Original article
Outcomes after laparoscopic conversion of failed adjustable gastric banding to sleeve gastrectomy or Roux-en-Y gastric bypass P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal The Bariatric and Metabolic Institute, Section of Minimally Invasive Surgery, Department of General Surgery, Cleveland Clinic Florida, 2950 Cleveland Clinic Boulevard, Weston, Florida 33331, USA Correspondence to: Professor R. J. Rosenthal (e-mail: [email protected])
Background: Laparoscopic adjustable gastric banding (LAGB) has a high incidence of long-term
complications and failures. The best procedure to handle these failures and the optimal number of stages in such cases is still controversial. The aim of this retrospective study was to compare the results of conversions of LAGB to either laparoscopic sleeve gastrectomy (LSG) or laparoscopic Roux-en-Y gastric bypass (LRYGB) in failed LAGB using a single-stage approach. Methods: All patients who underwent conversion from LAGB to either LRYGB or LSG between January 2005 and March 2012 were included in the study. Early and late complications were reviewed. The percentage excess weight loss (%EWL) between the two groups was compared at 3, 6, 12 and 24 months of follow-up. Results: Fifty-nine patients, 11 men and 48 women, were included in the study. The most frequent indication was insufficient weight loss or weight regain (non-responders group), in 44 patients (75 per cent); 15 patients had a revision for complicated LAGB. The early complication rate in the nonresponders group was 7 per cent (3 of 44 patients), compared with 13 per cent (2 of 15) in the complicated LAGB group. Mean(s.d.) %EWL in the non-responders group was 55(22) per cent in patients converted to LRYGB and 28(25) in those converted to LSG (P = 0·001). Conclusion: LRYGB and LSG are both safe and feasible options for failed or complicated LAGB. In the non-responders group, %EWL was superior for conversion to LRYGB. The surgical morbidity rate was highest in patients having revision for band complications. Paper accepted 18 September 2013 Published online in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.9344
Introduction
Bariatric surgery is considered the only effective longterm treatment for severe obesity1,2 . In the late 1990s, laparoscopic adjustable gastric banding (LAGB) was one of the most popular bariatric procedures. This popularity was due to the perception of its reversibility and to initial good results in terms of weight loss (percentage excess weight loss (%EWL) 54–58 per cent at 5 years), resolution of co-morbidities and low operative morbidity rates3 – 5 . However, extended follow-up showed high failure (20–56 per cent) and removal (10–50 per cent) rates as a result of failure of weight loss or complications6,7 . The most common failures are related to pouch dilatation or slippage. There are several options after band failure, such as redo banding, conversion to laparoscopic sleeve gastrectomy (LSG), laparoscopic Roux-en-Y gastric bypass 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
(LRYGB), duodenal switch (DS) or biliopancreatic diversion (BPD)8 – 13 . Bariatric conversion surgery is associated with a higher complication rate than primary procedures, independent of the technique used14 . It is not clear which is the best surgical solution for a failed band, and a comprehensive approach is needed. Recently, LSG has become a popular option as both a primary and a revisional procedure14 – 16 . There is still controversy regarding whether conversion of a failed or complicated LAGB should be performed in a singleor two-stage approach. Those who advocate a two-stage approach state that the capsule (reactive scar tissue that normally builds up around an implanted adjustable gastric band) will become thinner, allowing better staple-line formation and decreasing the incidence of staple line disruption15 . The aim of this study was to report the present authors’ experience, and to compare LRYGB and BJS 2014; 101: 254–260
Laparoscopic conversion of failed adjustable gastric banding
LSG performed in a single stage as a conversion procedure for failed LAGB – patients with insufficient weight loss or weight regain (non-responders) or patients with band complications. Methods
After institutional review board approval, data from a prospectively maintained database at the Bariatric and Metabolic Institute of the Cleveland Clinic Florida were reviewed for all patients undergoing conversion from LAGB to LSG or LRYGB between January 2005 and March 2012. Patients underwent conversion to LSG or LRYGB if they were non-responders or for complications after LAGB. Slippage, erosion, gastro-oesophageal reflux, dysphagia and food intolerance were considered LAGBrelated complications. Insufficient weight loss (nonresponders), defined as %EWL of less than 25 per cent at 2 years, was also an indication for revisional surgery. Singlestage laparoscopic conversion was the preferred approach. However, in patients who presented after the band had been removed by another surgeon, the two-stage approach of LAGB conversion to LRYGB or LSG was employed. The conversion procedure of choice was LRYGB. LSG was performed mainly as a result of patient choice. All patients had a preoperative assessment with upper gastrointestinal contrast X-ray series and upper endoscopy. A tailored multidisciplinary approach (psychologist, nutritionist) within the framework of the local bariatric unit was used. Clinical data collected included: age, sex, body mass index (BMI) at primary and revisional operations, indications for revision, duration of surgery (measured from the time when the surgeon cut the skin to closure of the last trocar site), blood loss, hospital stay, early morbidity and mortality, late complications and degree of weight reduction. Complications were defined as expected or unexpected adverse events that required medical or surgical intervention and resulted in a prolonged hospital stay. %EWL was calculated from the time of reoperation by the formula: %EWL = (initial weight − final weight)/(initial weight − ideal weight) × 100. A bariatric operation was defined as successful when %EWL was at least 50 per cent.
Surgical technique Patients were placed in the supine position. The abdominal cavity was accessed through a supraumbilical incision with the OPTIVIEW trocar (Ethicon Endo-Surgery, Cincinnati, Ohio, USA). Seven trocars were placed in the subxiphoid area, right and left upper quadrants. The access port of the band was removed during placement of the 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
255
left upper quadrant trocar. Adhesiolysis was performed, particularly between the left lateral liver segment and the band, with a Harmonic Scalpel (Ethicon Endo-Surgery) or by blunt dissection, and the band was easily identified by the tubing system. The band was kept in place until: the fundic wrap had been taken down and freed from the upper and posterior adhesions; the buckle of the band had been carefully exposed; and the right and left crus of the diaphragm had been clearly dissected to rule out or repair a hiatal hernia and better identify the anatomy. To avoid accidental injury to the gastric wall, the band was cut with scissors and removed, together with the port. In all patients, the capsule was then dissected and excised on the greater curvature side of the gastric wall, to allow better staple-line formation and to reduce staple-line disruptions.
