Accepted Article
Title page
Magnetic Compression Anastomosis for Treatment of Benign Biliary Strictures1 Sung Ill Jang1 MD, Jungran Choi2 MD, Dong Ki Lee2 MD, PhD
1
Department of Internal Medicine, Kangnam Sacred Heart Hospital Hallym University College of Medicine, Seoul, Korea 2
Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea
Correspondence: Dong Ki Lee MD, PhD Address: Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, 211 Eonju-Ro, Gangnam-Gu, Seoul, Korea Phone: 82-2-2019-3214 This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/den.12319 This article is protected by copyright. All rights reserved.
Accepted Article
Fax: 82-2-3463-3882 E-mail: [email protected] Article type: Review Received date: 25-Apr-2014 Accepted date: 30-May-2014 Abstract
Endoscopic and percutaneous procedures have shown high success rates when used to
treat benign biliary strictures (BBS). However, cases in which a guidewire cannot be passed through a refractory stricture or a complete obstruction are difficult to treat using conventional methods. Magnetic compression anastomosis (MCA) has emerged as a nonsurgical alternative avoiding operational mortality and morbidity. The feasibility and safety of MCA have been experimentally and clinically verified in cases of biliobiliary and bilioenteric anastomosis. However, no pre-MCA assessment modality capable of predicting outcomes is yet available, and no universally effective magnet delivery method has yet been established, rendering it difficult to identify patients for whom MCA is appropriate. Various experimental studies seeking to overcome these limitations are underway. Such work will improve our indepth understanding of MCA, which has been trialed in various fields. Upon further development, MCA may become a groundbreaking option for treatment of benign strictures that are difficult to resolve using conventional methods, and MCA may be expected to be minimally traumatic and highly effective. The aim of the present study was to discuss the current status of MCA and the direction of MCA development by reviewing clinical and experimental MCA data. Key words: Magnetic compression anastomosis, compressive anastomosis, benign biliary This article is protected by copyright. All rights reserved.
Accepted Article
stricture, biliary obstruction, anastomosis stricture.
This article is protected by copyright. All rights reserved.
Accepted Article
Introduction Advances in non-surgical methods, including endoscopic or percutaneous approaches,
have enabled recanalization of severe biliary strictures, benign or malignant, and strictures that develop postoperatively.1-6 However, non-surgical methods may be of limited efficacy when used to treat severe biliary strictures or complete obstructions; placement and maintenance of drainage catheters are required in patients who fail recanalization using nonsurgical methods. Recently, magnetic compression anastomosis (MCA) has been developed as a non-surgical alternative treatment for patients in whom a conventional endoscopic or percutaneous method has failed (Fig. 1).7-14 The clinical feasibility, safety, and utility of MCA have been established in various cases involving stenosis and occlusions, without any need for surgery.15-17 This review focuses on novel applications of MCA for the biliobiliary and bilioenteric anastomosis of stenosis and occlusions.
History
The concept of compression anastomosis was first proposed by Denan in 1826, who
described a newly formed non-sutureless anastomotic fistula caused by ischemic compression of tissue.18 Denan’s spring-loaded device was revised and further developed by Murphy in 1892 and became known as Murphy’s button.19-22 Use of the technique enables formation of a circular gastrointestinal anastomosis via ischemic compression of tissue held between two buttons by a spring, and was the first surgical device used to prepare functioning end-to-end and side-to-side sutureless anastomoses via ischemic compression.23 In 1991, an animal study used compression buttons and modified Murphy’s buttons to perform gastrojejunostomy, aided by endoscopy.24 The mechanical device forms a compression anastomosis via contact of screws, snaps, or springs of the two components. Such physical contact can be replaced by magnetic attraction, using a field-mediated magnetic force. The effects of transluminal
This article is protected by copyright. All rights reserved.
Accepted Article
magnetic force transduction via both a fistula and a free perforation have been analyzed in several children who had ingested magnets.25-28 Jansen et al.29 performed the first human experiments in 1980; tissue compression was achieved via magnetic attraction achieved using mechanical compression. Mucosa-to-mucosa anastomosis using MCA was successfully performed in five patients with colonic resections. Subsequently, in 1993, Salvelev et al.30 performed clinical and laboratory studies on mongrel dogs. Successful cholecystoenteric, enteroenteric, and magnetic cholecystogastric anastomoses were formed. In addition, preliminary data from four patients who underwent cholecystogastric anastomoses and one who underwent a cholecystoduodenal anastomosis have demonstrated the feasibility of endoscopic magnetic cholecystodigestive anastomosis.31 Together, the work to date supports both the concept and the clinical feasibility of compression anastomosis using magnets.32 In 1998, Yamanouchi33 was the first to use a modern MCA method in a clinical trial, successfully generating a bile duct-small intestinal fistula and introducing novel uses of MCA. Other clinical outcomes have also been reported.6-14,34-40 Magnets
Magnetic power is critical contributor if MCA is to successful. Rare-earth magnets are
classified into neodymium iron-boron magnets and samarium-cobalt (Sm-Co) magnets. Both have high magnetic flux densities and retention forces, rendering them suitable for MCA. However, the retention force of a Sm-Co magnet is stronger than that of a neodymium ironboron magnet, and the former type of magnet is thus more frequently used. 7-10,34,41 To quantify magnetic power, the magnetic powers of magnets used in published studies were calculated using a magnetic force determination algorithm (MAGDA).31,41 The work suggested that such calculations helped to predict the success of MCA. Variables including magnet shape, dimensions, the nature of the magnetic material, the magnetic grade or
This article is protected by copyright. All rights reserved.
Accepted Article
strength, and experimentally derived or estimated in vivo magnetic separation forces, may be entered into the MAGDA. Animal studies In the 1990s, efforts were made to induce compression anastomosis via magnetic
attraction between powerful rare-earth magnets. In 1995, Cope demonstrated the feasibility and safety of MCA by creating bilioenteric and enteroenteric anastomoses in pigs.15,16 Cope used neodymium-iron-boron or rare-earth Sm-Co magnets to perform cholecystogastric and cholecystojejunal MCA in swine, and a bilioenteric anastomosis formed via magnet approximation after 9-16 weeks.15 A preliminary study revealed that magnets could be used to create leak-free short-term enteroenteric anastomoses in swine.16 The shape of magnets used
in subsequent MCA experiments was modified to amplify the magnetic effects, and further animal studies were performed. Jamshidi et al.19 performed MCA using both uniform and gradient compression methods, and compared stapled anastomoses to those that had been additionally hand-sutured. Also, all of gross appearance, histology, functionality (assessed radiologically), and mechanical integrity were evaluated. Although one stapled anastomosis leaked, no magnetic anastomosis did so. No severe complication and no appreciable stenosis were observed. The burst pressures of MCA-formed and surgically formed anastomoses did not differ. Upon pathological examination, MCA-formed anastomoses evidenced continuity of the serosal, submucosal, and mucosal layers, and neither ischemia nor necrosis was present. Thus, MCA was safe, and equivalent or superior to anastomoses created by traditional suturing or stapling techniques.19 Further, the same team showed that MCA-assisted enteroenterostomy was feasible, using modified magnets in the form of two convex-concave radially symmetrical halves.42 The design and development of a controlled MCA system (magnamosis) that optimizes all of magnetic coupling, the distance between magnets, and This article is protected by copyright. All rights reserved.
Accepted Article
mating of surface geometries, are required to achieve reliable enteric anastomosis.19,42,43 A magnamosis device has the following three main features: (1) two convex-concave radially symmetric rings that self-align magnetically; (2) a ring-shaped magnet that allows immediate patency; and (3) a mating surface with a radial topography specifically engineered to promote necrosis at the center and healing at the periphery; this ensures that the anastomosis formed is free of any perforation. Apart from modifications to magnets, animal studies have been conducted to optimize
endoscopic magnet delivery.44 A modular flexible magnetic anastomotic device has been developed, the use of which ensured absence of leakage. One study used partially covered stents to improve both the modular shapes of the magnets and the patency of an MCA-formed fistula.45 When partially covered stents were inserted into MCA-formed gastroenteric anastomoses, patency was maintained for more than 7 weeks. Compression anastomosis using magnets has been experimentally attempted in all of the vascular, biliary, and gastroenterological tracts.46 Studies in humans Materials and Methods
Bilio-biliary anastomosis The magnets used in MCA are cylindrical Sm-Co rare-earth magnets, of various powers,
which can be delivered by a variety of methods.41 However, the most common delivery route is via the percutaneous and peroral bile ducts. The MCA procedure may be divided into the four following steps (Figs. 2, 3): (1) tract formation for magnet delivery; (2) magnet approximation; (3) magnet removal; and, (4) maintenance and removal of the internal catheter.
This article is protected by copyright. All rights reserved.
Accepted Article
(1) Tract formation for magnet delivery: A percutaneous transhepatic biliary drainage (PTBD) tract for magnet delivery is formed using an 18-Fr PTBD catheter. The PTBD catheter is exchanged for an 18-Fr sheath prior to MCA approximation, to allow convenient magnet insertion via the PTBD tract and to reduce duct injury. In the common bile duct (CBD) tract, full endoscopic sphincterotomy (EST) and balloon dilation or transient insertion of a retrievable, fully converted, self-expandable metal stent (FCSEM), are used to facilitate magnet delivery.
(2) Magnet approximation: A thread attached to one magnet is fixed to a polypectomy snare, and the magnet is moved to the anastomosis site via PTBD tract. The polypectomy snare is passed through the channel of an endoscopic retrograde cholangiopancreatography (ERCP) scope, and the other magnet is fixed in front of the scope. The magnet is moved to the anastomosis site through the FCSEM, and MCA approximation occurs via attraction between the two magnets. To better approximate the magnets, a balloon catheter may be used to advance the magnets through both the PTBD and ERCP tracts. Approximation of the two magnets is confirmed radiographically. Next, the long sheath tube is removed and the indwelling PTBD catheter inserted. The FCSEM inserted in the CBD is removed immediately after magnet approximation.
(3) Magnet removal: When a fistula forms because of ischemic necrosis caused by the approximated magnets, the magnets spontaneously migrate to the duodenum. However, if spontaneous migration does not occur after ~8 to 10 weeks, the magnets can be pushed out using a guidewire or catheter. The magnets can also be removed through the PTBD tract via percutaneous transhepatic cholangioscopy (PTCS).
(4) Maintenance and removal of the internal catheter: After magnet removal, a 14-16-Fr
This article is protected by copyright. All rights reserved.
