating a gastroenteric anastomosis. The success rate was 86.6 %, there were no immediate complications, and there was no mortality related to the procedure.

Introduction

The use of magnets for creating anastomoses without the use of sutures was described by Murphy 100 years ago [1]. The same principle is used today for making anastomoses between the stomach and the small bowel [2] [3]. The first experience in this area was a successful application in an animal model, with minimal complications, by Cope in 1995 [4]. The purpose of the present pilot study was to evaluate the safety, efficacy, and long-term patency rates of anastomoses created using the novel nonsurgical technique of endoscopic gastroenteric anastomosis using magnets (EGAM), in patients with malignant gastrointestinal obstruction of the duodenum (gastric outlet obstruction) [5].

Patients and Methods

Between December 2001 and May 2003, 15 patients with malignant biliary and duodenal obstruction caused by pancreatic, duodenal, and other cancers were included in this study (see Table [1]). Surgery was considered to be inappropriate in all of these patients because of their advanced disease and poor status according to the American Joint Committee on Cancer (AJCC) classification.

Informed consent was obtained from all patients, and the protocol was approved by the hospital ethical committee.

As originally described by Cope et al. [6], magnetic gastroenteric compression was obtained using coated rare-earth magnets with an attractive force of 1000 gm. The diameters of the gastric and duodenal magnets were 14 and 12 mm respectively. The two magnets were attracted to each other and mated spontaneously over a separation distance of 2 -5 mm (Figure [1]).

For stenting we used a prototype ”yo-yo”-shaped 45-mm covered stent, consisting of a central stainless steel Z stent with wide conical extensions to minimize the potential for stent migration. The stent has a lumen 12 mm in diameter and is covered except for the distal prongs of the duodenal cone extension (Wilson-Cook) (Figure [2]). It can be loaded into a 30-Fr esophageal stent introduction sheath over a spring-tip guide wire and tapered obturator.

Placement of Magnets

All the patients presented as outpatients, with vomiting, weight loss, and jaundice. Olympus GIF-Q 145 video endoscopes (Olympus, Tokyo, Japan) were used in all procedures. The patients were given premedication of midazolam and fentanyl. All patients were admitted for 24-h observation following the procedure.

The patients underwent biliary drainage, either by endoscopic retrograde or percutaneous transhepatic cholangiography, before the EGAM procedure. Duodenal strictures were dilated to 15 - 18 mm using an Eclipse balloon (Wilson-Cook Inc., Winston-Salem, North Carolina, USA) (Figure [3 a]). A Hybrid guide wire (4.8 m, 0.035 inch; Wilson-Cook) was passed through the endoscope into the jejunum. The distal 12-mm magnet was then pushed under fluoroscopic guidance to the distal duodenum (D2) with a 10-Fr catheter that was advanced over the guide wire (Figure [3 b]). The 14-mm gastric magnet was placed in the stomach using the same technique, and it was then maneuvered using the endoscope and a foreign-body forceps. Sufficient distension with air was applied to cause the body of the stomach to overlap the duodenal-jejunal junction, and the gastric magnet was maneuvered under fluoroscopic and endoscopic guidance so that it mated transmurally with the duodenal magnet (Figure [3 c]). Endoscopic clips (HX-3/4, MD-850; Olympus) were applied to the gastric mucosa around the magnet to aid subsequent location of the fistula. Tattooing with India ink can be used, in addition to clipping, in order to identify the site of the anastomosis should stenosis occur.

Removal of Magnets and Placement of Stent

The coupled magnet buttons were withdrawn 7 - 10 days after placement. An extractor magnet was introduced endoscopically and held by a foreign-body forceps. This third magnet was coupled to the gastric magnet, and then all three magnets were withdrawn under endoscopic and radiologic control (Figure [3 d]).

The fistula was located endoscopically, catheterized, and opacified with contrast medium under fluoroscopy to ensure that there was no anastomotic leakage. A Savary guide wire was then passed through the anastomosis into the distal jejunum (Figure [3 e]). The yo-yo stent was loaded over the guide wire within the distal introduction sheath with the olive tip in place.

The sheath-stent assembly was then advanced under endoscopic and fluoroscopic guidance until the mid-section of the stent was in the fistula. This section was kept in position with a pusher tube while the sheath was slowly withdrawn to allow the distal stent cone to fully deploy within the duodenum. While the duodenal cone was kept in contact with the distal end of the fistula, the mid-section of the stent was deployed within the anastomotic channel and the proximal cone was opened within the gastric lumen (Figure [3 f]). The placement of the magnets and the delivery of the yo-yo stent took no more than 45 minutes.

The patients received a soft diet for 24 hours and were then allowed a normal diet.