Sleeve gastrectomy The vascular supply of the greater curvature of the stomach was divided with the Harmonic Scalpel from the antrum 5 cm proximal to the pylorus to the angle of His. The phreno-oesophageal membrane was dissected in an attempt to mobilize the fundus and identify any hiatal hernia, which was then repaired posteriorly. The LSG was calibrated with a 38-Fr gastric bougie and created using a linear cutting stapler with 4·1-mm (green) and 3·5-mm (blue) staples. Special attention was paid to the proximal third of the sleeve. For all LSG procedures, the transection was performed using green cartridges because of the increased thickness of the gastric wall at the level of the previous band, although the sleeve channel was not made wider at the band removal site. When resecting the stomach, the authors made sure to staple laterally from the fat pad that covers the gastro-oesophageal junction. The staple line was buttressed with a seroserosal 2/0 polyglactin 910 running suture in all cases. A drain was left in place at the end of the procedure in all patients. It is the authors’ routine to drain primary and reoperative bariatric procedures. The stomach specimen and band were removed through the supraumbilical trocar site. Roux en-Y gastric bypass LRYGB was performed using a seven-port approach according to the authors’ standard technique. All procedures had a 100-cm alimentary limb. Proximal gastric pouches had a capacity of approximately 15–30 ml. The stomach was transected below the left gastric artery to maintain an optimal blood supply to the pouch and anastomosis, as well as medially and proximally to the scar tissue of the band to avoid staple-line disruption. The gastrojejunostomy was performed with a linear stapler with 3·5-mm staples (blue cartridge) and handsewn closed www.bjs.co.uk
BJS 2014; 101: 254–260
256
P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal
with 2/0 polyglactin 910. The jejunojejunostomy was performed with a completely stapled technique. The enteric limb was positioned in an antecolic and antegastric fashion. An intraoperative leak test with a combination of pneumatic, methylene blue and intraoperative endoscopy was performed in all patients to rule out leakage and bleeding. A drain was placed near the gastrojejunostomy in all patients.
Perioperative and postoperative care All patients had an upper gastrointestinal X-ray series with water-soluble contrast within 2 days. If no leak or stenosis was found and good transit was shown, the drain was removed, a soft diet was started and the patient was discharged the following day.
Indications for conversion following laparoscopic adjustable gastric banding
Table 1
Sleeve Roux-en-Y gastrectomy gastric bypass (n = 39) (n = 20) Non-responders* Band complications Slippage/prolapse Erosion Gastro-oesophageal reflux disease Intolerance
31 (1) 8 2 (1) 3 (2) 2 (1) 1 (1)
Results
During the study period, 59 patients, 11 men and 48 women, underwent laparoscopic conversion of LAGB: 39 patients (66 per cent; 6 men) were converted to LRYGB and 20 (34 per cent; 5 men) to LSG. Initial band placement was performed at the authors’ centre in 28 patients (47 per cent) and at another institution in 31 (53 per cent). The indications for revisional surgery are detailed in Table 1. Of the 59 patients, 44 (75 per cent) were non-responders to LAGB (31 were converted to LRYGB and 13 to LSG) and 15 (25 per cent) had band complications (8 patients converted to LRYGB and 7 to LSG). The mean interval between the primary operation and conversion was 50(35) months in the LRYGB group and 31(23) months in the LSG group. At the time of revision, the mean age of patients in the LRYGB group was 49(14) (range 25–80) years and their mean BMI was 42(6) (range 31–57) kg/m2 ; in the LSG group the mean age was 44(17) (19–71) years and mean BMI 39(6) (30–51) kg/m2 . Mean BMI before LAGB was 44(6) (35–58) kg/m2 in the LRYGB group and 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
0·226‡ 0·131 0·569 0·466 1·000
Values in parentheses are number of stages for the surgical approach. *Insufficient weight loss or weight regain. †Fisher’s exact test, except ‡χ2 test.
Characteristics of patients in the two conversion groups for both non-responders and those with band complications after laparoscopic adjustable gastric banding
Table 2
Statistical analysis All data for baseline characteristics and %EWL are reported as mean(s.d.) values unless otherwise specified. A paired t test was performed to compare %EWL before conversion and at follow-up in patients undergoing LRYGB and LSG procedures. Comparison of categorical variables between groups was by means of the χ2 test or Fisher’s exact test. Statistical analysis of data was performed using SPSS version 12·0 for Windows (IBM, Armonk, New York, USA). P < 0·050 was considered statistically significant.
13 (1) 7 5 (2) 1 (2) 0 (1) 1 (1)
P†
Age (years) Sex ratio (F : M) Time between LAGB and revision (months) Length of follow-up after revision (months) BMI (kg/m2 ) Before LAGB Before conversion surgery At last follow-up %EWL after revisional surgery At 12 months At 24 months
LRYGB (n = 39)
LSG (n = 20)
P*
49(14) 33 : 6 50(35) 22(18)
44(17) 15 : 5 31(23) 33(29)
0·217 0·369† 0·028 0·073
44(6) 42(6) 31(7)
45(6) 39(6) 33(6)
0·719 0·101 0·164
59(20) 55(22)
35(20) 28(25)
0·011 0·001
Values are mean(s.d.). LRYGB, laparoscopic Roux-en-Y gastric bypass; LSG, laparoscopic sleeve gastrectomy; LAGB, laparoscopic adjustable gastric banding; BMI, body mass index; %EWL, percentage excess weight loss. *t test, except †χ2 test.