Accepted Article
internal catheter is inserted into the fistula and maintained in place for 4-6 months to reduce the probability of re-stenosis of the fistulous tract. A FCSEM is more useful than an internal catheter, because the FCSEM allows formation of a larger fistula. However, a FCSEM is more likely to migrate than an indwelling catheter. Further work is necessary to confirm the efficacy of FCSEMs.
Bilio-enteric anastomosis In general, the MCA methods and concepts used in bilioenteric anastomosis are similar to
those of biliobiliary anastomosis. However the delivery routes of magnets differ (Fig. 4). Several different routes including a surgically formed percutaneous-jejunum tract;8,35 a
percutaneous-percutaneous tract;8 or a percutaneous-peroral tract (the most common approach), may be used to deliver magnets. The method of magnet delivery to the percutaneous tract is identical to that described above, but a forward-viewing endoscope is used in any peroral approach. However, an endoscopic approach is difficult in patients with long afferent loops. In such cases, magnets may be delivered via a surgically created skin/intestinal fistula (Fig. 5A).8,35 One study has suggested that single-balloon enteroscopy is useful in such situations.40 Rarely, the left intrahepatic duct (IHD) and the right IHD may be separately anastomosed to the jejunum, with the stricture evident on the right side (Fig. 5B). Two PTCS scopes are used in this context; one to deliver a magnet through the right IHD via the PTBD tract, and the other to approximate the second magnet at the stricture site via the left IHD tract.8 Results
Bilio-biliary anastomosis Recent reports on human MCA have described 22 MCA biliobiliary anastomoses (in 13
This article is protected by copyright. All rights reserved.
Accepted Article
males and 9 females) (Table 1). The most common causes of benign biliary stricture requiring MCA were postoperative complications, especially after living-donor liver transplantation (17 cases). In addition to those occurring post-operatively, some strictures were caused by ischemic injury by radiation and embolization. The mean distance between the two magnets was 6.4 mm (range, 2–15 mm), and the mean time to magnet removal 53.3 days (range, 9– 181 days). During the follow-up period (mean, 12 months; range, 2-36 months), re-stenosis occurred in only a single case who was re-cannulated using a conventional method. No obstructive jaundice or related symptoms was/were found in the remaining 21 cases. Bilio-enteric anastomosis Forty-two bilioenteric MCA procedures have been performed in patients with 8 benign and
34 malignant biliary strictures (Table 2). The causes of benign biliary stricture (BBB) were post-operative complications, and the principal anastomosis method used was Roux-en-Y
reconstruction. Magnets were delivered via an endoscope or balloon endoscope in patients undergoing Roux-en-Y reconstruction, but if the anastomosis site was not attained, magnet delivery was performed via a surgically created skin fistula.8,35 In addition, two PTBD tracts
were used for magnet delivery to stricture sites.8 These cases highlight the various delivery
methods that may be employed. The mean distance between the two magnets was 4 mm (range, 2–7 mm), and the time to magnet removal 7–40 days. Of the 42 subjects, 41 (97.6%) developed complete anastomoses and no MCA-related complications were noted. During the follow-up period (mean, 40 months; range, 2–53 months), re-stenosis occurred in a single
case 6 days following removal of the drainage tube, but re-cannulation after balloon dilation was performed without difficulty.35 Discussion
This article is protected by copyright. All rights reserved.
Accepted Article
BBB can be caused by infection, inflammation, an autoimmune disease, ischemic injury,
trauma, and intra- or post-operative complications.47 The feasibility and safety of MCA for a BBB that cannot be recanalized using a conventional method continue to be verified in various experimental and clinical studies.7-10,12,34,40 MCA-associated patency is clinically similar to that of an anastomosis formed by conventional surgery.17 In an animal model, the anastomosis site exhibited continuity of the serosal, submucosal, and mucosal layers. Ischemia and necrosis were absent, and the burst strengths of anastomosis formed by MCA was equaled or exceeded that of a surgical approach.42,44 Doppler ultrasound-based screening and follow-up are often performed because of the
possibility of rupture during MCA if blood vessels are placed between two magnets. 34,35 However, no blood vessel rupture or other complication has yet been reported in any clinical trial. This is believed to be because fistula formation after MCA requires a relatively long time. Use of two magnets renders the closing gradual. Thus, even if an intervening vessel is present, neither compression nor rupture is to be expected. In previous studies, the mean time to magnet removal after successful approximation was 53.3 days (range, 9–181 days) in cases
of biliobiliary anastomosis and 7–40 days in cases of bilioenteric anastomosis; these times are rather long. The time to magnet removal after successful approximation depends on the distance between the two magnets, the magnetic field strength, and the histological situation around the obstruction.39 Consequently, the relatively short duration required for bilioenteric anastomosis is explained by the fact that the distance between the two magnets (2–7 mm) is less than in biliobiliary anastomosis (2–15 mm). Typically, partial recanalization requires at
least 10 days for a short obstruction and ~1 month for a long obstruction.39 Currently, no long-term clinical follow-up data on new fistulas formed via MCA are
available. However, MCA forms fistulas via tissue necrosis, without dilation of fibrotic tissue. This article is protected by copyright. All rights reserved.
Accepted Article
Therefore, the risk of restenosis upon recoiling of fibrotic tissue should be low. In one case of biliobiliary anastomosis, no restenosis was reported after 3 years.38 In patients with biliary stenosis developing after living-donor liver transplantation, restenosis was observed in 1 of 12 subjects over 331 days, but recanalization was easily performed via PTBD.7 In studies on bilioenteric anastomosis, one case exhibited no restenosis for 50 months 8, and no recurrence was observed in malignant patients to 30 days after MCA.12 The suggested low recurrence rate after MCA will be validated in larger studies with long-term follow-up. Pre-MCA assessment is limited to predicting outcomes and planning delivery methods;
these issues need to be addressed. The success of MCA depends on several factors such as a length of the stricture, a shape of bile duct, a pattern of direction of each magnet and bile duct axis. The main causes in unsuccessful MCA cases are a long length of the stricture, a tapered or tortuous duct and/or the parallel axis of alignment.7,8 The magnetic force weakens as the
length of a stricture increases. In such circumstances, compressive necrosis will not occur, inhibiting fistula formation. Thus, accurate evaluation of stricture length prior to MCA is important. However, currently, the stricture length cannot be accurately evaluated using noninvasive imaging modalities such as computed tomography, ultrasonography, or magnetic resonance cholangiopancreatography. Duct evaluation via cholangiography is reasonably accurate, but has the disadvantage that an invasive procedure such as ERCP or PTBD must be performed. Apart from stricture length, the shape and axis of the bile duct are also important in terms of pre-MCA assessment. If the duct is tapered or tortuous (Fig. 6A), the magnets cannot maximally approach the stricture site despite a short stricture length, and the actual length between the two magnets is thus longer than measured during pre-MCA assessment, leading to MCA failure (Fig. 6B).7 Moreover, the bile duct axis determines the direction of magnet alignment. Even if the distance between magnets is short, as in most instances of
This article is protected by copyright. All rights reserved.
Accepted Article
successful MCA, the technique may still fail if the alignment is parallel; the magnetic power becomes weak.7,8 As no non-invasive method can accurately evaluate stricture length, duct
shape, or the duct axis, all of which are required to plan successful MCA, a limitation in identifying suitable MCA candidates becomes apparent. The result can be known only when magnets are actually applied. The magnet delivery method chosen depends on the type of anastomosis required, prior
operative history, and patient characteristics in both biliobiliary and bilioenteric anastomosis. Biliobiliary anastomosis delivers magnets via PTBD and ERCP. To prevent hepatic duct injury during magnet delivery through the PTBD tract, it is preferable to employ 16- or 18-Fr sheathes for insertion. Magnet delivery into the CBD is more difficult with ERCP than PTBD, because delivery through the papilla of Vater is challenging and may sometimes fail. Delivery of a 5-mm magnet is difficult via EST alone, and balloon dilation is often used, but manipulation of the magnet becomes difficult at times.7 To solve this problem, a metal stent can be transiently inserted into the papilla.7,36 The stent can be inserted on the day of magnet approximation, or 1 day prior, to minimize stent migration and pancreatitis. In other words, the stent indwelling time should be minimal. Generally, during Roux-en-Y bilioenteric anastomosis, magnet delivery to the stricture site via ERCP is difficult because of the long lengths of the efferent loop and afferent limbs, and the risk of intestinal perforation. In such cases, transparent cap-installed forward colonoscopy8 and balloon endoscopy48 can be helpful, but do not guarantee success in all cases, and the techniques are associated with a risk of intestinal perforation. Recently, MCA magnet delivery via surgically formed skin/intestinal fistulas, has been reported; this may be a useful alternative method.8,35 In summary, various magnet delivery methods have been developed by consideration of patient characteristics, prior operative history, and the type of anastomosis required at the stricture site. Further
This article is protected by copyright. All rights reserved.
Accepted Article
development of effective and safe delivery methods is needed. The feasibility and safety of biliobiliary and bilioenteric anastomoses created via MCA
have been verified in both human and animal studies. In addition, Avalinani et al.12 used MCA to form anastomoses between the normal bile duct and the duodenum or jejunum in 34 patients with malignant strictures, but not for recanalization of malignant obstruction. Re-intervention was required in six subjects.12 However, MCA is not normally indicated to treat malignant biliary obstructions, which are often treatable using conventional peroral or percutaneous methods. Entero-enteric fistulous tract formation via MCA has been successful in experimental animal models, and has been reported to be compatible with a surgically formed anastomosis.42,44 However, other than a small amount of work in Japan, no study has
yet sought to construct human enteroenteric anastomoses via MCA. Gastrojejunostomy via MCA might be more efficacious in the management of malignant gastric outlet obstructions than is the current use of bare self-expandable metal stents.45 In summary, MCA is expected
to have diverse clinical applications including formation of gastrojejunal and jejunojejunal anastomoses creating gastric bypasses in obese subjects; removal of short strictures; esophageal reconstruction; closure of ileostomies and colostomies; bypassing of malignant obstructions; and others.
Conclusions MCA is a non-surgical alternative for treatment of occluded BBS that are difficult to
resolve using conventional endoscopic or percutaneous methods, and is both feasible and safe. For effective MCA and successful recanalization, a pre-MCA assessment method predicting outcomes, smaller and more powerful magnets, and an effective magnet delivery system, need to be developed. Moreover, endoscopists should fully understand the mechanisms and principles of MCA and expand the clinical indications for MCA, thereby enabling application This article is protected by copyright. All rights reserved.
