Experimental studies of transgastric gallbladder surgery: cholecystectomy and cholecystogastric anastomosis (videos)
Article Outline
- Abstract
- Material and methods
- Discussion
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- Appendix. Supplementary data
- References
- Copyright
Background
Transgastric flexible endoscopic surgery might offer advantages over open and laparoscopic surgery. The aim of this study was to develop methods for performing transgastric biliary endosurgery.
Methods
Cholecystectomies and biliary anastomoses were performed in 8 anesthetized pigs (27-30 kg) in nonsurvival studies. Two endoscopes passed perorally were inserted through the stomach wall after needle-knife incision. Endoscope-induced pneumoperitoneum allowed viewing and manipulation of the gallbladder with both endoscopes independently. The cystic duct was dissected, clipped, and transected. Cholecystectomy was performed with one of two methods: either by using two endoscopes, or a single endoscope and a 5-mm-diameter grasping instrument inserted transabdominally. Clips and sutures were used to attach the gallbladder to the stomach wall, and an incision was made to form a cholecystogastrostomy. In survival experiments in 8 pigs, transgastric incisions were closed with endoscopic sutures.
Results
The gallbladder was successfully removed in 8 pigs (nonsurvival experiments). The time for the procedure ranged from 2.5 hours to 40 minutes and decreased with experience. At postmortem examination, clips placed on the cystic duct and the artery were secure. An anastomosis was successfully formed between gallbladder and stomach in 3 pigs. In 8 pigs, full-thickness incisions in the stomach wall were closed with two to 4 stitches. All 8 pigs survived (median follow-up, 22 days; range 14-28 days).
Conclusions
Transgastric gallbladder surgery, including cholecystectomy and biliary anastomosis, is feasible. Full-thickness gastric incisions were safely closed in survival studies. The efficacy and the safety of transgastric surgery merits further study.
What is already known on this topic
What this study adds to our knowledge
A peroral transgastric approach to the abdominal cavity by using flexible endoscopes might offer advantages for selected surgical procedures and for patients in whom conventional transabdominal or laparoscopic procedures are unattractive: for example, morbidly obese patients and patients with extensive scars, burns, or infections in the abdominal wall. By using the transgastric route, the abdominal cavity can be accessed from within.
The endoscopes and the accessories used for flexible endoscopic surgery may have some disadvantages compared with the rigid instruments used at open or laparoscopic surgery. In particular, the support from the wall of the GI tract, which allows the successful controlled deployment of accessories through the accessory channel, is absent. The forces that can be exerted by a curved flexible endoscope in an air-filled abdominal cavity and by currently available flexible devices passed through the accessory channel might impose severe restrictions on access, control, and, especially, retraction of organs that obstruct the view of the surgical field.
To explore the possibilities and the limitations, a study was conducted of the feasibility of operating on the gallbladder by using a transgastric approach. Cholecystectomy was chosen, because it is a standard, widely performed operation of moderate complexity. It requires adequate exposure, controlled dissection, and effective hemostasis. The aim of this study was to test the feasibility of performing cholecystectomy and gallbladder anastomosis by using a transgastric approach at flexible endoscopy.
Material and methods
Transgastric biliary procedures were performed in 8 pigs (27-35 kg) in nonsurvival studies. Survival studies of full-thickness gastric incisions, with subsequent sutured closure, were carried out in 8 animals. The protocol for these studies was reviewed and approved by ethical review boards in the two institutions (Goteborg, Sweden, London, UK), where the experiments were performed. A Home Office (a government regulatory authority in the United Kingdom) project license was approved for these experiments. The transgastric biliary procedures were performed in Sweden; the survival studies of full-thickness gastric incisions were carried out in London. Three of the operators who performed the procedures were surgeons experienced in laparoscopic cholecystectomy, and two were gastroenterologists. All had extensive experience with flexible endosurgical techniques.
Equipment
For some procedures, two endoscopes (either two GIF Q140, or a GIF Q140 plus a GIF 2T160; Olympus, Solna, Sweden) were passed into the esophagus and then into the peritoneal cavity through a transgastric incision. In other procedures, a double-channel endoscope (GIF 2T160; Olympus) was used without the second endoscope. For one cholecystectomy, a double-channel endoscope was used, together with a 5-mm diameter, rigid forceps retractor inserted transabdominally.