Follow-up was done at 1-month intervals, by endoscopy, radiology, evaluation of symptoms and nutritional status, and by rating according to the Karnofsky scale.

Results

Between December 2001 and May 2003, 15 patients (13 men, two women; mean age 64.5 years, range 49 - 82) underwent EGAM. The causes of obstruction of the distal gastric and duodenal tract were pancreatic carcinoma with duodenal infiltration in 12 patients, cholangiocarcinoma in two patients, and adenocarcinoma of the gastric antrum in one.

The procedure was successful in 13 patients (88.66 %; 12 men, one woman) and failed in two (13.33 %). The failures were due to technical limitations in one instance and a complication in the other. In the first patient it was not possible to dilate a duodenal stricture. In the second patient, a perforation occurred as a result of manipulation of the recently formed and likely immature fistula, and immediate surgical intervention and gastrointestinal bypass was required.

To date, of the 13 patients who were successfully treated, 12 (92.3 %) have died because of the underlying malignancy. Their mean survival was 5.23 months (range 1 - 10 months). The patient who is alive at 10 months continues to take solid food. All the patients maintained their Karnofsky scores, except for one whose score improved by 2 points, and in all the patients oral intake of solids was maintained until their deaths.

There were four minor complications (30.76 %) during the follow-up period: two instances of distal migration of the stent (the patients found them in their stools); one proximal migration; and one obstruction of the stent by solid food. The latter two complications were resolved endoscopically.

Discussion

Obstruction of the upper digestive tract is a late complication of unresectable biliopancreatic, duodenal, and gastric prepyloric malignant neoplasms [7]. Currently, management includes surgical gastrointestinal bypass (open or laparoscopic gastroenteric bypass) or endoscopic insertion of a self-expandable metal stent (SEMS).

Surgery has major shortcomings including high morbidity and mortality, due in part to advanced stage of disease, the patients’ poor general condition, and the long hospital stay after surgery. We believe surgery is a valid option for patients with good performance status and offers the possibility of curative resection, and who present intraoperative criteria for nonresectability, such as vascular encasement, that were not detected in preoperative investigations [8].

Many series in the literature have evaluated the results of palliative surgical bypass [9] [10], which can be performed either prophylactically or in symptomatic patients [11] [12], reporting a morbidity of 30 %, a mortality of 15 % [8] [13] [14], and long hospitalizations with a mean hospital stay of at least 2 weeks [14].

The second option is the endoscopic insertion of a SEMS [15]. At least 25 studies have been published on the safety and efficacy of SEMS for palliation in patients with malignant gastric outlet, duodenal and small-bowel obstruction [16] [17] [18]. Reported complications of enteral stent insertion include ulceration caused by the stent wires, bleeding, migration, occlusion, and perforation [17] [18] [19]. The mean length of hospitalization for those who underwent enteral stent placement was 4 days [20].

In this study we present a third option, that is the creation of a gastroenteric anastomosis endoscopically by means of magnetic compression. This procedure was performed by an interventional gastroenterologist in an interventional radiology suite.

Our results demonstrate the feasibility, safety, and efficacy of EGAM, with an 86.6 % success rate, no immediate complications, and no mortality related to the procedure. In addition, all patients were able to resume a normal intake of solids 48 hours after the procedure. On long-term follow-up, all the patients maintained their pre-procedure Karnofsky score except for one patient whose score improved by two points, and all were able to ingest solid food by mouth until death.

In summary, we present a new option for palliation of malignant obstruction of the upper digestive tract, that has low morbidity and mortality, requires only a short hospital stay, and permits a good quality of life for these patients. More studies are required, involving a large number of patients and a longer follow-up, to confirm our preliminary results.

Acknowledgment

The investigators would like to thank Irving Waxman, M. D., of the University of Chicago for his critical review of this manuscript.

Support for this study was provided by Wilson-Cook Inc.