Accepted Article
and development of the technique in various fields. Despite the limited number of cases reported to date, MCA is effective, safe, has a low recurrence rate, and is less traumatic than other treatments for occluded BBS. Reference 1
Dumonceau JM, Deviere J, Delhaye M, Baize M, Cremer M. Plastic and metal stents
for postoperative benign bile duct strictures: the best and the worst. Gastrointest
Endosc. 1998; 47 (1): 8-17.
2
Bonnel DH, Liguory CL, Lefebvre JF, Cornud FE. Placement of metallic stents for
treatment of postoperative biliary strictures: long-term outcome in 25 patients. AJR
Am J Roentgenol. 1997; 169 (6): 1517-22.
3
Baba S, Yoshioka U, Sotome T et al. [Case of von Hippel-Lindau disease with
pancreatic exocrine dysfunction and benign biliary stenosis due to multiple pancreatic serous cystadenoma]. Nihon Naika Gakkai Zasshi. 1996; 85 (10): 1765-7.
4
Ernst O, Sergent G, Mizrahi D, Delemazure O, L'Hermine C. Biliary leaks: treatment
by means of percutaneous transhepatic biliary drainage. Radiology. 1999; 211 (2):
345-8.
5
Vogel SB, Howard RJ, Caridi J, Hawkins IF, Jr. Evaluation of percutaneous
transhepatic balloon dilatation of benign biliary strictures in high-risk patients. Am J
Surg. 1985; 149 (1): 73-9.
6
Mita A, Hashikura Y, Masuda Y et al. Nonsurgical policy for treatment of bilioenteric
anastomotic stricture after living donor liver transplantation. Transpl Int. 2008; 21 (4):
320-7.
7
Jang SI, Kim JH, Won JY et al. Magnetic compression anastomosis is useful in biliary
anastomotic strictures after living donor liver transplantation. Gastrointest Endosc.
This article is protected by copyright. All rights reserved.
Accepted Article
2011; 74 (5): 1040-8.
8
Jang SI, Rhee K, Kim H et al. Recanalization of refractory benign biliary stricture
using magnetic compression anastomosis. Endoscopy. 2014; 46 (1): 70-4.
9
Takao S, Matsuo Y, Shinchi H et al. Magnetic compression anastomosis for benign
obstruction of the common bile duct. Endoscopy. 2001; 33 (11): 988-90.
10
Itoi T, Yamanouchi E, Ikeda T et al. Magnetic compression anastomosis: a novel
technique for canalization of severe hilar bile duct strictures. Endoscopy. 2005; 37
(12): 1248-51.
11
Okajima H, Kotera A, Takeichi T et al. Magnet compression anastomosis for bile duct
stenosis after duct-to-duct biliary reconstruction in living donor liver transplantation. Liver Transpl. 2005; 11 (4): 473-5.
12
Avaliani M, Chigogidze N, Nechipai A, Dolgushin B. Magnetic compression biliary-
enteric anastomosis for palliation of obstructive jaundice: initial clinical results. J
Vasc Interv Radiol. 2009; 20 (5): 614-23.
13
Oya H, Sato Y, Yamanouchi E et al. Magnetic compression anastomosis for bile duct
stenosis after donor left hepatectomy: a case report. Transplant Proc. 2012; 44 (3):
806-9.
14
Suyama K, Takamori H, Yamanouchi E et al. Recanalization of obstructed
choledochojejunostomy using the magnet compression anastomosis technique. Am J Gastroenterol. 2010; 105 (1): 230-1.
15
Cope C. Evaluation of compression cholecystogastric and cholecystojejunal
anastomoses in swine after peroral and surgical introduction of magnets. J Vasc Interv
Radiol. 1995; 6 (4): 546-52.
16
Cope C. Creation of compression gastroenterostomy by means of the oral,
percutaneous, or surgical introduction of magnets: feasibility study in swine. J Vasc
This article is protected by copyright. All rights reserved.
Accepted Article
Interv Radiol. 1995; 6 (4): 539-45.
17
Yamanouchi E, Sugiura K, Yasuda Y, al e. Development of new technology in
interventional radiology of pancreas and biliary system. Jpn J Intervent Radiol. 2000;
15: 56-61.
18
Kaidar-Person O, Rosenthal RJ, Wexner SD, Szomstein S, Person B. Compression anastomosis: history and clinical considerations. Am J Surg. 2008; 195 (6): 818-26.
19
Jamshidi R, Stephenson JT, Clay JG, Pichakron KO, Harrison MR. Magnamosis:
magnetic compression anastomosis with comparison to suture and staple techniques. J
Pediatr Surg. 2009; 44 (1): 222-8.
20
Aggarwal R, Darzi A. Compression anastomoses revisited. J Am Coll Surg. 2005; 201
(6): 965-71.
21
Kopelman D, Hatoum OA, Kimmel B et al. Compression gastrointestinal anastomosis. Expert Rev Med Devices. 2007; 4 (6): 821-8.
22
Hardy KJ. Non-suture anastomosis: the historical development. Aust N Z J Surg.
1990; 60 (8): 625-33.
23
JB M. Cholecysto-intestinal, gastro-intestinal, entero-intestinal anastomosis, and
approximation without sutures. . Med Rec NY. 1892; 42: 665-76.
24
Swain CP, Mills TN. Anastomosis at flexible endoscopy: an experimental study of
compression button gastrojejunostomy. Gastrointest Endosc. 1991; 37 (6): 628-31.
25
Pryor HI, 2nd, Lange PA, Bader A, Gilbert J, Newman K. Multiple magnetic foreign
body ingestion: a surgical problem. J Am Coll Surg. 2007; 205 (1): 182-6.
26
Centers for Disease C, Prevention. Gastrointestinal injuries from magnet ingestion in
children--United States, 2003-2006. MMWR Morb Mortal Wkly Rep. 2006; 55 (48):
1296-300.
27
Cauchi JA, Shawis RN. Multiple magnet ingestion and gastrointestinal morbidity.
This article is protected by copyright. All rights reserved.
Accepted Article
Arch Dis Child. 2002; 87 (6): 539-40.
28
Honzumi M, Shigemori C, Ito H, Mohri Y, Urata H, Yamamoto T. An intestinal fistula
in a 3-year-old child caused by the ingestion of magnets: report of a case. Surg Today. 1995; 25 (6): 552-3.
29
Jansen A, Keeman JN, Davies GA, Klopper PJ. Early experiences with magnetic rings
in resection of the distal colon. Neth J Surg. 1980; 32 (1): 20-7.
30
Saveliev VS, Avaliani MV, Bashirov AD. Endoscopic magnetic cholecystodigestive
anastomoses: personal technique for palliative treatment of distal bile duct obstruction. J Laparoendosc Surg. 1993; 3 (2): 99-112.
31
. MAGDA : Magnetic force Determination Algorithm. : http://magda.ucc.ie.
32
Savelev VS, Avaliani MV, Kapranov SA, Bashirov AD, Boldina TB. [Endoscopic
biliodigestive anastomosis with the use of magnets (experimental and clinical study)]. Khirurgiia (Mosk). 1993 (3): 10-8.
33
Yanamouchi E, Kawaguchi H, Endo I et al. A new interventional method : magnetic
compression anastomosis with rare-earth magnets. Cardiovas Intervent Radiol. 1998; 22: S155, abstract.
34
Mimuro A, Tsuchida A, Yamanouchi E et al. A novel technique of magnetic
compression anastomosis for severe biliary stenosis. Gastrointest Endosc. 2003; 58 (2): 283-7.
35
Muraoka N, Uematsu H, Yamanouchi E et al. Yamanouchi magnetic compression
anastomosis
for
bilioenteric
anastomotic
stricture
after
living-donor
liver
transplantation. J Vasc Interv Radiol. 2005; 16 (9): 1263-7.
36
Akita H, Hikita H, Yamanouchi E et al. Use of a metallic-wall stent in the magnet
compression anastomosis technique for bile duct obstruction after liver transplantation. Liver Transpl. 2008; 14 (1): 118-20.
This article is protected by copyright. All rights reserved.
Accepted Article 37
Yukawa N, Rino Y, Yamanouchi E et al. [A case of magnetic compression
anastomosis between the common bile duct and the duodenum after distal gastrectomy with Roux-Y reconstruction and cholecystectomy]. Nihon Shokakibyo Gakkai Zasshi. 2008; 105 (10): 1523-8.
38
Matsuno N, Uchiyama M, Nakamura Y et al. A nonsuture anastomosis using magnetic
compression for biliary stricture after living donor liver transplantation. Hepatogastroenterology. 2009; 56 (89): 47-9.
39
Itoi T, Yamanouchi E, Ikeuchi N, Kasuya K, Iwamoto H, Tsuchida A. Magnetic
Compression Duct-to-duct Anastomosis for Biliary Obstruction in a Patient with Living Donor Liver Transplantation. Gut Liver. 2010; 4 Suppl 1: S96-8.
40
Itoi T, Kasuya K, Sofuni A et al. Magnetic compression anastomosis for biliary
obstruction: review and experience at Tokyo Medical University Hospital. J Hepatobiliary Pancreat Sci. 2011; 18 (3): 357-65.
41
Lambe T, Riordain MG, Cahill RA, Cantillon-Murphy P. Magnetic Compression in
Gastrointestinal and Bilioenteric Anastomosis: How Much Force? Surg Innov. 2013.
42
Pichakron KO, Jelin EB, Hirose S et al. Magnamosis II: Magnetic compression
anastomosis for minimally invasive gastrojejunostomy and jejunojejunostomy. J Am Coll Surg. 2011; 212 (1): 42-9.
43
Wall J, Diana M, Leroy J et al. MAGNAMOSIS IV: magnetic compression
anastomosis for minimally invasive colorectal surgery. Endoscopy. 2013; 45 (8): 643-
8.
44
Gonzales KD, Douglas G, Pichakron KO et al. Magnamosis III: delivery of a
magnetic compression anastomosis device using minimally invasive endoscopic techniques. J Pediatr Surg. 2012; 47 (6): 1291-5.
45
Cope C, Clark TW, Ginsberg G, Habecker P. Stent placement of gastroenteric
This article is protected by copyright. All rights reserved.
Accepted Article
anastomoses formed by magnetic compression. J Vasc Interv Radiol. 1999; 10 (10):
1379-86.
46
Yan X, Fan C, Ma J et al. Portacaval shunt established in six dogs using magnetic
compression technique. PLoS One. 2013; 8 (9): e76873.