45(6) (37–60) kg/m2 in the LSG group. At the time of conversion, four of the 39 patients having LRYGB and one of the 20 patients having LSG were superobese (BMI above 50 kg/m2 ). Patient characteristics are shown in Table 2. The mean duration of surgery was 142(50) (range 60–300) min in the LRYGB group and 121(23) (85–180) min in the LSG group (Table 3). All procedures were performed laparoscopically. There were no conversions to an open procedure. Median hospital stay was 3 (range 3–47) days for LRYGB and 3 (3–20) days for LSG. Five patients, three in the LRYGB group and two in the LSG group, remained in hospital for more than 7 days. Thirty-six (92 per cent) of the 39 patients in the LRYGB group had a one-stage approach, compared with 14 (70 per cent) of the 20 patients in the LSG group (Table 1). Mean follow-up was 22(18) and 33(29) months in the LRYGB and LSG group respectively. The proportion of patients attending for follow-up at 3, 6, 12 and 24 months www.bjs.co.uk
BJS 2014; 101: 254–260
Laparoscopic conversion of failed adjustable gastric banding
Outcome following conversion surgery
Operating time (min) Blood loss (ml) Hospital stay (days) Total no. of complications Early complications Leakage Stenosis Small bowel obstruction Intussusception Late complications Small bowel obstruction Marginal ulcer Stricture Chronic abdominal pain Dilatation Deep vein thrombosis Conversion to open procedure Death
Non-responders to LRYGB Non-responders to LSG
LRYGB (n = 39)
LSG (n = 20)
P*
142(50) 76(56) 5(7) 10
121(23) 74(52) 4(4) 4
< 0·132† 0·852† 0·644† 0·629
0 0 1 1
1 2 0 0
0·339 0·111 0·661 0·661
4 1 1 1 0 1 0 0
0 0 0 0 1 0 0 0
0·180 0·661 0·661 0·661 0·661 0·661
Complicated LAGB to LRYGB 70
50
40
30
20
10
Values are mean(s.d.). LRYGB, laparoscopic Roux-en-Y gastric bypass; LSG, laparoscopic sleeve gastrectomy. *Fisher’s exact test, except †t test.
was 16 of 39, 21 of 39, 23 of 34 and 13 of 27 respectively in the LRYGB group, and six of 19, ten of 19, seven of 18 and six of 15 in the LSG group. No statistically significant difference in %EWL was seen between the two conversion groups at less than 12 months’ follow-up, but at 12 and 24 months there was a statistically significant difference between the groups (P = 0·011 and P = 0·001 respectively) (Table 2). When separate analyses were made of non-responders and patients with band complications, mean %EWL was 55(22) per cent at 24 months in non-responders converted to LRYGB and 26(31) per cent in non-responders converted to LSG (P = 0·004). In contrast, mean %EWL was 42(15) per cent at 24 months in patients with band complications who were converted to LRYGB and 42(13) per cent in those converted to LSG (P = 0·629) (Fig. 1). There were no perioperative (30-day) deaths in either group. Five postoperative complications were recorded, two (13 per cent) in the 15 patients who had band complications. One patient, following removal of an eroded band 3 years before the conversion, developed acute complete obstruction of the proximal gastric sleeve that required conversion to LRYGB after 2 days. Another patient with conversion of a slipped gastric band developed early short stenosis at the sleeve and mega-oesophagus that required reintervention 4 months after conversion. This patient had repair of a large para-oesophageal hernia with non-absorbable suture. Three (7 per cent) of the 44 patients in the non-responders group had complications. One patient had peritonitis secondary to 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
Complicated LAGB to LSG
60
% EWL
Table 3
257
0
3
6
12
24
Follow-up (months)
Mean percentage excess weight loss (%EWL) at 3, 6, 12 and 24 months in non-responders to laparoscopic adjustable gastric banding (LAGB) and patients with complicated LAGB who were converted to laparoscopic Roux-en-Y gastric bypass (LRYGB) or laparoscopic gastric bypass (LSG)
Fig. 1
necrosis of the gastric fundus and was treated with laparoscopic exploration, suture and omentum patch repair on postoperative day 3. After lavage and repair, this patient was subsequently managed with drainage, bowel rest and total parenteral nutrition. One patient with alimentary limb intussusception required resection and redo jejunojejunostomy. In the third patient a small bowel obstruction that required lysis of adhesions occurred 9 days after conversion. All early complications were in patients who had a one-stage procedure. There were nine late complications (15 per cent), of which eight were in the LRYGB group. Four of these patients had small bowel obstruction, two of whom required small bowel resection (no internal hernias were identified), one patient had a marginal ulcer with bleeding, one required dilatation of a stricture at the gastrojejunostomy, and another patient required diagnostic laparoscopy for chronic abdominal pain. One patient who had conversion to LRYGB had deep vein thrombosis. In the LSG group, one patient developed dilatation of the sleeve proximal to distal compression, which resolved with operative adhesiolysis. Another patient converted to www.bjs.co.uk
BJS 2014; 101: 254–260
258
P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal
LSG subsequently required a LRYGB revision because of insufficient weight loss after 3 years. Discussion
