88 Obesity Surgery, 13, 2003 © FD-Communications Inc.

Obesity Surgery, 13, 88-94

Background: Creating the proximal anastomosis inlaparoscopic biliopancreatic diversion with duodenalswitch (LBPD-DS) and Roux-en-Y gastric bypass(LRYGBP) is a critical step in ensuring the success ofthe procedures. The aim of this study was to assessthe safety and efficiency of performing this anasto-mosis using a flexible, computerized, circular staplingdevice.

Method: We prospectively monitored the use of anewly FDA-approved stapling device (SurgASSIST,Power Medical Intervention) for the construction ofthe proximal anastomosis by a variety of approachesand reviewed the charts of 10 patients.

Results: We successfully constructed 9 out of 10proximal anastomoses: 2 gastro-jejunostomies and 7duodeno-ileostomies, without any signs of leakage. In2 patients, the stapling technique involved a transpy-loric instrumental passage; both were complicated bythe difficulty to pass either the flexible scope or theanvil through the narrow pyloric lumen. In 5 patients,the anvil was placed directly through a duodenotomyand no technical problems were encountered. Themedian time for performing the proximal anastomosiswas 19 minutes (range 9-55). There were no postoper-ative complications in any patients.

Conclusions: Stapling using the SurgASSIST wasfeasible and safe for performing laparoscopic anasto-moses in bariatric bypass procedures. A duodenoto-my for direct placement of the anvil into the postpy-loric region seems to be most feasible for duodeno-ileostomies, while transoral passage of the anvil canbe recommended for gastro-jejunostomies. In its cur-rent form, we do not recommend transoral placementof the flexible shaft of the SurgASSIST device. Furtherclinical trials need to be performed for comparisonwith existing devices.

Key words: Flexible circular stapling, computerized sta-

pling technique, gastrointestinal anastomosis, biliopan-creatic diversion, Roux-en-Y gastric bypass, bariatric sur-gery, morbid obesity, laparoscopy

Introduction

Laparoscopic gastric bypass procedures for treatingmorbid obesity have evolved significantly duringthe last decade.1,2 With the advantages of minimallyinvasive techniques, the overall complication ratehas decreased, but technical complications such asleakage and bleeding from the proximal anastomo-sis still occur. Since a laparoscopic hand-sewn anas-tomosis is technically demanding and time-consum-ing, most surgeons prefer to use a stapled technique.

The circular end-to-end anastomotic device(EEA) originally designed for open surgery hasbeen reported to be suitable for this part of laparo-scopic bypass procedures. Still, there is a risk ofbleeding from the anastomotic site, and the leakagerates are as high as 1-3%.1-5

These complications may be life-threatening inthe morbidly obese patient, and technical improve-ments in stapling devices may decrease the rate ofthe anastomotic failures. The aim of this study wasto assess the safety and efficiency by performing theproximal anastomosis using a newly FDA-approvedflexible computer-mediated circular stapling device.

Methods

This prospective pilot study was performed fromDecember 2001 through January 2002, and includ-

Early Experience With Computer-Mediated FlexibleCircular Stapling Technique For UpperGastrointestinal Anastomosis

Anne Waage, MD; Michel Gagner, MD, FRCSC; FACS, John J. Feng, MD

Mount Sinai School of Medicine, Department of Surgery, Division of Laparoscopic Surgery, NewYork, NY, USA

Reprint requests to: Michel Gagner, MD, FRCS, FACS, MountSinai School of Medicine, Department of Surgery, MinimalInvasive Surgical Center, 5 East 98th St., Box 1103, 15th Floor,New York, NY 10029, USA.E-mail: michel.gagner@mountsinai.org

ed 10 laparoscopic intestinal bypass-proceduresperformed by one surgeon. In all the patients, theproximal anastomosis was attempted using theSurgASSIST, a newly FDA-approved circular, flex-ible computer-mediated stapling system (PowerMedical Interventions, New Hope, PA, USA).

The SurgASSIST system (Figure 1A-C) consistsof a Power Console (1), Flex shaft (2), RemoteControl Unit (RCU) (3), and Digital Loading Unit(DLU) (4). Currently, DLUs are available in 4 cir-cular sizes: 21 mm, 25 mm, 29 mm and 33 mm.