47
Zepeda-Gomez S, Baron TH. Benign biliary strictures: current endoscopic
management. Nat Rev Gastroenterol Hepatol. 2011; 8 (10): 573-81.
48
Itoi T, Ishii K, Sofuni A et al. Single-balloon enteroscopy-assisted ERCP in patients
with Billroth II gastrectomy or Roux-en-Y anastomosis (with video). Am J Gastroenterol. 2010; 105 (1): 93-9.
This article is protected by copyright. All rights reserved.
Accepted Article
Figures and legends Figure 1. Indications for use of magnetic compression anastomosis (MCA). The cholangiogram reveals a biliary stricture present after living donor liver transplantation. A, Endoscopy cannot be used to pass a guidewire because of a severe stricture at the site of anastomosis. B, After injection of contrast medium through a percutaneous transhepatic tract, the media cannot drain to the common bile duct because the obstruction is complete. As shown in these cases, MCA can be applied in instances of severe stenosis or complete obstruction through which a guidewire cannot be passed. Figure 2. A cholangiogram showing how magnetic compression anastomosis can be used to treat a biliobiliary stricture developing after living donor liver transplantation. A, After insertion of a percutaneous transhepatic biliary drainage (PTBD) catheter, the tract was dilated to 18 F, and a covered self-expandable metal stent was inserted into the common bile duct (CBD). B, A magnet attached to a polypectomy snare was delivered via endoscopic
retrograde cholangiopancreatography (ERCP) scope through the CBD. C, Another magnet was fixed to alligator forceps and moved toward the anastomosis site through the PTBD tract. The magnets were approximated, and the PTBD catheter inserted. D, After 6 weeks, the approximated magnets were removed by percutaneous transhepatic cholangioscopy through the PTBD tract. E, After the approximated magnets were removed, a retrievable, fully converted, self-expandable metal stent (FCSEM) was inserted for 6 months. F, After 6 months, the FCSEM was removed and a new fistula was formed. Figure 3. A cholangiogram showing how magnetic compression anastomosis can be used to treat a biliobiliary stricture developing after cholecystectomy (adapted from reference 8). A, A percutaneous transhepatic biliary drainage (PTBD) catheter was inserted because the cystic duct site was completely obstructed as a result of ischemic injury. B, A magnet attached to a This article is protected by copyright. All rights reserved.
Accepted Article
polypectomy snare was delivered via endoscopic retrograde cholangiopancreatography (ERCP) scope through a previously inserted a retrievable, fully converted, self-expandable metal stent (FCSEM). C, After the magnets were successfully approximated, a PTBD catheter was inserted and the FCSEM removed. D, After 7 weeks, the approximated magnets were removed by percutaneous transhepatic cholangioscopy through the PTBD tract. E, An FCSEM was next inserted via the ampulla into the CBD and maintained for 6 months. F, After 6 months, the FCSEM was removed; recanalization was complete. Figure 4. A cholangiogram showing how magnetic compression anastomosis can be used to treat a bilioenteric stricture developing after Whipple’s operation with Roun-en-Y
anastomosis to treat pancreatic cancer. A, A percutaneous transhepatic biliary drainage (PTBD) catheter was inserted because the anastomosis site was completely obstructed, and the tract was dilated to 18 F. B, A magnet attached to a polypectomy snare was delivered using a duodenoscope, and another magnet was delivered through the PTBD tract. C, The
magnets were successfully approximated, and a PTBD catheter inserted. D, After 6 weeks, the approximated magnets were removed by percutaneous transhepatic cholangioscopy through the PTBD tract. E, After the approximated magnets were removed, an internal drainage catheter (16 Fr) was inserted and maintained for 6 months. F, After 6 months, the catheter was removed and development of a new fistula was confirmed. Figure 5. Magnetic compression anastomosis using other delivery routes to form a bilioenteric anastomosis. A. An endoscopic approach to the anastomosis site failed because the afferent loop was long. Thus, magnet delivery featured a surgically created skin/intestinal fistula. B, Both the left intrahepatic duct (IHD) and the right IHD were separately anastomosed to the jejunum, and the right IHD was obstructed. Two percutaneous transhepatic cholangioscopy scopes were used to deliver a magnet via the PTBD tract via the This article is protected by copyright. All rights reserved.
Accepted Article
right IHD tract (one scope), and another magnet through the left IHD tract (the second scope), to allow of magnet approximation. Figure 6. A cholangiogram showing failure of magnetic compressive anastomosis (adapted from reference 7). A, The magnet could not be delivered to the stricture because the distal common bile duct was tortuous and angulated (like the letter W; white arrow). B, Between-
magnet attraction across the long stricture (~30 mm) was weak, and the magnets would not come together.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 1. Outcomes of recanalization using magnetic compression anastomosis in biliobiliary stricture.
Year
Authors
Type of report
Age/sex (no.)
Reason for operation (no.)
Previous operation
Type of biliary obstruction
Distance between the two magnets (mm)
Time to removal of magnets (days)
Conduction of anastomosis
Duration of biliary drainage
Compli cation
Follow-up period /symptom
2003
Mimuro et al.34
Case
76/F
Pancreatic cancer
DP
Benign
12
21
Partial
NA
No
NA
2005
Itoi et al.10
Case
76/F
Hilar bile duct cancer
None (radiation)†
Malignant
8
14
Partial
NA
No
NA
2005
Okajima et al.11
Case
44/F
Fulminant hepatitis
LDLT (right lobe)
Benign
2
42
Complete
3 months
No
15 months asymptomatic
2008
Akita et al.36
Case
34/F
NA
LDLT (right lobe)
Benign
2
24
Complete
NA
No
NA
2009
Matsumo et al.38
Case
53/M
NA
LDLT (right lobe)
Benign
2
10
Complete
30 days
No
3 years asymptomatic
2010
Itoi et al.39
Case
60/M
NA
LDLT (right lobe)
Benign
NA
9
Complete
6 months
No
2 months asymptomatic
2011
Itoi et al.40
Case series
40/F
Metastasis live tumor derived from colon cancer
Right three segmental + S3 partial hepatectomy
Benign
15
30
Complete
6 months
No
2 years asymptomatic
2011
Jang et al.7
Original
Mean 53.8 (range 33-64) M:F=9:3
LC (3),HCC (7), HF (1)
LDLT (right lobe)
Benign
NA
Mean 74.2 (range 14-181)
Complete
Mean 183 days (range 99-581 days)
No
Mean 303 days
24/M
NA
8 months
No
2012
Oya
Case
LDLT
This article is protected by copyright. All rights reserved.
Benign
NA
60
Partial
Restenosis (1) NA
Accepted Article et al.13
2014
Jang et al.8
(right lobe) Case series
Case#1: 45/M Case#2: 38/F
#1: Abdominal trauma #2: Cholecystitis
#1: embolization‡ #2 : cholecystecomy
Benign
#1: #2:
4 6
#1: #2:
63 32
Complete
NA
No
#1: 468 days #2: 80 days asymptomatic
M, male; F, female; NA, not available; LC, liver cirrhosis; HCC, hepatocellular carcinoma; HF, hepatic failure; DP, dorsal pancreatectomy; LDLT, living donor liver transplantation. † Bile duct obstruction occurred after intraluminal radiation to treat bile duct cancer. ‡ Bile duct obstruction occurred because of ischemic injury to the bile duct after embolization of the hepatic artery.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 2. Outcomes of recannalization using magnetic compression anastomosis in bilioenteric stricture..
Year
Authors
Type of report
Age/sex (no.)
Reason for operation (no.)
Previous operation
Type of biliary obstruction
Distance between the two magnets (mm)
Time to removal of magnets (days)
Conduction of anastomosis
Duration of biliary drainage
Comp licati on
Follow-up period /symptoms
2001
Takao et al.9
Case
70/M
Gastric cancer
Subtotal gastrectomy (B-II)
Benign
2
32
Complete
90 days
No
2 years
2005
Muraoka et al.35
Case
Case #1: 1/F
#1: Fulminant hepatitis
#1: LDLT (left lobe) with R-Y
Benign
#1:
2
#1:
12
Complete
8 weeks
No
#2:
5
#2:
13
Case#2: 57/M
#2: LC+HCC
#2: LDLT (right lobe) with R-Y
#1: 54 months /asymptomatic #2: 6 days /recurrence and 6 months /asymptomatic
2008
Yukawa et al.37
Case
83/M
Gastric and gallbladder cancer
Distal gastrectomy with R-Y and cholecystectomy
Benign
NA
21
Complete
NA
Slight fever
NA
2009
Avaliani et al.12
Original
Mean 64 (range 4682) M:F=9:25
AOV cancer (7) Pancreatic cancer (21) CCC (6)
None†
Malignant ‡
NA
7-10
Complete (except one case)
No drain
No
Re-intervention (6 cases) after 9.6 months (mean; range 2-36 months)
2010
Suyama et al.14
Case
78/M
Gallbladder cancer
Radical cholecystectomy with R-Y
Benign
NA
34
Complete
6 months
No
NA
2011
Itoi et al.40
Case
60/F
CCC
Expanded left lobectomy with
Benign
2
17
Complete
Not removed
No
NA
This article is protected by copyright. All rights reserved.
Accepted Article 2014
Jang et al.8
R-Y Case series
Case#1: 49/M
#1: Pancreatic NET #2: Choledochal cyst
Case#2: 27/M
#3: Pancreatic NET
Case#3: 63/F
#1: PPPD with HJstomy #2: Excision of cyst with R-Y #3: Whipple operation with R-Y
Benign
#1:
5
#1:
36
#2:
5
#2:
40
#3:
7
#3: 14
Complete
NA
No
#1:
202 days
#2:
1,573 days
#3: 103 days /asymptomatic
M, male; F, female; NA, not available; LC, liver cirrhosis; HCC, hepatocellular carcinoma; AOV, ampulla of Vator; CCC, cholangiocellular carcinoma; NET, neuroendocrine tumor; B-II, Billroth II; LDLT, living donor liver transplantation; R-Y, Roux-en-Y anastomosis; PPPD, pylorus-preserving pancreatico-duodenectomy. † The tumor was not treated with radical surgery. ‡ The obstructions were caused by cancer, but the sites requiring anastomoses in the bile duct and duodenum were benign.
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f1A_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f1B_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f2_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f3_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f4_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f5A_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f5B_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f6A_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f6B_AA
This article is protected by copyright. All rights reserved.