References

• 1 Murphy J B. Cholecysto-intestinal anastomosis, and approximation without sutures (original research).  Med Rec NY. 1892;  42 665-676
  PubMedSearch in Google Scholar
• 2 Forde K A, McLarty A J, Tsai J. et al . Murphy’s button revisited: clinical experience with the biofragmentable anastomotic ring.  Ann Surg. 1993;  217 78-81
  PubMedSearch in Google Scholar
• 3 Saveliev V S, Avaliani M V, Bashirov A D. Endoscopic magnetic cholecystodigestive anastomoses: personal technique for palliative treatment of distal bile duct obstruction.  J Laparoendosc Surg. 1993;  3 99-112
  PubMedSearch in Google Scholar
• 4 Cope C. Creation of compression gastroenterostomy by means of the oral, percutaneous, or surgical introduction of magnets: feasibility study in swine.  J Vasc Interv Radiol. 1995;  6 539-545
  CrossrefPubMedSearch in Google Scholar
• 5 Cope C, Ginsberg G G. Long-term patency of experimental magnetic compression gastroenteric anastomoses achieved with covered stents.  Gastrointest Endosc. 2001;  53 780-784
  CrossrefPubMedSearch in Google Scholar
• 6 Cope C, Clark T WI, Ginsberg G, Habecker P. Stent placement of gastroenteric anastomosis formed by magnetic compression.  J Vasc Interv Radiol. 1999;  10 1379-1386
  CrossrefPubMedSearch in Google Scholar
• 7 Lillemoe K D, Pitt H A. Palliation, surgical and otherwise.  Cancer. 1996;  78 605-614
  CrossrefPubMedSearch in Google Scholar
• 8 Nasif T, Prat F. et al . Endoscopic palliation of malignant gastric outlet obstruction using SEMS.  Endoscopy. 2003;  35 483-489
  Thieme ConnectPubMedSearch in Google Scholar
• 9 Di Fronzo L A, Cymerman J. et al . Unresectable pancreatic carcinoma: correlating length of survival with choice of palliative bypass.  Am Surg. 1999;  65 955-958
  PubMedSearch in Google Scholar
• 10 Deziel D J, Wilhelmi B. et al . Surgical palliation for ductal adenocarcinoma of the pancreas.  Am Surg. 1996;  62 582-588
  PubMedSearch in Google Scholar
• 11 Van Vangensveld B A, Coene P PLO. et al . Outcome of palliative biliary and gastric bypass surgery for pancreatic head carcinoma in 126 patients.  Br J Surg. 1997;  84 1402-1406
  CrossrefPubMedSearch in Google Scholar
• 12 Watanapa P, Williamson R CN. Surgical palliation for pancreatic cancer: developments during the past two decades.  Br J Surg. 1992;  79 8-20
  CrossrefPubMedSearch in Google Scholar
• 13 Nagy A, Brosseuk D. et al . Laparoscopic gastroenterostomy for duodenal obstruction.  Am J Surg. 1995;  169 539-542
  CrossrefPubMedSearch in Google Scholar
• 14 Lillemoe K D, Cameron J N. et al . Is prophylactic gastrojejunostomy indicated for unresectable periampullary cancer? A prospective randomised trial.  Ann Surg. 1999;  230 322-328
  CrossrefPubMedSearch in Google Scholar
• 15 Carr-Locke D L. Role of endoscopic stenting in the duodenum.  Ann Oncol. 1999;  10 (Suppl 4) 261-264
  PubMedSearch in Google Scholar
• 16 Bethge N, Breitkreutz C, Vakil N. Metal stents for the palliation of inoperable upper gastrointestinal stenoses.  Am J Gastroenterol. 1998;  93 643-645
  CrossrefPubMedSearch in Google Scholar
• 17 Soetikno R M, Lichtenstein D R, Vandervoort J. et al . Palliation of malignant gastric outlet obstruction using an endoscopically placed Wallstent.  Gastrointest Endosc. 1998;  47 267-270
  PubMedSearch in Google Scholar
• 18 Yates M R, Morgan D E, Baron T H. Palliation of malignant gastric and small intestinal strictures with self-expandable metal stents.  Endoscopy. 1998;  30 266-272
  PubMedSearch in Google Scholar
• 19 De Baere T, Harry G, Ducreux M. et al . Self-expanding metallic stents as palliative treatment of malignant gastroduodenal stenosis.  AJR Am J Roentgenol. 1997;  169 1079-1083
  PubMedSearch in Google Scholar
• 20 Yim H B, Jacobson B C, Saltzman J R. et al . Clinical outcome of the use of enteral stents for palliation of patients with malignant upper GI obstruction.  Gastrointest Endosc. 2001;  53 329-332
  CrossrefPubMedSearch in Google Scholar
• 21 Baron T H, Harewood G C. Enteral self-expandable stents.  Gastrointest Endosc. 2003;  58 421-433
  CrossrefPubMedSearch in Google Scholar