Despite the popularity of gastric banding, many patients require reoperative surgery for band removal or conversion to another procedure because of insufficient weight loss and/or complications. One of the main controversies surrounding this clinical conundrum of need for conversion is selection of the procedure. Specific decision points include which procedure will yield the best outcomes, and which approach, single- or two-stage, best avoids perioperative complications. The literature lacks reports to help elucidate both of these questions. Highlighting the safety of laparoscopic single-stage conversion of a failed or complicated LAGB to LRYGB or LSG, this series showed no mortality and acceptable morbidity rates for both procedures, all of which were completed laparoscopically. LSG as a primary procedure has rapidly gained acceptance among surgeons owing to several advantages that it offers compared with other, more complex, procedures17 – 20 . In fact, LSG does not alter bowel continuity and is associated with a relatively low risk of mineral and vitamin deficiencies, with the exception of a still significant risk of vitamin B12 deficiency13,14 . Other advantages include the avoidance of dumping syndrome, marginal ulceration, malabsorption, internal hernia and foreign body material. Another advantage is that it can be used as the first step of a staged procedure in patients who fail to lose weight after an initial bariatric procedure. One of the major disadvantages of LSG is the risk of leakage at the oesophagogastric junction, which has been reported to be increased following revision of LAGB procedures to LSG20 – 22 . This is probably because stapling of the stomach is performed in scarred tissue and dissection of the left crus might jeopardize the vascular supply of the oesophagogastric junction20 – 22 . The overall complication rate for primary LSG is acceptable (5–10 per cent), and the mortality rate associated with this procedure is low17,21 . LSG is also a viable option for staged surgery in high-risk, superobese patients6,11 . In the present series, one of 20 patients undergoing LSG after failed LAGB developed a leak from the staple line at the level of the oesophagogastric junction. At 24 months, the overall (combining nonresponders and band complication groups) mean %EWL was 28(25) per cent, which is comparable to earlier published results for LSG as a revisional option7,20,22 . The weight loss data are at 24-month follow-up, so the results should be interpreted with caution, as longer follow-up 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
might show some weight regain. These results are worse than those for LRYGB, where %EWL of up to 60 per cent is achievable22 . In the present series there was a statistically significant difference in %EWL at 24 months for nonresponders converted to LRYGB and those converted to LSG (55(22) versus 26(31) per cent respectively). Based on the results presented, LSG does not seem to be the ideal operation when converting a failed LAGB. In the present series, 39 (66 per cent) of the 59 patients with a failed LAGB were converted to LRYGB, which is a technically more challenging procedure than LSG owing to the need to dissect both the angle of His and the lesser curve – the area of the greatest scarring and adhesions. Conversion to LRYGB is associated with a complication rate of less than 19 per cent22,23 . In the present study, only two (5 per cent) major postoperative complications occurred. The mean operating time was longer than for the average primary LRYGB, which is usually 110 min22 . This observation was noted in two other studies22,23 . With a %EWL of 55 per cent at 24 months, the present results are comparable to those already published23 . The overall complication rate in the present series was 22 per cent (13 of 59) with no deaths, which is in keeping with the literature (13–34 per cent)22,23 . The leak rate was 2 per cent (1 of 59), in a patient in whom the capsule was not taken down and 3·5-mm staples (blue cartridge linear staplers) were used. The necrosis of the gastric fundus was, in the authors’ opinion, most likely due to lack of blood supply and/or trauma. Patients who fail to lose weight or who regain weight after gastric banding (non-responders) with no apparent band complications are considered to have failure of restrictive procedures and the rescue procedure should include a malabsorptive component. LSG is probably a less desirable option in this scenario. Most bariatric surgeons do not consider converting a failed restrictive procedure such as LAGB to another restrictive procedure such as LSG to be a good strategy. Patients with bandassociated complications, such as slippage, pouch dilatation or band erosion, that occurred after successful weight loss can undergo conversion to any of the rescue procedures mentioned. In order to choose the most successful strategy, it is important to review the overall condition and comorbidities of the patient. Symptomatic reflux and diabetes are considered an indication for LRYGB, unless the patient refuses the procedure. Another alternative for revision of failed LAGB is BPD–DS, which is a more complex conversion with higher morbidity and mortality rates. Some authors have reported the safety of BPDDS as a reoperative intervention10 , but others11 have found higher morbidity and mortality rates (62 and 8 per cent respectively). However, a small study13 directly www.bjs.co.uk
BJS 2014; 101: 254–260
Laparoscopic conversion of failed adjustable gastric banding
comparing the standard revisional procedures of LRYGB and BPD–DS failed to demonstrate any superiority for one procedure over the other. BPD–DS, as either a primary or revisional procedure, is not routinely offered in the present authors’ practice. In their centre, the present authors perform both LRYGB and LSG as primary and revisional bariatric procedures. They routinely recommend a single-stage approach, and also recommend LRYGB. Patients have a choice of procedure provided they do not have severe gastro-oesophageal reflux disease, when the authors will perform only LRYGB, not LSG. In patients with eroded bands, the authors always undertake a two-stage approach. The overall complication rate for LSG as a revision of LAGB is higher than that of LSG as a primary procedure (13–34 per cent)18 – 20 . The different complication rates in primary versus revisional procedures are probably explained by tissue trauma and ischaemia. In addition, a compromised vascular supply of the superior aspect of the staple line, due to dissection of the left crus, may be a factor. Surgeons should balance the possibility of increased cost when managing a complicated conversion of LAGB to LSG performed in a single stage in comparison with an uncomplicated two-stage approach. Several weaknesses and strengths of this study need to be addressed. This series is relatively small and retrospective, with an intermediate follow-up of 24 months. Poor followup in this institution is related to many factors, including the fact that patients come from other countries or states to undergo reoperation. Additionally, in the USA, owing to the current economic situation, patients tend to move out of state, which can make it difficult to attain meaningful data on long-term success. Disclosure