Title page
Magnetic Compression Anastomosis for Treatment of Benign Biliary Strictures1 Sung Ill Jang1 MD, Jungran Choi2 MD, Dong Ki Lee2 MD, PhD
1
Department of Internal Medicine, Kangnam Sacred Heart Hospital Hallym University College of Medicine, Seoul, Korea 2
Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea
Correspondence: Dong Ki Lee MD, PhD Address: Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, 211 Eonju-Ro, Gangnam-Gu, Seoul, Korea Phone: 82-2-2019-3214 This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/den.12319 This article is protected by copyright. All rights reserved.
Accepted Article
Fax: 82-2-3463-3882 E-mail: [email protected] Article type: Review Received date: 25-Apr-2014 Accepted date: 30-May-2014 Abstract
Endoscopic and percutaneous procedures have shown high success rates when used to
treat benign biliary strictures (BBS). However, cases in which a guidewire cannot be passed through a refractory stricture or a complete obstruction are difficult to treat using conventional methods. Magnetic compression anastomosis (MCA) has emerged as a nonsurgical alternative avoiding operational mortality and morbidity. The feasibility and safety of MCA have been experimentally and clinically verified in cases of biliobiliary and bilioenteric anastomosis. However, no pre-MCA assessment modality capable of predicting outcomes is yet available, and no universally effective magnet delivery method has yet been established, rendering it difficult to identify patients for whom MCA is appropriate. Various experimental studies seeking to overcome these limitations are underway. Such work will improve our indepth understanding of MCA, which has been trialed in various fields. Upon further development, MCA may become a groundbreaking option for treatment of benign strictures that are difficult to resolve using conventional methods, and MCA may be expected to be minimally traumatic and highly effective. The aim of the present study was to discuss the current status of MCA and the direction of MCA development by reviewing clinical and experimental MCA data. Key words: Magnetic compression anastomosis, compressive anastomosis, benign biliary This article is protected by copyright. All rights reserved.
Accepted Article
stricture, biliary obstruction, anastomosis stricture.
This article is protected by copyright. All rights reserved.
Accepted Article
Introduction Advances in non-surgical methods, including endoscopic or percutaneous approaches,
have enabled recanalization of severe biliary strictures, benign or malignant, and strictures that develop postoperatively.1-6 However, non-surgical methods may be of limited efficacy when used to treat severe biliary strictures or complete obstructions; placement and maintenance of drainage catheters are required in patients who fail recanalization using nonsurgical methods. Recently, magnetic compression anastomosis (MCA) has been developed as a non-surgical alternative treatment for patients in whom a conventional endoscopic or percutaneous method has failed (Fig. 1).7-14 The clinical feasibility, safety, and utility of MCA have been established in various cases involving stenosis and occlusions, without any need for surgery.15-17 This review focuses on novel applications of MCA for the biliobiliary and bilioenteric anastomosis of stenosis and occlusions.
History
The concept of compression anastomosis was first proposed by Denan in 1826, who
described a newly formed non-sutureless anastomotic fistula caused by ischemic compression of tissue.18 Denan’s spring-loaded device was revised and further developed by Murphy in 1892 and became known as Murphy’s button.19-22 Use of the technique enables formation of a circular gastrointestinal anastomosis via ischemic compression of tissue held between two buttons by a spring, and was the first surgical device used to prepare functioning end-to-end and side-to-side sutureless anastomoses via ischemic compression.23 In 1991, an animal study used compression buttons and modified Murphy’s buttons to perform gastrojejunostomy, aided by endoscopy.24 The mechanical device forms a compression anastomosis via contact of screws, snaps, or springs of the two components. Such physical contact can be replaced by magnetic attraction, using a field-mediated magnetic force. The effects of transluminal
This article is protected by copyright. All rights reserved.
Accepted Article
magnetic force transduction via both a fistula and a free perforation have been analyzed in several children who had ingested magnets.25-28 Jansen et al.29 performed the first human experiments in 1980; tissue compression was achieved via magnetic attraction achieved using mechanical compression. Mucosa-to-mucosa anastomosis using MCA was successfully performed in five patients with colonic resections. Subsequently, in 1993, Salvelev et al.30 performed clinical and laboratory studies on mongrel dogs. Successful cholecystoenteric, enteroenteric, and magnetic cholecystogastric anastomoses were formed. In addition, preliminary data from four patients who underwent cholecystogastric anastomoses and one who underwent a cholecystoduodenal anastomosis have demonstrated the feasibility of endoscopic magnetic cholecystodigestive anastomosis.31 Together, the work to date supports both the concept and the clinical feasibility of compression anastomosis using magnets.32 In 1998, Yamanouchi33 was the first to use a modern MCA method in a clinical trial, successfully generating a bile duct-small intestinal fistula and introducing novel uses of MCA. Other clinical outcomes have also been reported.6-14,34-40 Magnets
Magnetic power is critical contributor if MCA is to successful. Rare-earth magnets are
classified into neodymium iron-boron magnets and samarium-cobalt (Sm-Co) magnets. Both have high magnetic flux densities and retention forces, rendering them suitable for MCA. However, the retention force of a Sm-Co magnet is stronger than that of a neodymium ironboron magnet, and the former type of magnet is thus more frequently used. 7-10,34,41 To quantify magnetic power, the magnetic powers of magnets used in published studies were calculated using a magnetic force determination algorithm (MAGDA).31,41 The work suggested that such calculations helped to predict the success of MCA. Variables including magnet shape, dimensions, the nature of the magnetic material, the magnetic grade or
This article is protected by copyright. All rights reserved.
Accepted Article
strength, and experimentally derived or estimated in vivo magnetic separation forces, may be entered into the MAGDA. Animal studies In the 1990s, efforts were made to induce compression anastomosis via magnetic
attraction between powerful rare-earth magnets. In 1995, Cope demonstrated the feasibility and safety of MCA by creating bilioenteric and enteroenteric anastomoses in pigs.15,16 Cope used neodymium-iron-boron or rare-earth Sm-Co magnets to perform cholecystogastric and cholecystojejunal MCA in swine, and a bilioenteric anastomosis formed via magnet approximation after 9-16 weeks.15 A preliminary study revealed that magnets could be used to create leak-free short-term enteroenteric anastomoses in swine.16 The shape of magnets used
in subsequent MCA experiments was modified to amplify the magnetic effects, and further animal studies were performed. Jamshidi et al.19 performed MCA using both uniform and gradient compression methods, and compared stapled anastomoses to those that had been additionally hand-sutured. Also, all of gross appearance, histology, functionality (assessed radiologically), and mechanical integrity were evaluated. Although one stapled anastomosis leaked, no magnetic anastomosis did so. No severe complication and no appreciable stenosis were observed. The burst pressures of MCA-formed and surgically formed anastomoses did not differ. Upon pathological examination, MCA-formed anastomoses evidenced continuity of the serosal, submucosal, and mucosal layers, and neither ischemia nor necrosis was present. Thus, MCA was safe, and equivalent or superior to anastomoses created by traditional suturing or stapling techniques.19 Further, the same team showed that MCA-assisted enteroenterostomy was feasible, using modified magnets in the form of two convex-concave radially symmetrical halves.42 The design and development of a controlled MCA system (magnamosis) that optimizes all of magnetic coupling, the distance between magnets, and This article is protected by copyright. All rights reserved.
Accepted Article
mating of surface geometries, are required to achieve reliable enteric anastomosis.19,42,43 A magnamosis device has the following three main features: (1) two convex-concave radially symmetric rings that self-align magnetically; (2) a ring-shaped magnet that allows immediate patency; and (3) a mating surface with a radial topography specifically engineered to promote necrosis at the center and healing at the periphery; this ensures that the anastomosis formed is free of any perforation. Apart from modifications to magnets, animal studies have been conducted to optimize
endoscopic magnet delivery.44 A modular flexible magnetic anastomotic device has been developed, the use of which ensured absence of leakage. One study used partially covered stents to improve both the modular shapes of the magnets and the patency of an MCA-formed fistula.45 When partially covered stents were inserted into MCA-formed gastroenteric anastomoses, patency was maintained for more than 7 weeks. Compression anastomosis using magnets has been experimentally attempted in all of the vascular, biliary, and gastroenterological tracts.46 Studies in humans Materials and Methods
Bilio-biliary anastomosis The magnets used in MCA are cylindrical Sm-Co rare-earth magnets, of various powers,
which can be delivered by a variety of methods.41 However, the most common delivery route is via the percutaneous and peroral bile ducts. The MCA procedure may be divided into the four following steps (Figs. 2, 3): (1) tract formation for magnet delivery; (2) magnet approximation; (3) magnet removal; and, (4) maintenance and removal of the internal catheter.
This article is protected by copyright. All rights reserved.
Accepted Article
(1) Tract formation for magnet delivery: A percutaneous transhepatic biliary drainage (PTBD) tract for magnet delivery is formed using an 18-Fr PTBD catheter. The PTBD catheter is exchanged for an 18-Fr sheath prior to MCA approximation, to allow convenient magnet insertion via the PTBD tract and to reduce duct injury. In the common bile duct (CBD) tract, full endoscopic sphincterotomy (EST) and balloon dilation or transient insertion of a retrievable, fully converted, self-expandable metal stent (FCSEM), are used to facilitate magnet delivery.
(2) Magnet approximation: A thread attached to one magnet is fixed to a polypectomy snare, and the magnet is moved to the anastomosis site via PTBD tract. The polypectomy snare is passed through the channel of an endoscopic retrograde cholangiopancreatography (ERCP) scope, and the other magnet is fixed in front of the scope. The magnet is moved to the anastomosis site through the FCSEM, and MCA approximation occurs via attraction between the two magnets. To better approximate the magnets, a balloon catheter may be used to advance the magnets through both the PTBD and ERCP tracts. Approximation of the two magnets is confirmed radiographically. Next, the long sheath tube is removed and the indwelling PTBD catheter inserted. The FCSEM inserted in the CBD is removed immediately after magnet approximation.
(3) Magnet removal: When a fistula forms because of ischemic necrosis caused by the approximated magnets, the magnets spontaneously migrate to the duodenum. However, if spontaneous migration does not occur after ~8 to 10 weeks, the magnets can be pushed out using a guidewire or catheter. The magnets can also be removed through the PTBD tract via percutaneous transhepatic cholangioscopy (PTCS).
(4) Maintenance and removal of the internal catheter: After magnet removal, a 14-16-Fr
This article is protected by copyright. All rights reserved.