To operate the system, both the Flex shaft andRCU are connected to the Power Console. Afterturning on the power, the system prompts the sur-geon to press the “on” button on the RCU. Once thepower button is depressed, the system goes througha hardware check, ensuring proper functioning ofthe Power Console. At this point, the system asksthe surgeon to attach a new DLU. The DLU isattached to the Flex shaft, which is 14 mm in diam-eter and 2 meters in length. The Power Consoleidentifies the type of DLU attached to the Flex shaftand sends a ready signal. This notification is vocal-ized by the Power Console, as well as displayedvisually on the LCD screen. The position of theDLU can be controlled by depressing one of fourdirectional arrows on the RCU. Not only does theRCU steer the Flex shaft 90° in either direction, butalso individual “Open” and “Close” buttons respec-tively control the separation and approximation ofthe DLU anvil and cartridge, to allow proper sta-pling with the “Fire” key. If the DLU is not proper-ly aligned or approximated, an alarm alerts the sur-geon that the DLU is out of stapling range. The cir-cular DLUs have a range of staple closure of 1.5mm to 2.0 mm, which is the internal measurementof a closed staple. Once the system has been fired,the anvil of the DLU is opened to a predeterminedheight, to facilitate simple removal of the expendedcartridge system from the anastomosis.

Operation Technique

All of the procedures were performed bylaparoscopy. In two of the eight laparoscopic bil-iopancreatic diversion with duodenal switch(LBPD-DS) patients, a second stage (duodeno-ileostomy with ileo-ileostomy) was only performed,because the restrictive part of the procedure (sleeve

Obesity Surgery, 13, 2003 89

Computer-Mediated Flexible Circular Stapling for Upper GI Anastomoses

Figure 1. A. SurgASSIST: 1=Power Console, 2=Flex,3=Digital Loading Unit (DLU), 4= Remote Control Unit(RCU); B. DLU; C. RCU.

A

B

C

gastrectomy) had already been done on a prior date.Two laparoscopic Roux-en-Y gastric bypass(LRYGB) procedures were performed.

The distal entero-enterostomy was completedside-to-side by a standard laparoscopic staplingtechnique using Endoscopic GIA (US surgical,Norwalk, CT) in all patients. For the BPD-DS pro-cedure, the placement of the trocar and the restric-tive part of the procedures were performed by thestandard technique for sleeve gastrectomy, by verti-cal resection of the stomach and transection of theduodenum using the Endoscopic GIA.6

The LRYGBPs were performed according to stan-dard laparoscopic techniques, creating the pouch bytransection of the stomach using the GIA.

For the proximal anastomosis, different approach-es using SurgASSIST were assessed after detachingthe anvil from the DLU (Table1).

Attachment of the anvil to a nasogastric tube andextracting the tube through a proximal duodenal orgastric staple-line allowed transoral passage of theanvil in three patients. By opening a 1-2 cm lengthof the duodenal postpyloric staple-line, the anvilwas inserted directly into the duodenum in five ofthe BPD-DS cases (Figure 2A and B). In both ofthese approaches the detachable trocar of theSurgASSIST was attached to the anvil, therebyfacilitating perforation of the anterior duodenalwall. In one BPD-DS, the anvil was entered into thesmall bowel lumen and perforated the anti-mesen-

teric wall of the ileum by using the trocar spike(Figure 2C).

After anvil positioning, the Flex shaft with theDLU was passed transabdominally, after removal ofthe upper left trocar. Depending on the approach ofanvil placement, the Flex shaft was entered eitherintraluminally into the small bowel or through thestomach wall (Figure 2E), thereby performing anend-to-side anastomosis. Using the RCU to proper-ly adjust Flex shaft position, the DLU was thenattached to the anvil, and the anastomosis was com-pleted by the computer-mediated control system(Figure 2F).

After performing the anastomosis, the resectedanastomotic rings were inspected for any disconti-nuity. In one case, an attempt to pass the Flex shaftwith the DLU transorally was unsuccessful.

In all procedures, methylene blue dye (60-120 cc)was inserted transorally after clamping the smallbowel distal to the anastomosis, in order to detectany leakage of the proximal anastomosis intraoper-atively.

The variables analyzed included the time con-sumed for performing the anastomosis as well asany intraoperative complications or instrument fail-ure. All patients were followed for 3 months post-operatively to evaluate any complications.

An unpaired Student’s t-test was used for statisti-cal analysis.