The authors declare no conflict of interest. References 1 Suter M, Calmes JM, Paroz A, Giusti V. A 10-year experience with laparoscopic gastric banding for morbid obesity: high long-term complication and failure rates. Obes Surg 2006; 16: 829–835. 2 van Gemert WG, van Wersch MM, Greve JW, Soeters PB. Revisional surgery after vertical banded gastroplasty: restoration of vertical banded gastroplasty or conversion to gastric bypass. Obes Surg 1998; 8: 21–28. 3 Favretti F, Segato G, Ashton D, Busetto L, De Luca M, Mazza M et al. Laparoscopic adjustable gastric banding in 1791 consecutive obese patients: 12-year results. Obes Surg 2007; 17: 168–175. 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
259
4 Gagner M, Gumbs AA. Gastric banding: conversion to sleeve, bypass, or DS. Surg Endosc 2007; 21: 1931–1935. 5 Cunneen SA. Review of meta-analytic comparisons of bariatric surgery with a focus on laparoscopic adjustable gastric banding. Surg Obes Relat Dis 2008; 4(Suppl): S47–S55. 6 Foletto M, Bernante P, Busetto L, Pomerri F, Vecchiato G, Prevedello L et al. Laparoscopic gastric rebanding for slippage with pouch dilation: results on 29 consecutive patients. Obes Surg 2008; 18: 1099 –1103. 7 Biertho L, Steffen R, Branson R, Potoczna N, Ricklin T, Piec G et al. Management of failed adjustable gastric banding. Surgery 2005; 137: 33–41. 8 Dapri G, Cadi`ere GB, Himpens J. Feasibility and technique of laparoscopic conversion of adjustable gastric banding to sleeve gastrectomy. Surg Obes Relat Dis 2009; 5: 72– 76. 9 Dargent J. Surgical treatment of morbid obesity by adjustable gastric band: the case for a conservative strategy in the case of failure – a 9-year series. Obes Surg 2004; 14: 986–990. 10 Topart P, Becouarn G, Ritz P. Biliopancreatic diversion with duodenal switch or gastric bypass for failed gastric banding: retrospective study from two institutions with preliminary results. Surg Obes Relat Dis 2007; 3: 521–525. 11 Spivak H, Beltran OR, Slavchev P, Wilson EB. Laparoscopic revision from LAP-BAND to gastric bypass. Surg Endosc 2007; 21: 1388–1392. 12 van Wageningen B, Berends FJ, Van Ramshorst B, Janssen IF. Revision of failed laparoscopic adjustable gastric banding to Roux-en-Y gastric bypass. Obes Surg 2006; 16: 137–141. 13 Di Betta E, Mittempergher F, Di Fabio F, Casella C, Terraroli C, Salerni B. Duodenal switch without gastric resection after failed gastric restrictive surgery for morbid obesity. Obes Surg 2006; 16: 258–261. 14 Patel S, Szomstein S, Rosenthal RJ. Reasons and outcomes of reoperative bariatric surgery for failed and complicated procedures (excluding adjustable gastric banding). Obes Surg 2011; 21: 1209–1219. 15 Patel S, Eckstein J, Acholonu E, Abu-Jaish W, Szomstein S, Rosenthal RJ. Reasons and outcomes of laparoscopic revisional surgery after laparoscopic adjustable gastric banding for morbid obesity. Surg Obes Relat Dis 2010; 6: 391–398. 16 Abu-Jaish W, Rosenthal RJ. Sleeve gastrectomy: a new surgical approach for morbid obesity. Expert Rev Gastroenterol Hepatol 2010; 4: 101–119. 17 Acholonu E, McBean E, Court I, Bellorin O, Szomstein S, Rosenthal RJ. Safety and short-term outcomes of laparoscopic sleeve gastrectomy as a revisional approach for failed laparoscopic adjustable gastric banding in the treatment of morbid obesity. Obes Surg 2009; 19: 1612–1616. 18 Buchwald H, Oien DM. Metabolic/bariatric surgery worldwide 2008. Obes Surg 2009; 19: 1605–1611.
www.bjs.co.uk
BJS 2014; 101: 254–260
260
P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal
19 Deitel M, Crosby RD, Gagner M. The First International Consensus Summit for Sleeve Gastrectomy (SG), New York City, October 25–27, 2007. Obes Surg 2008; 18: 487–496. 20 Rubin M, Yehoshua RT, Stein M, Lederfein D, Fichman S, Bernstine H et al. Laparoscopic sleeve gastrectomy with minimal morbidity. Early results in 120 morbidly obese patients. Obes Surg 2008; 18: 1567–1570. 21 Frezza EE, Jaramillo-de la Torre EJ, Enriquez CC, Gee L, Wachtel MS, Lopez Corvala JA. Laparoscopic sleeve
gastrectomy after gastric banding removal: a feasibility study. Surg Innov 2009; 16: 68–72. 22 Tucker O, Sucandy I, Szomstein S, Rosenthal RJ. Revisional surgery after failed laparoscopic adjustable gastric banding. Surg Obes Relat Dis 2008; 4: 740–747. 23 Mognol P, Chosidow D, Marmuse JP. Laparoscopic conversion of laparoscopic gastric banding to Roux-en-Y gastric bypass: a review of 70 patients. Obes Surg 2004; 14: 1349–1353.
Snapshot quiz
Snapshot quiz 14/2 Answer: Anastomotic leak at the enteroenterostomy. During the initial procedure, the integrity of the gastrojejunostomy was assessed using methylene blue dye. At reoperation, an anastomotic leak secondary to staple-line failure at the enteroenterostomy was identified. The blue discoloration of the skin resulted from leakage of the dye at this site and subsequent extravasation through the port site. In addition to the physiological deterioration on day 1, the bright blue ‘bruising’ was a further cutaneous indication of an underlying anastomotic leak.