Accepted Article
internal catheter is inserted into the fistula and maintained in place for 4-6 months to reduce the probability of re-stenosis of the fistulous tract. A FCSEM is more useful than an internal catheter, because the FCSEM allows formation of a larger fistula. However, a FCSEM is more likely to migrate than an indwelling catheter. Further work is necessary to confirm the efficacy of FCSEMs.
Bilio-enteric anastomosis In general, the MCA methods and concepts used in bilioenteric anastomosis are similar to
those of biliobiliary anastomosis. However the delivery routes of magnets differ (Fig. 4). Several different routes including a surgically formed percutaneous-jejunum tract;8,35 a
percutaneous-percutaneous tract;8 or a percutaneous-peroral tract (the most common approach), may be used to deliver magnets. The method of magnet delivery to the percutaneous tract is identical to that described above, but a forward-viewing endoscope is used in any peroral approach. However, an endoscopic approach is difficult in patients with long afferent loops. In such cases, magnets may be delivered via a surgically created skin/intestinal fistula (Fig. 5A).8,35 One study has suggested that single-balloon enteroscopy is useful in such situations.40 Rarely, the left intrahepatic duct (IHD) and the right IHD may be separately anastomosed to the jejunum, with the stricture evident on the right side (Fig. 5B). Two PTCS scopes are used in this context; one to deliver a magnet through the right IHD via the PTBD tract, and the other to approximate the second magnet at the stricture site via the left IHD tract.8 Results
Bilio-biliary anastomosis Recent reports on human MCA have described 22 MCA biliobiliary anastomoses (in 13
This article is protected by copyright. All rights reserved.
Accepted Article
males and 9 females) (Table 1). The most common causes of benign biliary stricture requiring MCA were postoperative complications, especially after living-donor liver transplantation (17 cases). In addition to those occurring post-operatively, some strictures were caused by ischemic injury by radiation and embolization. The mean distance between the two magnets was 6.4 mm (range, 2–15 mm), and the mean time to magnet removal 53.3 days (range, 9– 181 days). During the follow-up period (mean, 12 months; range, 2-36 months), re-stenosis occurred in only a single case who was re-cannulated using a conventional method. No obstructive jaundice or related symptoms was/were found in the remaining 21 cases. Bilio-enteric anastomosis Forty-two bilioenteric MCA procedures have been performed in patients with 8 benign and
34 malignant biliary strictures (Table 2). The causes of benign biliary stricture (BBB) were post-operative complications, and the principal anastomosis method used was Roux-en-Y
reconstruction. Magnets were delivered via an endoscope or balloon endoscope in patients undergoing Roux-en-Y reconstruction, but if the anastomosis site was not attained, magnet delivery was performed via a surgically created skin fistula.8,35 In addition, two PTBD tracts
were used for magnet delivery to stricture sites.8 These cases highlight the various delivery
methods that may be employed. The mean distance between the two magnets was 4 mm (range, 2–7 mm), and the time to magnet removal 7–40 days. Of the 42 subjects, 41 (97.6%) developed complete anastomoses and no MCA-related complications were noted. During the follow-up period (mean, 40 months; range, 2–53 months), re-stenosis occurred in a single
case 6 days following removal of the drainage tube, but re-cannulation after balloon dilation was performed without difficulty.35 Discussion
This article is protected by copyright. All rights reserved.
Accepted Article
BBB can be caused by infection, inflammation, an autoimmune disease, ischemic injury,
trauma, and intra- or post-operative complications.47 The feasibility and safety of MCA for a BBB that cannot be recanalized using a conventional method continue to be verified in various experimental and clinical studies.7-10,12,34,40 MCA-associated patency is clinically similar to that of an anastomosis formed by conventional surgery.17 In an animal model, the anastomosis site exhibited continuity of the serosal, submucosal, and mucosal layers. Ischemia and necrosis were absent, and the burst strengths of anastomosis formed by MCA was equaled or exceeded that of a surgical approach.42,44 Doppler ultrasound-based screening and follow-up are often performed because of the
possibility of rupture during MCA if blood vessels are placed between two magnets. 34,35 However, no blood vessel rupture or other complication has yet been reported in any clinical trial. This is believed to be because fistula formation after MCA requires a relatively long time. Use of two magnets renders the closing gradual. Thus, even if an intervening vessel is present, neither compression nor rupture is to be expected. In previous studies, the mean time to magnet removal after successful approximation was 53.3 days (range, 9–181 days) in cases
of biliobiliary anastomosis and 7–40 days in cases of bilioenteric anastomosis; these times are rather long. The time to magnet removal after successful approximation depends on the distance between the two magnets, the magnetic field strength, and the histological situation around the obstruction.39 Consequently, the relatively short duration required for bilioenteric anastomosis is explained by the fact that the distance between the two magnets (2–7 mm) is less than in biliobiliary anastomosis (2–15 mm). Typically, partial recanalization requires at
least 10 days for a short obstruction and ~1 month for a long obstruction.39 Currently, no long-term clinical follow-up data on new fistulas formed via MCA are
available. However, MCA forms fistulas via tissue necrosis, without dilation of fibrotic tissue. This article is protected by copyright. All rights reserved.
Accepted Article
Therefore, the risk of restenosis upon recoiling of fibrotic tissue should be low. In one case of biliobiliary anastomosis, no restenosis was reported after 3 years.38 In patients with biliary stenosis developing after living-donor liver transplantation, restenosis was observed in 1 of 12 subjects over 331 days, but recanalization was easily performed via PTBD.7 In studies on bilioenteric anastomosis, one case exhibited no restenosis for 50 months 8, and no recurrence was observed in malignant patients to 30 days after MCA.12 The suggested low recurrence rate after MCA will be validated in larger studies with long-term follow-up. Pre-MCA assessment is limited to predicting outcomes and planning delivery methods;
these issues need to be addressed. The success of MCA depends on several factors such as a length of the stricture, a shape of bile duct, a pattern of direction of each magnet and bile duct axis. The main causes in unsuccessful MCA cases are a long length of the stricture, a tapered or tortuous duct and/or the parallel axis of alignment.7,8 The magnetic force weakens as the
length of a stricture increases. In such circumstances, compressive necrosis will not occur, inhibiting fistula formation. Thus, accurate evaluation of stricture length prior to MCA is important. However, currently, the stricture length cannot be accurately evaluated using noninvasive imaging modalities such as computed tomography, ultrasonography, or magnetic resonance cholangiopancreatography. Duct evaluation via cholangiography is reasonably accurate, but has the disadvantage that an invasive procedure such as ERCP or PTBD must be performed. Apart from stricture length, the shape and axis of the bile duct are also important in terms of pre-MCA assessment. If the duct is tapered or tortuous (Fig. 6A), the magnets cannot maximally approach the stricture site despite a short stricture length, and the actual length between the two magnets is thus longer than measured during pre-MCA assessment, leading to MCA failure (Fig. 6B).7 Moreover, the bile duct axis determines the direction of magnet alignment. Even if the distance between magnets is short, as in most instances of
This article is protected by copyright. All rights reserved.
Accepted Article
successful MCA, the technique may still fail if the alignment is parallel; the magnetic power becomes weak.7,8 As no non-invasive method can accurately evaluate stricture length, duct
shape, or the duct axis, all of which are required to plan successful MCA, a limitation in identifying suitable MCA candidates becomes apparent. The result can be known only when magnets are actually applied. The magnet delivery method chosen depends on the type of anastomosis required, prior
operative history, and patient characteristics in both biliobiliary and bilioenteric anastomosis. Biliobiliary anastomosis delivers magnets via PTBD and ERCP. To prevent hepatic duct injury during magnet delivery through the PTBD tract, it is preferable to employ 16- or 18-Fr sheathes for insertion. Magnet delivery into the CBD is more difficult with ERCP than PTBD, because delivery through the papilla of Vater is challenging and may sometimes fail. Delivery of a 5-mm magnet is difficult via EST alone, and balloon dilation is often used, but manipulation of the magnet becomes difficult at times.7 To solve this problem, a metal stent can be transiently inserted into the papilla.7,36 The stent can be inserted on the day of magnet approximation, or 1 day prior, to minimize stent migration and pancreatitis. In other words, the stent indwelling time should be minimal. Generally, during Roux-en-Y bilioenteric anastomosis, magnet delivery to the stricture site via ERCP is difficult because of the long lengths of the efferent loop and afferent limbs, and the risk of intestinal perforation. In such cases, transparent cap-installed forward colonoscopy8 and balloon endoscopy48 can be helpful, but do not guarantee success in all cases, and the techniques are associated with a risk of intestinal perforation. Recently, MCA magnet delivery via surgically formed skin/intestinal fistulas, has been reported; this may be a useful alternative method.8,35 In summary, various magnet delivery methods have been developed by consideration of patient characteristics, prior operative history, and the type of anastomosis required at the stricture site. Further
This article is protected by copyright. All rights reserved.
Accepted Article
development of effective and safe delivery methods is needed. The feasibility and safety of biliobiliary and bilioenteric anastomoses created via MCA
have been verified in both human and animal studies. In addition, Avalinani et al.12 used MCA to form anastomoses between the normal bile duct and the duodenum or jejunum in 34 patients with malignant strictures, but not for recanalization of malignant obstruction. Re-intervention was required in six subjects.12 However, MCA is not normally indicated to treat malignant biliary obstructions, which are often treatable using conventional peroral or percutaneous methods. Entero-enteric fistulous tract formation via MCA has been successful in experimental animal models, and has been reported to be compatible with a surgically formed anastomosis.42,44 However, other than a small amount of work in Japan, no study has
yet sought to construct human enteroenteric anastomoses via MCA. Gastrojejunostomy via MCA might be more efficacious in the management of malignant gastric outlet obstructions than is the current use of bare self-expandable metal stents.45 In summary, MCA is expected
to have diverse clinical applications including formation of gastrojejunal and jejunojejunal anastomoses creating gastric bypasses in obese subjects; removal of short strictures; esophageal reconstruction; closure of ileostomies and colostomies; bypassing of malignant obstructions; and others.
Conclusions MCA is a non-surgical alternative for treatment of occluded BBS that are difficult to
resolve using conventional endoscopic or percutaneous methods, and is both feasible and safe. For effective MCA and successful recanalization, a pre-MCA assessment method predicting outcomes, smaller and more powerful magnets, and an effective magnet delivery system, need to be developed. Moreover, endoscopists should fully understand the mechanisms and principles of MCA and expand the clinical indications for MCA, thereby enabling application This article is protected by copyright. All rights reserved.