90 Obesity Surgery, 13, 2003

Waage et al

Table1. Proximal anastomosis performed by SurgASSIST

Patient Procedure Site of Insertion Time for anvil Time for IntraoperativeNumber performed Anvil/Flex shaft insertion (min) Anastomosis complications

performance* (min)

1 BPD-DS Transoral /ileum 30 15 Pyloric dilatation necessary2 BPD-DS ileum /transgastric 6 41 Pyloric dilatation necessary3 BPD-DS ileum /failed transoral - - Tearing of cervical

esopageal mucosa4 BPD-DS II Duodenotomy /ileum 4 5 -5 BPD-DS Duodenotomy /ileum 7 6 -6 BPD-DS II Duodenotomy /ileum 8 10 -7 RYGBP Transoral /jejunum 10 45 Anvil not attaching/anvil

slipping off8 RYGBP Transoral /jejunum 7 12 Anvil slipping off9 BPD-DS Duodenotomy /ileum 8 12 -10 BPD-DS Duodenotomy /ileum 6 10 -

*After inserting the anvil

Obesity Surgery, 13, 2003 91

Computer-Mediated Flexible Circular Stapling for Upper GI Anastomoses

Figure 2.A. Anvil entering duodenum after opening of

proximal staple-line.B. Closure of the duodenotomy using Endo-

GIA.C. Anvil with spike perforating wall of small

bowel.D. SurgASSIST Flex shaft and Digital Loading

Unit entered transgastrically.E. Attachment of anvil to the Digital Loading

Unit.

A B

C D

E

Results

The average age of the patients was 42.3 years(range 31-55 years). The average body mass index(BMI) was 51.4 kg/m2 (range 44.2-59.0). Five of thepatients had co-morbidities: sleep apnea, hyperten-sion, venous thrombosis or diabetes.

Nine of ten proximal anastomoses were success-fully performed by laparoscopy using differentapproaches for SurgASSIST (Table 1). None ofthese showed any signs of leakage intraoperativelyusing blue dye testing. The 21-mm DLUs were usedin all procedures. In one case, the attempt to intro-duce the Flexible Shaft with the DLU transorallyfailed because of esophageal resistance, and thisprocedure was completed as a laparoscopic BPD-DS using an endoscopic end-to-end anastomotic sta-pling device (EEA, US Surgical). In this patient,intraoperative endoscopy revealed superficial, lineartearing of the esophageal mucosa. This patient wasdischarged from hospital 2 days postoperatively anddid not develop any postoperative complications.

In one case of BPD-DS, the anvil was passed tran-sorally into the duodenum, but due to the narrowlumen of the pylorus, the anvil was unable to passinto the duodenum without a dilatation being per-formed.

For the two LRYGBP patients, transoral passageof the anvil was done without any resistance in theesophagus. In five BPD-DS patients, a transduode-nal approach for the anvil was chosen and the anas-tomosis was performed without any technical com-plications. In one BPD-DS case, the DLU connect-ed to the Flexible Shaft was inserted into the stom-ach through a gastrotomy after transjejunal place-ment of the anvil. To allow DLU passage throughthe pylorus, a transgastric dilatation of the pyloruswas required. The median operation time for per-forming the anastomosis by SurgASSIST was 19minutes (range 9-55 minutes). For the BPD-DS pro-cedures, the time to perform the anastomosis wassignificantly shorter when transpyloric passage ofthe anvil or Flex shaft was not involved (13 versus46 minutes, P<0.05)

One technical failure was noted, in which theSurgASSIST device refused to approximate theanvil with the DLU, and a new anvil had to beinserted, thereby successfully completing the proce-

dure. In two cases, the anvil was detached from theDLU after firing the anastomosis and had to beremoved from the proximal duodenum by a laparo-scopic grasper.

The average postoperative hospital stay was 2.8days (range 1-3). There were no postoperative com-plications in the group. No patients necessitated anupper gastrointestinal x-ray for possible leaks dur-ing the hospital stay or in the follow-up period. At 3months follow-up, no symptoms were reported thatwould indicate stenosis of the anastomosis in anypatient.