2014 BJS Society Ltd Published by John Wiley & Sons Ltd
www.bjs.co.uk
BJS 2014; 101: 254–260
Outcomes after laparoscopic conversion of failed adjustable gastric banding to sleeve gastrectomy or Roux-en-Y gastric bypass P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal The Bariatric and Metabolic Institute, Section of Minimally Invasive Surgery, Department of General Surgery, Cleveland Clinic Florida, 2950 Cleveland Clinic Boulevard, Weston, Florida 33331, USA Correspondence to: Professor R. J. Rosenthal (e-mail: [email protected])
Background: Laparoscopic adjustable gastric banding (LAGB) has a high incidence of long-term
complications and failures. The best procedure to handle these failures and the optimal number of stages in such cases is still controversial. The aim of this retrospective study was to compare the results of conversions of LAGB to either laparoscopic sleeve gastrectomy (LSG) or laparoscopic Roux-en-Y gastric bypass (LRYGB) in failed LAGB using a single-stage approach. Methods: All patients who underwent conversion from LAGB to either LRYGB or LSG between January 2005 and March 2012 were included in the study. Early and late complications were reviewed. The percentage excess weight loss (%EWL) between the two groups was compared at 3, 6, 12 and 24 months of follow-up. Results: Fifty-nine patients, 11 men and 48 women, were included in the study. The most frequent indication was insufficient weight loss or weight regain (non-responders group), in 44 patients (75 per cent); 15 patients had a revision for complicated LAGB. The early complication rate in the nonresponders group was 7 per cent (3 of 44 patients), compared with 13 per cent (2 of 15) in the complicated LAGB group. Mean(s.d.) %EWL in the non-responders group was 55(22) per cent in patients converted to LRYGB and 28(25) in those converted to LSG (P = 0·001). Conclusion: LRYGB and LSG are both safe and feasible options for failed or complicated LAGB. In the non-responders group, %EWL was superior for conversion to LRYGB. The surgical morbidity rate was highest in patients having revision for band complications. Paper accepted 18 September 2013 Published online in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.9344
Introduction
Bariatric surgery is considered the only effective longterm treatment for severe obesity1,2 . In the late 1990s, laparoscopic adjustable gastric banding (LAGB) was one of the most popular bariatric procedures. This popularity was due to the perception of its reversibility and to initial good results in terms of weight loss (percentage excess weight loss (%EWL) 54–58 per cent at 5 years), resolution of co-morbidities and low operative morbidity rates3 – 5 . However, extended follow-up showed high failure (20–56 per cent) and removal (10–50 per cent) rates as a result of failure of weight loss or complications6,7 . The most common failures are related to pouch dilatation or slippage. There are several options after band failure, such as redo banding, conversion to laparoscopic sleeve gastrectomy (LSG), laparoscopic Roux-en-Y gastric bypass 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
(LRYGB), duodenal switch (DS) or biliopancreatic diversion (BPD)8 – 13 . Bariatric conversion surgery is associated with a higher complication rate than primary procedures, independent of the technique used14 . It is not clear which is the best surgical solution for a failed band, and a comprehensive approach is needed. Recently, LSG has become a popular option as both a primary and a revisional procedure14 – 16 . There is still controversy regarding whether conversion of a failed or complicated LAGB should be performed in a singleor two-stage approach. Those who advocate a two-stage approach state that the capsule (reactive scar tissue that normally builds up around an implanted adjustable gastric band) will become thinner, allowing better staple-line formation and decreasing the incidence of staple line disruption15 . The aim of this study was to report the present authors’ experience, and to compare LRYGB and BJS 2014; 101: 254–260
Laparoscopic conversion of failed adjustable gastric banding
LSG performed in a single stage as a conversion procedure for failed LAGB – patients with insufficient weight loss or weight regain (non-responders) or patients with band complications. Methods
After institutional review board approval, data from a prospectively maintained database at the Bariatric and Metabolic Institute of the Cleveland Clinic Florida were reviewed for all patients undergoing conversion from LAGB to LSG or LRYGB between January 2005 and March 2012. Patients underwent conversion to LSG or LRYGB if they were non-responders or for complications after LAGB. Slippage, erosion, gastro-oesophageal reflux, dysphagia and food intolerance were considered LAGBrelated complications. Insufficient weight loss (nonresponders), defined as %EWL of less than 25 per cent at 2 years, was also an indication for revisional surgery. Singlestage laparoscopic conversion was the preferred approach. However, in patients who presented after the band had been removed by another surgeon, the two-stage approach of LAGB conversion to LRYGB or LSG was employed. The conversion procedure of choice was LRYGB. LSG was performed mainly as a result of patient choice. All patients had a preoperative assessment with upper gastrointestinal contrast X-ray series and upper endoscopy. A tailored multidisciplinary approach (psychologist, nutritionist) within the framework of the local bariatric unit was used. Clinical data collected included: age, sex, body mass index (BMI) at primary and revisional operations, indications for revision, duration of surgery (measured from the time when the surgeon cut the skin to closure of the last trocar site), blood loss, hospital stay, early morbidity and mortality, late complications and degree of weight reduction. Complications were defined as expected or unexpected adverse events that required medical or surgical intervention and resulted in a prolonged hospital stay. %EWL was calculated from the time of reoperation by the formula: %EWL = (initial weight − final weight)/(initial weight − ideal weight) × 100. A bariatric operation was defined as successful when %EWL was at least 50 per cent.
Surgical technique Patients were placed in the supine position. The abdominal cavity was accessed through a supraumbilical incision with the OPTIVIEW trocar (Ethicon Endo-Surgery, Cincinnati, Ohio, USA). Seven trocars were placed in the subxiphoid area, right and left upper quadrants. The access port of the band was removed during placement of the 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
255
left upper quadrant trocar. Adhesiolysis was performed, particularly between the left lateral liver segment and the band, with a Harmonic Scalpel (Ethicon Endo-Surgery) or by blunt dissection, and the band was easily identified by the tubing system. The band was kept in place until: the fundic wrap had been taken down and freed from the upper and posterior adhesions; the buckle of the band had been carefully exposed; and the right and left crus of the diaphragm had been clearly dissected to rule out or repair a hiatal hernia and better identify the anatomy. To avoid accidental injury to the gastric wall, the band was cut with scissors and removed, together with the port. In all patients, the capsule was then dissected and excised on the greater curvature side of the gastric wall, to allow better staple-line formation and to reduce staple-line disruptions.