Accepted Article
and development of the technique in various fields. Despite the limited number of cases reported to date, MCA is effective, safe, has a low recurrence rate, and is less traumatic than other treatments for occluded BBS. Reference 1
Dumonceau JM, Deviere J, Delhaye M, Baize M, Cremer M. Plastic and metal stents
for postoperative benign bile duct strictures: the best and the worst. Gastrointest
Endosc. 1998; 47 (1): 8-17.
2
Bonnel DH, Liguory CL, Lefebvre JF, Cornud FE. Placement of metallic stents for
treatment of postoperative biliary strictures: long-term outcome in 25 patients. AJR
Am J Roentgenol. 1997; 169 (6): 1517-22.
3
Baba S, Yoshioka U, Sotome T et al. [Case of von Hippel-Lindau disease with
pancreatic exocrine dysfunction and benign biliary stenosis due to multiple pancreatic serous cystadenoma]. Nihon Naika Gakkai Zasshi. 1996; 85 (10): 1765-7.
4
Ernst O, Sergent G, Mizrahi D, Delemazure O, L'Hermine C. Biliary leaks: treatment
by means of percutaneous transhepatic biliary drainage. Radiology. 1999; 211 (2):
345-8.
5
Vogel SB, Howard RJ, Caridi J, Hawkins IF, Jr. Evaluation of percutaneous
transhepatic balloon dilatation of benign biliary strictures in high-risk patients. Am J
Surg. 1985; 149 (1): 73-9.
6
Mita A, Hashikura Y, Masuda Y et al. Nonsurgical policy for treatment of bilioenteric
anastomotic stricture after living donor liver transplantation. Transpl Int. 2008; 21 (4):
320-7.
7
Jang SI, Kim JH, Won JY et al. Magnetic compression anastomosis is useful in biliary
anastomotic strictures after living donor liver transplantation. Gastrointest Endosc.
This article is protected by copyright. All rights reserved.
Accepted Article
2011; 74 (5): 1040-8.
8
Jang SI, Rhee K, Kim H et al. Recanalization of refractory benign biliary stricture
using magnetic compression anastomosis. Endoscopy. 2014; 46 (1): 70-4.
9
Takao S, Matsuo Y, Shinchi H et al. Magnetic compression anastomosis for benign
obstruction of the common bile duct. Endoscopy. 2001; 33 (11): 988-90.
10
Itoi T, Yamanouchi E, Ikeda T et al. Magnetic compression anastomosis: a novel
technique for canalization of severe hilar bile duct strictures. Endoscopy. 2005; 37
(12): 1248-51.
11
Okajima H, Kotera A, Takeichi T et al. Magnet compression anastomosis for bile duct
stenosis after duct-to-duct biliary reconstruction in living donor liver transplantation. Liver Transpl. 2005; 11 (4): 473-5.
12
Avaliani M, Chigogidze N, Nechipai A, Dolgushin B. Magnetic compression biliary-
enteric anastomosis for palliation of obstructive jaundice: initial clinical results. J
Vasc Interv Radiol. 2009; 20 (5): 614-23.
13
Oya H, Sato Y, Yamanouchi E et al. Magnetic compression anastomosis for bile duct
stenosis after donor left hepatectomy: a case report. Transplant Proc. 2012; 44 (3):
806-9.
14
Suyama K, Takamori H, Yamanouchi E et al. Recanalization of obstructed
choledochojejunostomy using the magnet compression anastomosis technique. Am J Gastroenterol. 2010; 105 (1): 230-1.
15
Cope C. Evaluation of compression cholecystogastric and cholecystojejunal
anastomoses in swine after peroral and surgical introduction of magnets. J Vasc Interv
Radiol. 1995; 6 (4): 546-52.
16
Cope C. Creation of compression gastroenterostomy by means of the oral,
percutaneous, or surgical introduction of magnets: feasibility study in swine. J Vasc
This article is protected by copyright. All rights reserved.
Accepted Article
Interv Radiol. 1995; 6 (4): 539-45.
17
Yamanouchi E, Sugiura K, Yasuda Y, al e. Development of new technology in
interventional radiology of pancreas and biliary system. Jpn J Intervent Radiol. 2000;
15: 56-61.
18
Kaidar-Person O, Rosenthal RJ, Wexner SD, Szomstein S, Person B. Compression anastomosis: history and clinical considerations. Am J Surg. 2008; 195 (6): 818-26.
19
Jamshidi R, Stephenson JT, Clay JG, Pichakron KO, Harrison MR. Magnamosis:
magnetic compression anastomosis with comparison to suture and staple techniques. J
Pediatr Surg. 2009; 44 (1): 222-8.
20
Aggarwal R, Darzi A. Compression anastomoses revisited. J Am Coll Surg. 2005; 201
(6): 965-71.
21
Kopelman D, Hatoum OA, Kimmel B et al. Compression gastrointestinal anastomosis. Expert Rev Med Devices. 2007; 4 (6): 821-8.
22
Hardy KJ. Non-suture anastomosis: the historical development. Aust N Z J Surg.
1990; 60 (8): 625-33.
23
JB M. Cholecysto-intestinal, gastro-intestinal, entero-intestinal anastomosis, and
approximation without sutures. . Med Rec NY. 1892; 42: 665-76.
24
Swain CP, Mills TN. Anastomosis at flexible endoscopy: an experimental study of
compression button gastrojejunostomy. Gastrointest Endosc. 1991; 37 (6): 628-31.
25
Pryor HI, 2nd, Lange PA, Bader A, Gilbert J, Newman K. Multiple magnetic foreign
body ingestion: a surgical problem. J Am Coll Surg. 2007; 205 (1): 182-6.
26
Centers for Disease C, Prevention. Gastrointestinal injuries from magnet ingestion in
children--United States, 2003-2006. MMWR Morb Mortal Wkly Rep. 2006; 55 (48):
1296-300.
27
Cauchi JA, Shawis RN. Multiple magnet ingestion and gastrointestinal morbidity.
This article is protected by copyright. All rights reserved.
Accepted Article
Arch Dis Child. 2002; 87 (6): 539-40.
28
Honzumi M, Shigemori C, Ito H, Mohri Y, Urata H, Yamamoto T. An intestinal fistula
in a 3-year-old child caused by the ingestion of magnets: report of a case. Surg Today. 1995; 25 (6): 552-3.
29
Jansen A, Keeman JN, Davies GA, Klopper PJ. Early experiences with magnetic rings
in resection of the distal colon. Neth J Surg. 1980; 32 (1): 20-7.
30
Saveliev VS, Avaliani MV, Bashirov AD. Endoscopic magnetic cholecystodigestive
anastomoses: personal technique for palliative treatment of distal bile duct obstruction. J Laparoendosc Surg. 1993; 3 (2): 99-112.
31
. MAGDA : Magnetic force Determination Algorithm. : http://magda.ucc.ie.
32
Savelev VS, Avaliani MV, Kapranov SA, Bashirov AD, Boldina TB. [Endoscopic
biliodigestive anastomosis with the use of magnets (experimental and clinical study)]. Khirurgiia (Mosk). 1993 (3): 10-8.
33
Yanamouchi E, Kawaguchi H, Endo I et al. A new interventional method : magnetic
compression anastomosis with rare-earth magnets. Cardiovas Intervent Radiol. 1998; 22: S155, abstract.
34
Mimuro A, Tsuchida A, Yamanouchi E et al. A novel technique of magnetic
compression anastomosis for severe biliary stenosis. Gastrointest Endosc. 2003; 58 (2): 283-7.
35
Muraoka N, Uematsu H, Yamanouchi E et al. Yamanouchi magnetic compression
anastomosis
for
bilioenteric
anastomotic
stricture
after
living-donor
liver
transplantation. J Vasc Interv Radiol. 2005; 16 (9): 1263-7.
36
Akita H, Hikita H, Yamanouchi E et al. Use of a metallic-wall stent in the magnet
compression anastomosis technique for bile duct obstruction after liver transplantation. Liver Transpl. 2008; 14 (1): 118-20.
This article is protected by copyright. All rights reserved.
Accepted Article 37
Yukawa N, Rino Y, Yamanouchi E et al. [A case of magnetic compression
anastomosis between the common bile duct and the duodenum after distal gastrectomy with Roux-Y reconstruction and cholecystectomy]. Nihon Shokakibyo Gakkai Zasshi. 2008; 105 (10): 1523-8.
38
Matsuno N, Uchiyama M, Nakamura Y et al. A nonsuture anastomosis using magnetic
compression for biliary stricture after living donor liver transplantation. Hepatogastroenterology. 2009; 56 (89): 47-9.
39
Itoi T, Yamanouchi E, Ikeuchi N, Kasuya K, Iwamoto H, Tsuchida A. Magnetic
Compression Duct-to-duct Anastomosis for Biliary Obstruction in a Patient with Living Donor Liver Transplantation. Gut Liver. 2010; 4 Suppl 1: S96-8.
40
Itoi T, Kasuya K, Sofuni A et al. Magnetic compression anastomosis for biliary
obstruction: review and experience at Tokyo Medical University Hospital. J Hepatobiliary Pancreat Sci. 2011; 18 (3): 357-65.
41
Lambe T, Riordain MG, Cahill RA, Cantillon-Murphy P. Magnetic Compression in
Gastrointestinal and Bilioenteric Anastomosis: How Much Force? Surg Innov. 2013.
42
Pichakron KO, Jelin EB, Hirose S et al. Magnamosis II: Magnetic compression
anastomosis for minimally invasive gastrojejunostomy and jejunojejunostomy. J Am Coll Surg. 2011; 212 (1): 42-9.
43
Wall J, Diana M, Leroy J et al. MAGNAMOSIS IV: magnetic compression
anastomosis for minimally invasive colorectal surgery. Endoscopy. 2013; 45 (8): 643-
8.
44
Gonzales KD, Douglas G, Pichakron KO et al. Magnamosis III: delivery of a
magnetic compression anastomosis device using minimally invasive endoscopic techniques. J Pediatr Surg. 2012; 47 (6): 1291-5.
45
Cope C, Clark TW, Ginsberg G, Habecker P. Stent placement of gastroenteric
This article is protected by copyright. All rights reserved.
Accepted Article
anastomoses formed by magnetic compression. J Vasc Interv Radiol. 1999; 10 (10):
1379-86.
46
Yan X, Fan C, Ma J et al. Portacaval shunt established in six dogs using magnetic
compression technique. PLoS One. 2013; 8 (9): e76873.