Numerous commercially available flexible endoscopic accessories were used and tested in the course of the study, including needle knives, papillotomes, guidewires, balloons, biopsy forceps, grasping forceps (all Microvasive Endoscopy, Boston Scientific Corp, Natick, Mass); “hot” biopsy forceps, grasping forceps, snares, suture cutters, scissors, stone baskets, and endoclips (all Olympus); Tri-clips and a 19-gauge EUS (Echotip) needle (all Wilson-Cook Medical Inc, Winston-Salem, NC); tags and polypropylene thread (retrieved from Endocinch kits), and locking devices (all Bard Interventional Products, Billerica, Mass).
Some new instruments, which are not commercially available, developed for other procedures, also were used. This nonstandard equipment included tags and thread, locking and thread-cutting devices, and devices for a new suturing method.9 A few new instruments were developed specifically for this study, including a needle-knife/guidewire combination device.
Two videoprocessors and monitors (Olympus) were used to acquire and display the images from both endoscopes. Initially, continuous recordings from both endoscopes were made by using a standard videotape format (SVHS); later the recording format was changed to digital. A videorecorder placed in the operating theater recorded the actions of the endoscopic surgeons. A commercially available endosurgical generator (Valleylab, Boulder, Colo) was used for needle-knife dissection and electrocoagulation.
Surgical technique
The surgical technique is illustrated in Appendix, Appendix, Appendix, Appendix, Appendix, Appendix, Appendix, Appendix, Appendix, Appendix, Appendix (online at www.mosby.com/gie). The methods studied included esophageal intubation with two flexible endoscopes, gastric incision, and penetration into the peritoneal cavity with one or two endoscopes. The air-insufflation mechanism of the endoscope systems was used to induce and to maintain the pneumoperitoneum. For cholecystectomy, the necessary surgical steps included the following: identification and exposure of the gallbladder; grasping and manipulation of the gallbladder to expose the cystic duct and artery, to allow dissection; hemostasis, including clipping of the cystic artery, clipping the cystic duct, dissection of the gallbladder from its bed; and transgastric/transesophageal removal of the gallbladder. For the cholecystogastric anastomosis, the gallbladder had to be grasped securely and pulled through a gastric incision for subsequent clipping or suturing to the gastric deep muscle, followed by incision of the gallbladder to complete the anastomosis.
The mucosa and the deep muscle of the stomach wall were incised by using a needle knife. The electrosurgical current was switched off as soon as the needle was felt to have penetrated the deep muscle; the catheter of the needle knife was pushed forward into the peritoneal cavity as the needle was withdrawn. A 0.035-inch guidewire (JAG wire; Microvasive Endoscopy) was inserted into the peritoneal cavity. In a majority of cases (n
=12), an 18-mm-diameter through-the-scope dilation balloon was passed over the guidewire and through the incision. The balloon was inflated to distend the muscle layer, and, by pushing the endoscope forward against the inflated balloon, the endoscope passed through the muscle layer into the peritoneal cavity. The insertion procedure was repeated with the second endoscope by using the intraperitoneal view provided by the first endoscope to assist passage of the second endoscope (Fig. 1A). It sometimes was helpful to make a cruciate incision in the gastric mucosa before penetrating the muscle, because this reduced mucosal drag on the balloon and the endoscope, as these were passed through the stomach wall. A sphincterotome was used for some transgastric incisions (n=4), which made use of the balloon unnecessary.
Figure 1.
A, Endoscopic view of second endoscope (guidewire in accessory channel) passing through stomach wall into peritoneal cavity parallel to falciform ligament. Insertion tube of first endoscope is seen at top, with small intestine above and colon to right and left. B, Endoscopic view of exposed gallbladder with grasping forceps applied to body to improve exposure. A snare grasping fundus of gallbladder (curving toward diaphragm) is ineffective in pushing it upward to improve exposure.
Passage of a balloon over a guidewire through the wall of the stomach tends to split the circular and the longitudinal muscle without cutting through the muscle. This allows the endoscope, if pressed against the balloon, to pass through the wall of the stomach. Sometimes this did not work well, because the accessory channel is not positioned centrally within the insertion tube of the endoscope. Thus, the balloon always is eccentric to the central axis of the insertion tube, and, with some angles of approach, the endoscope may catch on the stomach and not follow through into the peritoneal cavity.
The balloon method has the advantage (or the disadvantage) that when the endoscope is withdrawn, the muscle layers tend to spring back together, partially closing the defect. It can be difficult to get back through this defect with the endoscope. The sphincterotome method was found to be quicker, and, because the muscle is cut, there is no tendency for the hole to close spontaneously. If speed of entry and repeated crossings of the stomach wall are required, the sphincterotomy method was considered advantageous. But if this method is chosen, then secure methods of closure are of greater importance than with the balloon method.