Sleeve gastrectomy The vascular supply of the greater curvature of the stomach was divided with the Harmonic Scalpel from the antrum 5 cm proximal to the pylorus to the angle of His. The phreno-oesophageal membrane was dissected in an attempt to mobilize the fundus and identify any hiatal hernia, which was then repaired posteriorly. The LSG was calibrated with a 38-Fr gastric bougie and created using a linear cutting stapler with 4·1-mm (green) and 3·5-mm (blue) staples. Special attention was paid to the proximal third of the sleeve. For all LSG procedures, the transection was performed using green cartridges because of the increased thickness of the gastric wall at the level of the previous band, although the sleeve channel was not made wider at the band removal site. When resecting the stomach, the authors made sure to staple laterally from the fat pad that covers the gastro-oesophageal junction. The staple line was buttressed with a seroserosal 2/0 polyglactin 910 running suture in all cases. A drain was left in place at the end of the procedure in all patients. It is the authors’ routine to drain primary and reoperative bariatric procedures. The stomach specimen and band were removed through the supraumbilical trocar site. Roux en-Y gastric bypass LRYGB was performed using a seven-port approach according to the authors’ standard technique. All procedures had a 100-cm alimentary limb. Proximal gastric pouches had a capacity of approximately 15–30 ml. The stomach was transected below the left gastric artery to maintain an optimal blood supply to the pouch and anastomosis, as well as medially and proximally to the scar tissue of the band to avoid staple-line disruption. The gastrojejunostomy was performed with a linear stapler with 3·5-mm staples (blue cartridge) and handsewn closed www.bjs.co.uk
BJS 2014; 101: 254–260
256
P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal
with 2/0 polyglactin 910. The jejunojejunostomy was performed with a completely stapled technique. The enteric limb was positioned in an antecolic and antegastric fashion. An intraoperative leak test with a combination of pneumatic, methylene blue and intraoperative endoscopy was performed in all patients to rule out leakage and bleeding. A drain was placed near the gastrojejunostomy in all patients.
Perioperative and postoperative care All patients had an upper gastrointestinal X-ray series with water-soluble contrast within 2 days. If no leak or stenosis was found and good transit was shown, the drain was removed, a soft diet was started and the patient was discharged the following day.
Indications for conversion following laparoscopic adjustable gastric banding
Table 1
Sleeve Roux-en-Y gastrectomy gastric bypass (n = 39) (n = 20) Non-responders* Band complications Slippage/prolapse Erosion Gastro-oesophageal reflux disease Intolerance
31 (1) 8 2 (1) 3 (2) 2 (1) 1 (1)
Results
During the study period, 59 patients, 11 men and 48 women, underwent laparoscopic conversion of LAGB: 39 patients (66 per cent; 6 men) were converted to LRYGB and 20 (34 per cent; 5 men) to LSG. Initial band placement was performed at the authors’ centre in 28 patients (47 per cent) and at another institution in 31 (53 per cent). The indications for revisional surgery are detailed in Table 1. Of the 59 patients, 44 (75 per cent) were non-responders to LAGB (31 were converted to LRYGB and 13 to LSG) and 15 (25 per cent) had band complications (8 patients converted to LRYGB and 7 to LSG). The mean interval between the primary operation and conversion was 50(35) months in the LRYGB group and 31(23) months in the LSG group. At the time of revision, the mean age of patients in the LRYGB group was 49(14) (range 25–80) years and their mean BMI was 42(6) (range 31–57) kg/m2 ; in the LSG group the mean age was 44(17) (19–71) years and mean BMI 39(6) (30–51) kg/m2 . Mean BMI before LAGB was 44(6) (35–58) kg/m2 in the LRYGB group and 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
0·226‡ 0·131 0·569 0·466 1·000
Values in parentheses are number of stages for the surgical approach. *Insufficient weight loss or weight regain. †Fisher’s exact test, except ‡χ2 test.
Characteristics of patients in the two conversion groups for both non-responders and those with band complications after laparoscopic adjustable gastric banding
Table 2
Statistical analysis All data for baseline characteristics and %EWL are reported as mean(s.d.) values unless otherwise specified. A paired t test was performed to compare %EWL before conversion and at follow-up in patients undergoing LRYGB and LSG procedures. Comparison of categorical variables between groups was by means of the χ2 test or Fisher’s exact test. Statistical analysis of data was performed using SPSS version 12·0 for Windows (IBM, Armonk, New York, USA). P < 0·050 was considered statistically significant.
13 (1) 7 5 (2) 1 (2) 0 (1) 1 (1)
P†
Age (years) Sex ratio (F : M) Time between LAGB and revision (months) Length of follow-up after revision (months) BMI (kg/m2 ) Before LAGB Before conversion surgery At last follow-up %EWL after revisional surgery At 12 months At 24 months
LRYGB (n = 39)
LSG (n = 20)
P*
49(14) 33 : 6 50(35) 22(18)
44(17) 15 : 5 31(23) 33(29)
0·217 0·369† 0·028 0·073
44(6) 42(6) 31(7)
45(6) 39(6) 33(6)
0·719 0·101 0·164
59(20) 55(22)
35(20) 28(25)
0·011 0·001
Values are mean(s.d.). LRYGB, laparoscopic Roux-en-Y gastric bypass; LSG, laparoscopic sleeve gastrectomy; LAGB, laparoscopic adjustable gastric banding; BMI, body mass index; %EWL, percentage excess weight loss. *t test, except †χ2 test.