N. A. Chopita, M. D., Ph. D.

Department of Gastroenterology, San Martin Hospital

Calle 2, no. 76 · 1900 La Plata · Argentina

Fax: +54-221-4225111

Email: chopita@netverk.com.ar

References

• 1 Murphy J B. Cholecysto-intestinal anastomosis, and approximation without sutures (original research).  Med Rec NY. 1892;  42 665-676
  PubMedSearch in Google Scholar
• 2 Forde K A, McLarty A J, Tsai J. et al . Murphy’s button revisited: clinical experience with the biofragmentable anastomotic ring.  Ann Surg. 1993;  217 78-81
  PubMedSearch in Google Scholar
• 3 Saveliev V S, Avaliani M V, Bashirov A D. Endoscopic magnetic cholecystodigestive anastomoses: personal technique for palliative treatment of distal bile duct obstruction.  J Laparoendosc Surg. 1993;  3 99-112
  PubMedSearch in Google Scholar
• 4 Cope C. Creation of compression gastroenterostomy by means of the oral, percutaneous, or surgical introduction of magnets: feasibility study in swine.  J Vasc Interv Radiol. 1995;  6 539-545
  CrossrefPubMedSearch in Google Scholar
• 5 Cope C, Ginsberg G G. Long-term patency of experimental magnetic compression gastroenteric anastomoses achieved with covered stents.  Gastrointest Endosc. 2001;  53 780-784
  CrossrefPubMedSearch in Google Scholar
• 6 Cope C, Clark T WI, Ginsberg G, Habecker P. Stent placement of gastroenteric anastomosis formed by magnetic compression.  J Vasc Interv Radiol. 1999;  10 1379-1386
  CrossrefPubMedSearch in Google Scholar
• 7 Lillemoe K D, Pitt H A. Palliation, surgical and otherwise.  Cancer. 1996;  78 605-614
  CrossrefPubMedSearch in Google Scholar
• 8 Nasif T, Prat F. et al . Endoscopic palliation of malignant gastric outlet obstruction using SEMS.  Endoscopy. 2003;  35 483-489
  Thieme ConnectPubMedSearch in Google Scholar
• 9 Di Fronzo L A, Cymerman J. et al . Unresectable pancreatic carcinoma: correlating length of survival with choice of palliative bypass.  Am Surg. 1999;  65 955-958
  PubMedSearch in Google Scholar
• 10 Deziel D J, Wilhelmi B. et al . Surgical palliation for ductal adenocarcinoma of the pancreas.  Am Surg. 1996;  62 582-588
  PubMedSearch in Google Scholar
• 11 Van Vangensveld B A, Coene P PLO. et al . Outcome of palliative biliary and gastric bypass surgery for pancreatic head carcinoma in 126 patients.  Br J Surg. 1997;  84 1402-1406
  CrossrefPubMedSearch in Google Scholar
• 12 Watanapa P, Williamson R CN. Surgical palliation for pancreatic cancer: developments during the past two decades.  Br J Surg. 1992;  79 8-20
  CrossrefPubMedSearch in Google Scholar
• 13 Nagy A, Brosseuk D. et al . Laparoscopic gastroenterostomy for duodenal obstruction.  Am J Surg. 1995;  169 539-542
  CrossrefPubMedSearch in Google Scholar
• 14 Lillemoe K D, Cameron J N. et al . Is prophylactic gastrojejunostomy indicated for unresectable periampullary cancer? A prospective randomised trial.  Ann Surg. 1999;  230 322-328
  CrossrefPubMedSearch in Google Scholar
• 15 Carr-Locke D L. Role of endoscopic stenting in the duodenum.  Ann Oncol. 1999;  10 (Suppl 4) 261-264
  PubMedSearch in Google Scholar
• 16 Bethge N, Breitkreutz C, Vakil N. Metal stents for the palliation of inoperable upper gastrointestinal stenoses.  Am J Gastroenterol. 1998;  93 643-645
  CrossrefPubMedSearch in Google Scholar
• 17 Soetikno R M, Lichtenstein D R, Vandervoort J. et al . Palliation of malignant gastric outlet obstruction using an endoscopically placed Wallstent.  Gastrointest Endosc. 1998;  47 267-270
  PubMedSearch in Google Scholar
• 18 Yates M R, Morgan D E, Baron T H. Palliation of malignant gastric and small intestinal strictures with self-expandable metal stents.  Endoscopy. 1998;  30 266-272
  PubMedSearch in Google Scholar
• 19 De Baere T, Harry G, Ducreux M. et al . Self-expanding metallic stents as palliative treatment of malignant gastroduodenal stenosis.  AJR Am J Roentgenol. 1997;  169 1079-1083
  PubMedSearch in Google Scholar
• 20 Yim H B, Jacobson B C, Saltzman J R. et al . Clinical outcome of the use of enteral stents for palliation of patients with malignant upper GI obstruction.  Gastrointest Endosc. 2001;  53 329-332
  CrossrefPubMedSearch in Google Scholar
• 21 Baron T H, Harewood G C. Enteral self-expandable stents.  Gastrointest Endosc. 2003;  58 421-433
  CrossrefPubMedSearch in Google Scholar

N. A. Chopita, M. D., Ph. D.

Department of Gastroenterology, San Martin Hospital

Calle 2, no. 76 · 1900 La Plata · Argentina

Fax: +54-221-4225111

Email: chopita@netverk.com.ar