An experimental guidewire was made by removing the coating from the thin nitinol tip, thereby exposing the wire tip for use as a needle knife; a portion of the coating on the wire that would be external to the endoscope also was removed so that a connection could be made to the electrosurgical generator. Creation of an incision by using this wire with either a balloon or a sphincterotome worked well and shortened the time required to pass the endoscope through the stomach wall.
Some care was taken when using transillumination and abdominal compression to penetrate the stomach anteriorly; in early experience, the lesser sac was entered on two occasions. The sites chosen for the gastrostomy for the gallbladder work were in the body or the antrum. It was helpful to enter the peritoneal cavity a few centimeters away from the gallbladder instead of immediately adjacent to it. The extra distance appeared to allow better access to the gallbladder in a recurved position, and the relatively mobile body and antrum allowed the gallbladder to be easily dragged through the gastric incision to make the cholecystogastric anastomosis.
It was possible to find the gallbladder quickly in all of the animals. Exposing the cystic duct was more difficult. In part, this was because of the pig anatomy: 4 floppy hepatic lobes tend to cover the gallbladder when viewed from below with a pneumoperitoneum. Retraction of these lobes by pushing on catheters or instruments, in general, was ineffective. Pushing highly flexible instruments against the liver tended to cause them to buckle or the endoscope actually moved away from the liver (Fig. 1B). It was possible to expose the cystic duct and artery for dissection by placing the endoscope as close as possible to the gallbladder, followed by grasping the gallbladder with two forceps, and then moving each forceps alternately upward toward the cystic duct. It was possible to do this with two endoscopes, with a single double-channel endoscope, and with a single endoscope and a 5-mm-diameter transabdominal grasper.
Forceps were reasonably effective for grasping the gallbladder (Fig. 2A); this did not cause bile leaks; but, occasionally, a forceps slipped off and repositioning was required. Manipulation of the gallbladder with snares also was studied. Snares were especially effective for pulling the gallbladder into the stomach to form a cholecystogastric anastomosis but were less useful for dissection. It was possible to “exchange” the gallbladder between two transgastric endoscopes by passing the organ from one grasper or snare in one endoscope to an instrument inserted through the other endoscope. It also was possible to expose the cystic duct through a double-channel endoscope by alternately grasping the gallbladder with two flexible graspers; pulling downward on one allowed the other to be placed in a higher position.
Figure 2.
A, Endoscopic view showing cystic duct grasped with forceps (yellow sheath) and attachments of gallbladder to liver being severed with a needle knife. Clips have been applied to cystic duct and artery. B, Endoscopic view showing last attachment of gallbladder to liver being severed with needle-knife. Grasping forceps are holding neck of gallbladder.
Dissection for cholecystectomy was carried out by using a needle knife while the gallbladder was grasped with forceps (Fig. 2B). The utility of other devices for blunt dissection was tested, including endoscopic scissors and a suture cutter, but none were strong enough to be useful during dissection.
Occasionally, bleeding was encountered from the cystic artery in the course of the dissection. This either stopped spontaneously or was managed by grasping the artery and squeezing it for awhile, followed by placement of endoscopic clips (2-6 clips were applied per animal).
Once the cystic duct and artery were dissected free and were clipped, the gallbladder was pulled upward to expose the gallbladder bed, and the gallbladder then was separated from the liver by using needle-knife dissection. When it was free from the liver, the gallbladder was grasped with a snare, pulled back through the incision in the stomach wall, and removed via the esophagus (Fig. 3A).
Figure 3.
A, Gallbladder bed after transgastric removal of gallbladder; a clip is present on cystic duct and artery. B, Gallbladder after removal.
The time for the procedure ranged from 2.5 hours to 40 minutes. The time required decreased with experience.
Postmortem examination revealed that the clips on the cystic duct and artery were secure and that neither bile nor blood had leaked from this site (Fig. 3B). The gallbladder bed was dry.
An anastomosis was formed between the gallbladder and the stomach in 3 pigs. The gallbladder was pulled through the gastric incision (Fig. 4). It was stitched to the wall of the stomach in two experiments, and, in a third, it was attached to the stomach wall with clips (Olympus). The anastomosis was completed by incising the intragastric portion of the gallbladder with a needle knife.
Figure 4.
Endoscopic view of gallbladder being pulled through stomach wall to form an anastomosis.
In separate survival experiments in 8 pigs, full-thickness incisions in the wall of the stomach were closed with two to 4 stitches placed with a new needle, tag thread, and locking method that could be performed through a 2.8-mm-diameter accessory channel. All 8 pigs survived these closure experiments and were followed for a median of 22 days (range 14-28 days).