Discussion

The only effective treatment to achieve sustainedweight loss for patients with morbid obesity is sur-gery.7

Weight reduction is created by a restrictive proce-dure on the stomach and/or intestinal bypass.1-8

Isolated restrictive procedures (such as gastric band-ing) appear to have been less promising for weightloss on long-term follow-up.9 The most effectiveprocedure for bariatric surgery in North Americahas been the RYGBP, in which a small stomachpouch is created and anastomosed to a proximaljejunal segment.3-5 For super-obese patients, weightloss is improved by either a long-limb RYGBP or aBPD with or without duodenal switch (BPD-DS),which includes a sleeve gastrectomy in addition to aduodeno-ileostomy. The pylorus is preserved, and amajor segment of the small intestine is bypassed.6,8

Although effective in providing long-term weightloss and reversal of co-morbidities, the RYGBP per-formed via an open approach is associated with ahigher risk of incisional hernias and wound compli-cations.10

Laparoscopic bariatric procedures have evolvedduring the last decade. There is a need for continu-ous improvements on both laparoscopic techniquesand equipment, to avoid technical complicationssuch as anastomotic leakage and bleeding, becausethe consequences of these can be fatal in morbidlyobese patients.

Stapling as a substitute for suturing is widely pop-ular, especially in gastrointestinal tract surgery.Despite the technical ease with which the anasto-

92 Obesity Surgery, 13, 2003

Waage et al

mosis can be performed with stapling instruments,the complications of leakage, stricture and bleedinghave not been eliminated. A number of studies havereported complication-rates associated with stapledversus sutured anastomosis, but no appreciable dif-ferences have emerged.11-15

The SurgASSIST device introduces a new con-cept to the stapled anastomosis. Connecting the sta-pling device to a Flex shaft similar to a flexibleendoscopy facilitates insertion of the instrumentinto the lumen of the gastrointestinal channel forcreation of a stapled anastomosis in both open andlaparoscopic procedures. However, as demonstratedin one of our patients, despite the flexibility andsmaller 21-mm diameter of the Flex shaft, the non-flexible metallic DLU limits safe passage of thedevice into the smaller lumen of an organ, such asthe esophagus. In the current form of theSurgASSIST instrument, there is no light source orcamera attached to the instrument. Attempting blindintroduction of the non-flexible DLU through thenarrow-lumen esophagus is a potential hazard thatmay create lesions in the esophageal wall or evenperforate the organ.

The other technical problem encountered duringthis trial was due to the fixed angle between themushroom and the shaft of the anvil. This design iscontrary to the EEA anvil, which flexes after firing,thereby facilitating its intraluminal removal after fir-ing the staples. By pre-folding before entering thelumen, the EEA anvil also allows a smoother pas-sage through a narrow lumen such as the pylorusthan the SurgASSIST. This problem is demonstrat-ed in two patients in whom the anvil loosened fromthe DLU upon attempting removal after firing. Inaddition, the smooth surface of the top of the anvilmushroom makes it difficult to grasp and positionthe anvil using traditional laparoscopic instruments.We also had some problem attaching the anvil to theDLU, because precise positioning of the anvil usinga laparoscopic grasper required delicate handling ofthe hollow anvil shaft, to avoid narrowing and dam-aging the shaft.

In some patients, there were difficulties inmanoeuvring the flexible instrument through theabdominal wall which was quite thick in somepatients. This difficulty would suggest the need fora stiff outer shaft to guide the Flex shaft into theactual intestinal lumen. Leakage of the pneumoperi-

toneum through the wound around the Flex shaftalso created difficulty in sustaining proper insuffla-tion for the proximal anastomosis. Currently, thereare no trocar ports that can accommodate theSurgASSIST device that would seal the leakage andavoid wound contamination.

In general, we found the SurgASSIST systemeasy to handle. The RCU in connection with theFlexible shaft determines the optimal position of theDLU to approximate the anvil. The computer-assist-ed firing system eliminates any potential technicalerrors that would involve anvil misfire and incorrectposition. The computerized sensor of the stapler-anvil distance properly determines the thickness ofthe intestinal walls for approximation for anasto-moses. This system thereby uniquely provides a pre-cise, reproducible staple-line by digitally establish-ing stapler height. In a preclinical non-survivalporcine model study, a SurgASSIST-performedanastomosis was stronger compared to another cir-cular end-to-end device (FDA 510(k)). In our pilotstudy, all of the created anastomoses were patent astested intraoperatively, and there was no need forextra reinforcement or repair sutures at the anasto-motic site. No defects were found in the resectedanastomotic rings upon direct inspection.

For BPD-DS, the recommended approach forSurgASSIST stapling is direct implantation of theanvil through the duodenal stapling-line and perfo-ration of the anterior duodenal wall with the anvilspike. Average time for applying the anvil and per-forming the anastomosis in this manner was down to16 minutes. There were no problems by placing theanvil transorally in the LRYGBP group.