47
Zepeda-Gomez S, Baron TH. Benign biliary strictures: current endoscopic
management. Nat Rev Gastroenterol Hepatol. 2011; 8 (10): 573-81.
48
Itoi T, Ishii K, Sofuni A et al. Single-balloon enteroscopy-assisted ERCP in patients
with Billroth II gastrectomy or Roux-en-Y anastomosis (with video). Am J Gastroenterol. 2010; 105 (1): 93-9.
This article is protected by copyright. All rights reserved.
Accepted Article
Figures and legends Figure 1. Indications for use of magnetic compression anastomosis (MCA). The cholangiogram reveals a biliary stricture present after living donor liver transplantation. A, Endoscopy cannot be used to pass a guidewire because of a severe stricture at the site of anastomosis. B, After injection of contrast medium through a percutaneous transhepatic tract, the media cannot drain to the common bile duct because the obstruction is complete. As shown in these cases, MCA can be applied in instances of severe stenosis or complete obstruction through which a guidewire cannot be passed. Figure 2. A cholangiogram showing how magnetic compression anastomosis can be used to treat a biliobiliary stricture developing after living donor liver transplantation. A, After insertion of a percutaneous transhepatic biliary drainage (PTBD) catheter, the tract was dilated to 18 F, and a covered self-expandable metal stent was inserted into the common bile duct (CBD). B, A magnet attached to a polypectomy snare was delivered via endoscopic
retrograde cholangiopancreatography (ERCP) scope through the CBD. C, Another magnet was fixed to alligator forceps and moved toward the anastomosis site through the PTBD tract. The magnets were approximated, and the PTBD catheter inserted. D, After 6 weeks, the approximated magnets were removed by percutaneous transhepatic cholangioscopy through the PTBD tract. E, After the approximated magnets were removed, a retrievable, fully converted, self-expandable metal stent (FCSEM) was inserted for 6 months. F, After 6 months, the FCSEM was removed and a new fistula was formed. Figure 3. A cholangiogram showing how magnetic compression anastomosis can be used to treat a biliobiliary stricture developing after cholecystectomy (adapted from reference 8). A, A percutaneous transhepatic biliary drainage (PTBD) catheter was inserted because the cystic duct site was completely obstructed as a result of ischemic injury. B, A magnet attached to a This article is protected by copyright. All rights reserved.
Accepted Article
polypectomy snare was delivered via endoscopic retrograde cholangiopancreatography (ERCP) scope through a previously inserted a retrievable, fully converted, self-expandable metal stent (FCSEM). C, After the magnets were successfully approximated, a PTBD catheter was inserted and the FCSEM removed. D, After 7 weeks, the approximated magnets were removed by percutaneous transhepatic cholangioscopy through the PTBD tract. E, An FCSEM was next inserted via the ampulla into the CBD and maintained for 6 months. F, After 6 months, the FCSEM was removed; recanalization was complete. Figure 4. A cholangiogram showing how magnetic compression anastomosis can be used to treat a bilioenteric stricture developing after Whipple’s operation with Roun-en-Y
anastomosis to treat pancreatic cancer. A, A percutaneous transhepatic biliary drainage (PTBD) catheter was inserted because the anastomosis site was completely obstructed, and the tract was dilated to 18 F. B, A magnet attached to a polypectomy snare was delivered using a duodenoscope, and another magnet was delivered through the PTBD tract. C, The
magnets were successfully approximated, and a PTBD catheter inserted. D, After 6 weeks, the approximated magnets were removed by percutaneous transhepatic cholangioscopy through the PTBD tract. E, After the approximated magnets were removed, an internal drainage catheter (16 Fr) was inserted and maintained for 6 months. F, After 6 months, the catheter was removed and development of a new fistula was confirmed. Figure 5. Magnetic compression anastomosis using other delivery routes to form a bilioenteric anastomosis. A. An endoscopic approach to the anastomosis site failed because the afferent loop was long. Thus, magnet delivery featured a surgically created skin/intestinal fistula. B, Both the left intrahepatic duct (IHD) and the right IHD were separately anastomosed to the jejunum, and the right IHD was obstructed. Two percutaneous transhepatic cholangioscopy scopes were used to deliver a magnet via the PTBD tract via the This article is protected by copyright. All rights reserved.
Accepted Article
right IHD tract (one scope), and another magnet through the left IHD tract (the second scope), to allow of magnet approximation. Figure 6. A cholangiogram showing failure of magnetic compressive anastomosis (adapted from reference 7). A, The magnet could not be delivered to the stricture because the distal common bile duct was tortuous and angulated (like the letter W; white arrow). B, Between-
magnet attraction across the long stricture (~30 mm) was weak, and the magnets would not come together.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 1. Outcomes of recanalization using magnetic compression anastomosis in biliobiliary stricture.
Year
Authors
Type of report
Age/sex (no.)
Reason for operation (no.)
Previous operation
Type of biliary obstruction
Distance between the two magnets (mm)
Time to removal of magnets (days)
Conduction of anastomosis
Duration of biliary drainage
Compli cation
Follow-up period /symptom
2003
Mimuro et al.34
Case
76/F
Pancreatic cancer
DP
Benign
12
21
Partial
NA
No
NA
2005
Itoi et al.10
Case
76/F
Hilar bile duct cancer
None (radiation)†
Malignant
8
14
Partial
NA
No
NA
2005
Okajima et al.11
Case
44/F
Fulminant hepatitis
LDLT (right lobe)
Benign
2
42
Complete
3 months
No
15 months asymptomatic
2008
Akita et al.36
Case
34/F
NA
LDLT (right lobe)
Benign
2
24
Complete
NA
No
NA
2009
Matsumo et al.38
Case
53/M
NA
LDLT (right lobe)
Benign
2
10
Complete
30 days
No
3 years asymptomatic
2010
Itoi et al.39
Case
60/M
NA
LDLT (right lobe)
Benign
NA
9
Complete
6 months
No
2 months asymptomatic
2011
Itoi et al.40
Case series
40/F
Metastasis live tumor derived from colon cancer
Right three segmental + S3 partial hepatectomy
Benign
15
30
Complete
6 months
No
2 years asymptomatic
2011
Jang et al.7
Original
Mean 53.8 (range 33-64) M:F=9:3
LC (3),HCC (7), HF (1)
LDLT (right lobe)
Benign
NA
Mean 74.2 (range 14-181)
Complete
Mean 183 days (range 99-581 days)
No
Mean 303 days
24/M
NA
8 months
No
2012
Oya
Case
LDLT
This article is protected by copyright. All rights reserved.
Benign
NA
60
Partial
Restenosis (1) NA
Accepted Article et al.13
2014
Jang et al.8
(right lobe) Case series
Case#1: 45/M Case#2: 38/F
#1: Abdominal trauma #2: Cholecystitis
#1: embolization‡ #2 : cholecystecomy
Benign
#1: #2:
4 6
#1: #2:
63 32
Complete
NA
No
#1: 468 days #2: 80 days asymptomatic
M, male; F, female; NA, not available; LC, liver cirrhosis; HCC, hepatocellular carcinoma; HF, hepatic failure; DP, dorsal pancreatectomy; LDLT, living donor liver transplantation. † Bile duct obstruction occurred after intraluminal radiation to treat bile duct cancer. ‡ Bile duct obstruction occurred because of ischemic injury to the bile duct after embolization of the hepatic artery.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 2. Outcomes of recannalization using magnetic compression anastomosis in bilioenteric stricture..
Year
Authors
Type of report
Age/sex (no.)
Reason for operation (no.)
Previous operation
Type of biliary obstruction
Distance between the two magnets (mm)
Time to removal of magnets (days)
Conduction of anastomosis
Duration of biliary drainage
Comp licati on
Follow-up period /symptoms
2001
Takao et al.9
Case
70/M
Gastric cancer
Subtotal gastrectomy (B-II)
Benign
2
32
Complete
90 days
No
2 years
2005
Muraoka et al.35
Case
Case #1: 1/F
#1: Fulminant hepatitis
#1: LDLT (left lobe) with R-Y
Benign
#1:
2
#1:
12
Complete
8 weeks
No
#2:
5
#2:
13
Case#2: 57/M
#2: LC+HCC
#2: LDLT (right lobe) with R-Y
#1: 54 months /asymptomatic #2: 6 days /recurrence and 6 months /asymptomatic
2008
Yukawa et al.37
Case
83/M
Gastric and gallbladder cancer
Distal gastrectomy with R-Y and cholecystectomy
Benign
NA
21
Complete
NA
Slight fever
NA
2009
Avaliani et al.12
Original
Mean 64 (range 4682) M:F=9:25
AOV cancer (7) Pancreatic cancer (21) CCC (6)
None†
Malignant ‡
NA
7-10
Complete (except one case)
No drain
No
Re-intervention (6 cases) after 9.6 months (mean; range 2-36 months)
2010
Suyama et al.14
Case
78/M
Gallbladder cancer
Radical cholecystectomy with R-Y
Benign
NA
34
Complete
6 months
No
NA
2011
Itoi et al.40
Case
60/F
CCC
Expanded left lobectomy with
Benign
2
17
Complete
Not removed
No
NA
This article is protected by copyright. All rights reserved.
Accepted Article 2014
Jang et al.8
R-Y Case series
Case#1: 49/M
#1: Pancreatic NET #2: Choledochal cyst
Case#2: 27/M
#3: Pancreatic NET
Case#3: 63/F
#1: PPPD with HJstomy #2: Excision of cyst with R-Y #3: Whipple operation with R-Y
Benign
#1:
5
#1:
36
#2:
5
#2:
40
#3:
7
#3: 14
Complete
NA
No
#1:
202 days
#2:
1,573 days
#3: 103 days /asymptomatic
M, male; F, female; NA, not available; LC, liver cirrhosis; HCC, hepatocellular carcinoma; AOV, ampulla of Vator; CCC, cholangiocellular carcinoma; NET, neuroendocrine tumor; B-II, Billroth II; LDLT, living donor liver transplantation; R-Y, Roux-en-Y anastomosis; PPPD, pylorus-preserving pancreatico-duodenectomy. † The tumor was not treated with radical surgery. ‡ The obstructions were caused by cancer, but the sites requiring anastomoses in the bile duct and duodenum were benign.
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f1A_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f1B_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f2_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f3_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f4_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f5A_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f5B_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f6A_AA
This article is protected by copyright. All rights reserved.
Accepted Article
den_12319_f6B_AA
This article is protected by copyright. All rights reserved.