Discussion
This study confirms the feasiblity of removing the gallbladder by using flexible endoscopes passed through the wall of the stomach. It also shows that a cholecystogastric anastomosis can be made with a similar transgastric approach. Separate survival experiments confirmed that transgastric incisions can be closed successfully.
The introduction of two endoscopes into the stomach usually was straightforward. The use of a guidewire and lubrication was found to assist the introduction of the second endoscope. With one endoscope in the esophagus, the second was lubricated and placed in the pharynx. Then, an 0.035-inch guidewire (Jagwire; Microvasive Endoscopy) 450 cm in length was passed beside the first endoscope and, under direct vision, into the esophagus. Gentle pressure then was exerted to advance the second endoscope into the esophagus, which also advanced the first endoscope. Because the pigs had an endotracheal tube in place, as well as the first endoscope, both of which might rub against the second endoscope, the frictional resistance to insertion of the second endoscope was extremely high. It can be difficult to intubate the esophagus of the pig, which has a high pharyngeal pouch that might be perforated if the tip of the endoscope does not enter the esophagus. There was no evidence of esophageal trauma at the conclusion of the experiments.
The present study also revealed some limitations to the use of flexible endoscopes and instruments. The main problem encountered was the difficulty in exerting sufficient forward force to expose the biliary anatomy. When the endoscope is held in a retroflexed position unsupported in a pneumoperitoneum, there is a tendency for a forward force exerted on the insertion tube of the endoscope to move the endoscope away from the operative site. Another difficulty is in the reduced “triangulation” imposed by the use of instruments through one or two accessory channels in a flexible endoscope. At laparoscopy, the use of ports at different abdominal sites allows access from different angles and thereby facilitate the approach to the surgical field from different directions and passage of tissue or surgical instruments, e.g., a threaded needle, from one instrument to another.
Occasionally, the peritoneal cavity became overinflated and air had to be aspirated, either by applying suction with the endoscope or by passing a needle through the abdominal wall. A means of controlling or measuring intra-abdominal pressure through the flexible endoscope would be useful. Moving flexible instruments inserted through a double-channel endoscope apart and together is difficult. Many, but not all, of the limitations of the instrumentation that were used for the transgastric surgery could be overcome by innovative design and engineering improvements.
The concepts of “incisionless” surgery and a transgastric approach to intra-abdominal surgery are both attractive and alarming. The absence of an abdominal incision is likely to be attractive to patients. However, the proximity and the ease of access of certain organs via the stomach wall would likely be the reason for a potential for growth of transgastric surgery. Although most endoscopists regard perforation of the wall of the GI tract as a disaster, transgastric flexible endoscopic surgery is not new. Percutaneous endoscopic gastrostomy has been performed for 25 years,10 endoscopic pseudocyst drainage for 20 years,11 and transgastric retroperitoneal necrosectomy for 4 years.12 Needle puncture of structures through the stomach wall at EUS was introduced over 12 years ago.13 This procedure seems safe and useful, and has been used for transgastric endosurgical anastomosis of the gallbladder to the stomach.14 Kalloo et al.15 and Kantsevoy et al.16 used a peroral transgastric route in a porcine model for diagnostic peritoneoscopy, liver biopsy and gastrojejunostomy, and for tubal ligation, respectively.
Endoscopes have intentionally and safely been inserted into the peritoneal cavity in the course of surgical procedures, including enteroscopy for obscure bleeding,17 some bariatric procedures,18 and gastric resection.19 Gastric perforations during EMR have been successfully closed by using clips at flexible endoscopy without resorting to an abdominal incision.20
Although transgastric biliary surgery might prove to be useful in selected cases, it is likely that bariatric surgery may be the forcing ground for the development of transgastric surgical methods.
The stomach is a reasonably forgiving organ if perforated. Its thick muscular layer tends to contract and close the hole through which an endoscope penetrated. Moreover, the adjacent mesentery tends to plug leaks. However, the risk of infection, peritonitis, and adhesions, and problems consequent to such complications must be addressed in subsequent studies.
It is unclear at present how valuable transgastric surgical procedures will be when compared with laparoscopic or open procedures. The risks of peritoneal infection and the technical limitations of working through flexible endoscopes may be important barriers to the dissemination of this approach. The relative ease of access; the adjacency of relevant organs; and the attractiveness of incisionless, minimally invasive surgery may offer real advantages to patient care. Even if transgastric biliary surgery proves to be of little clinical value, the experience with closure of full-thickness incisions and perforations may be useful.