Unlike comparable endoscopes, the SurgASSISTFlex shaft does not contain any fibre-optic bundlesand cannot be damaged as such.

In its current form, we were dependent on gassterilization of the device, but this is being changedto allow steam sterilization in late 2002.

Concerning cost, the disposable part of the instru-ment, or DLU, each cost $360 US, which is compa-rable to a disposable circular stapler. The capitalprice for the system itself including the PC, Flexshaft and RCU approximates $50,000 US.

As with all digital instruments, the surgeon mustbe aware of the technical limitations and preciselyfollow device instructions. We recommend notattempting transoral or transpyloric passage the

Obesity Surgery, 13, 2003 93

Computer-Mediated Flexible Circular Stapling for Upper GI Anastomoses

94 Obesity Surgery, 13, 2003

Waage et al

Flexshaft/DLU in its current form. SurgASSIST introduces a new technological plat-

form in surgery. The flexible design has the advan-tages of allowing the surgeon to reach and repairanatomy inaccessible with conventional devicesthroughout the alimentary tract. Its computer-basedstapling mechanism has the potential to provideconsistent stapling, which can be difficult to obtainusing other devices. Minor improvements in thedesign will optimize its use for laparoscopic sur-gery, but further randomized clinical trails are need-ed to compare it with existing devices. Future prod-ucts, such as right-angled linear cutters and straightlinear cutter attached to the Flex shaft, can widenthe application of its use throughout the alimentarytract.

This work was supported by Mount Sinai Minimal InvasiveSurgical Center.

References

1. Wittgrove AC, Clark GW. Laparoscopic gastricbypass Roux-en-Y in 500 patients: technique andresults with 3-60 months follow up. Obes Surg 2000;10: 233-9.

2. Higa KD, Boone KB, Tienchin H. Complications oflaparoscopic Roux-en-Y gastric bypass: 1040 patients– what have we learned? Obes Surg 2000; 10: 509-13.

3. Schauer PR, Ikramuddin S: Laparoscopic surgery formorbid obesity. Probl Gen Surg 2000; 17: 39-54.

4. Nguyen NT, Goldman C, Rosenquist JC et al.Laparoscopic versus open gastric bypass: A random-ized study of outcomes, quality of life and costs. AnnSurg 2001; 234: 279-91.

5. Schauer PR, Ikramuddin S, Gourash W et al.

Outcomes after laparoscopic Roux-en-Y bypass formorbid obesity. Ann Surg 2000: 232; 515-29.

6. Ren C, Patterson E, Gagner M. Early results oflaparoscopic biliopancreatic diversion with duodenalswitch: A case series of 40 consecutive patients. ObesSurg 2000; 10: 514-23.

7. Balsiger BM, Murr MM, Poggio JL et al. Bariatricsurgery: surgery for weight control in patients withmorbid obesity. Med Clin North Am 2000; 84: 477-89.

8. Scopinaro N, Adami FG, Marinari GM et al.Biliopancreatic diversion: Two decades of experience.In: Deitel M, ed. Update: Surgery for the MorbidlyObese Patient. Toronto: FD-Communications 2000:227-58.

9. Doldi SB, Micheletto G, Lattuada E et al. Adjustablegastric banding: 5-year experience. Obes Surg 2000;10: 171-3.

10.Brolin RE, Kenler HA, Gorman JH et al. Long limbgastric bypass in the superobese: a prospective ran-domized study. Ann Surg 1992; 215: 387-95.

11.Di Matteo G, Cancrini A Jr, Palazzini G et al. Stapledsuture in digestive tract surgery. Int Surg 1988; 73:23-8.

12.Latimer RG, Doane WA, Mckittrick JE et al.Automatic staple suturing for gastrointestinal surgery.Am J Surg 1975; 130: 766-71.

13.Lowdon IMR, Gear MWL, Kilby JO. Stapling instru-ments in upper gastrointestinal surgery: A retrospec-tive study of 362 cases. Br J Surg 1982; 69: 333-5.

14.Fisher MG. Bleeding from stapled anastomosis. Am JSurg 1976; 131: 745-7.

15.Penninckx FM, Kerremans RP, Geboes KJ. The heal-ing of single and double-row stapled circular anasto-mosis; Dis Colon Rectum 1984; 27: 714-19.

(Received June 18, 2002; accepted November 2, 2002)