The present study demonstrates that a commonly performed surgical procedure of moderate complexity can be completed via a transoral-transgastric route. It also demonstrates certain limitations of the standard devices used with flexible endoscopes for this type of work. The design of many of these needs to be improved. This present work gives encouragement to further experimental studies of transgastric surgery.
Appendix. Supplementary data
Video-clip 1 A needle knife penetrates the wall of the stomach and a guidewire is passed into the peritoneal cavity.
Appendix. Supplementary data
Video-clip 2 A modified needle knife guidewire and sphincterotome combination is used to cut a hole large enough to allow the endoscope to enter the peritoneal cavity.
Appendix. Supplementary data
Video-clip 3 A balloon over a guidewire is used to dilate a track in the stomach wall to allow the endoscope to enter the peritoneal cavity.
Appendix. Supplementary data
Video-clip 4 A second endoscope is following a guidewire into the peritoneal cavity.
Appendix. Supplementary data
Video-clip 5 A snare and forceps are used to grasp the gallbladder.
Appendix. Supplementary data
Video-clip 6 Dissecting the cystic duct.
Appendix. Supplementary data
Video-clip 7 Clipping the cystic duct.
Appendix. Supplementary data
Video-clip 8 Cutting the cystic duct.
Appendix. Supplementary data
Video-clip 9 Dissecting the gallbladder from the liver.
Appendix. Supplementary data
Video-clip 10 The gallbladder has been freed from the liver and is pulled through the stomach wall.
Appendix. Supplementary data
Video-clip 11 The gallbladder is removed through the mouth.
References
- . Lost among giants. J Invest Surg. 1999;12:241–243
- . Ein fall von exterpation der gallenblase wegen chronischer cholelithiasis: heilung. [German] Klin Wochenschr. 1882;19:725–727
- . The Langenbuch paper. I. An historical perspective and comments of the translators. Gastroenterology. 1983;85:1426–1433
- . Cholecystotomy and cholecystectomy. Medical News. 1885;46:472–473
- . Laparoscopic biliary surgery. Oxford: Blackwell Science; 1992;p. vii–ix
- . Die erste cholezystektomie durch das laparoscop. [Abstract in German] Langenb Arch Klin Chir. 1986;369:804
- . The first laparoscopic cholecystectomy. JSLS. 2001;5:89–94
- . Profiles in laparoscopy: Mouret, Dubois, and Perissat: the laparoscopic breakthrough in Europe (1987-1988). JSLS. 1999;3:163–167
- . A through-the-scope device for suturing and tissue approximation under endoscopic ultrasound control. Gastrointest Endosc. 2002;56:737–742
- . Gastrostomy without laparotomy: a percutaneous endoscope technique. J Pediatr Surg. 1980;15:872–875
- . Endoscopic drainage of pancreatic pseudocysts. Gastrointest Endosc. 1985;31:322–327
- . Retroperitoneal endoscopic debridement for infected peripancreatic necrosis. Lancet. 2000;356:653–655
- Endoscopic ultrasonography as an adjunct to fine needle aspiration cytology of the upper and lower gastrointestinal tract. Gastrointest Endosc. 1992;38:35–39
- . Transluminal endosurgery: single lumen access anastomotic device for flexible endoscopy. Gastrointest Endosc. 2003;58:585–591
- Flexible transgastric peritoneoscopy: a novel approach to diagnosis and therapeutic intervention in the peritoneal cavity. Gastrointest Endosc. 2004;60:114–117
- Per-oral transgastric endoscopic ligation of fallopian tubes with long survival in a porcine model. [abstract] Gastrointest Endosc. 2004;59:AB112
- . Intraoperative endoscopy. Gastrointest Endosc Clin N Am. 1999;9:61–69
- . Laparoscopic gastric bypass surgery: equipment and necessary tools. J Laparoendosc Adv Surg Tech A. 2003;13:241–245
- Combined endoscopic intragastric resection of a posterior stromal gastric tumor using an original technique. Surg Endosc. 2002;16:537
- Endoscopic closure of perforations caused by EMR in the stomach by the application of metal clips. Gastrointest Endosc. 2003;75:949–951
This work was presented to the American Society of Gastrointestinal Endoscopy at Digestive Diseases Week, May 15-20, 2004, New Orleans, Louisiana (Gastrointest Endosc 2004;59:AB 113).
PII: S0016-5107(04)02774-9
© 2005 American Society for Gastrointestinal Endoscopy. Published by Elsevier Inc. All rights reserved.