45(6) (37–60) kg/m2 in the LSG group. At the time of conversion, four of the 39 patients having LRYGB and one of the 20 patients having LSG were superobese (BMI above 50 kg/m2 ). Patient characteristics are shown in Table 2. The mean duration of surgery was 142(50) (range 60–300) min in the LRYGB group and 121(23) (85–180) min in the LSG group (Table 3). All procedures were performed laparoscopically. There were no conversions to an open procedure. Median hospital stay was 3 (range 3–47) days for LRYGB and 3 (3–20) days for LSG. Five patients, three in the LRYGB group and two in the LSG group, remained in hospital for more than 7 days. Thirty-six (92 per cent) of the 39 patients in the LRYGB group had a one-stage approach, compared with 14 (70 per cent) of the 20 patients in the LSG group (Table 1). Mean follow-up was 22(18) and 33(29) months in the LRYGB and LSG group respectively. The proportion of patients attending for follow-up at 3, 6, 12 and 24 months www.bjs.co.uk
BJS 2014; 101: 254–260
Laparoscopic conversion of failed adjustable gastric banding
Outcome following conversion surgery
Operating time (min) Blood loss (ml) Hospital stay (days) Total no. of complications Early complications Leakage Stenosis Small bowel obstruction Intussusception Late complications Small bowel obstruction Marginal ulcer Stricture Chronic abdominal pain Dilatation Deep vein thrombosis Conversion to open procedure Death
Non-responders to LRYGB Non-responders to LSG
LRYGB (n = 39)
LSG (n = 20)
P*
142(50) 76(56) 5(7) 10
121(23) 74(52) 4(4) 4
< 0·132† 0·852† 0·644† 0·629
0 0 1 1
1 2 0 0
0·339 0·111 0·661 0·661
4 1 1 1 0 1 0 0
0 0 0 0 1 0 0 0
0·180 0·661 0·661 0·661 0·661 0·661
Complicated LAGB to LRYGB 70
50
40
30
20
10
Values are mean(s.d.). LRYGB, laparoscopic Roux-en-Y gastric bypass; LSG, laparoscopic sleeve gastrectomy. *Fisher’s exact test, except †t test.
was 16 of 39, 21 of 39, 23 of 34 and 13 of 27 respectively in the LRYGB group, and six of 19, ten of 19, seven of 18 and six of 15 in the LSG group. No statistically significant difference in %EWL was seen between the two conversion groups at less than 12 months’ follow-up, but at 12 and 24 months there was a statistically significant difference between the groups (P = 0·011 and P = 0·001 respectively) (Table 2). When separate analyses were made of non-responders and patients with band complications, mean %EWL was 55(22) per cent at 24 months in non-responders converted to LRYGB and 26(31) per cent in non-responders converted to LSG (P = 0·004). In contrast, mean %EWL was 42(15) per cent at 24 months in patients with band complications who were converted to LRYGB and 42(13) per cent in those converted to LSG (P = 0·629) (Fig. 1). There were no perioperative (30-day) deaths in either group. Five postoperative complications were recorded, two (13 per cent) in the 15 patients who had band complications. One patient, following removal of an eroded band 3 years before the conversion, developed acute complete obstruction of the proximal gastric sleeve that required conversion to LRYGB after 2 days. Another patient with conversion of a slipped gastric band developed early short stenosis at the sleeve and mega-oesophagus that required reintervention 4 months after conversion. This patient had repair of a large para-oesophageal hernia with non-absorbable suture. Three (7 per cent) of the 44 patients in the non-responders group had complications. One patient had peritonitis secondary to 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
Complicated LAGB to LSG
60
% EWL
Table 3
257
0
3
6
12
24
Follow-up (months)
Mean percentage excess weight loss (%EWL) at 3, 6, 12 and 24 months in non-responders to laparoscopic adjustable gastric banding (LAGB) and patients with complicated LAGB who were converted to laparoscopic Roux-en-Y gastric bypass (LRYGB) or laparoscopic gastric bypass (LSG)
Fig. 1
necrosis of the gastric fundus and was treated with laparoscopic exploration, suture and omentum patch repair on postoperative day 3. After lavage and repair, this patient was subsequently managed with drainage, bowel rest and total parenteral nutrition. One patient with alimentary limb intussusception required resection and redo jejunojejunostomy. In the third patient a small bowel obstruction that required lysis of adhesions occurred 9 days after conversion. All early complications were in patients who had a one-stage procedure. There were nine late complications (15 per cent), of which eight were in the LRYGB group. Four of these patients had small bowel obstruction, two of whom required small bowel resection (no internal hernias were identified), one patient had a marginal ulcer with bleeding, one required dilatation of a stricture at the gastrojejunostomy, and another patient required diagnostic laparoscopy for chronic abdominal pain. One patient who had conversion to LRYGB had deep vein thrombosis. In the LSG group, one patient developed dilatation of the sleeve proximal to distal compression, which resolved with operative adhesiolysis. Another patient converted to www.bjs.co.uk
BJS 2014; 101: 254–260
258
P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal
LSG subsequently required a LRYGB revision because of insufficient weight loss after 3 years. Discussion
Despite the popularity of gastric banding, many patients require reoperative surgery for band removal or conversion to another procedure because of insufficient weight loss and/or complications. One of the main controversies surrounding this clinical conundrum of need for conversion is selection of the procedure. Specific decision points include which procedure will yield the best outcomes, and which approach, single- or two-stage, best avoids perioperative complications. The literature lacks reports to help elucidate both of these questions. Highlighting the safety of laparoscopic single-stage conversion of a failed or complicated LAGB to LRYGB or LSG, this series showed no mortality and acceptable morbidity rates for both procedures, all of which were completed laparoscopically. LSG as a primary procedure has rapidly gained acceptance among surgeons owing to several advantages that it offers compared with other, more complex, procedures17 – 20 . In fact, LSG does not alter bowel continuity and is associated with a relatively low risk of mineral and vitamin deficiencies, with the exception of a still significant risk of vitamin B12 deficiency13,14 . Other advantages include the avoidance of dumping syndrome, marginal ulceration, malabsorption, internal hernia and foreign body material. Another advantage is that it can be used as the first step of a staged procedure in patients who fail to lose weight after an initial bariatric procedure. One of the major disadvantages of LSG is the risk of leakage at the oesophagogastric junction, which has been reported to be increased following revision of LAGB procedures to LSG20 – 22 . This is probably because stapling of the stomach is performed in scarred tissue and dissection of the left crus might jeopardize the vascular supply of the oesophagogastric junction20 – 22 . The overall complication rate for primary LSG is acceptable (5–10 per cent), and the mortality rate associated with this procedure is low17,21 . LSG is also a viable option for staged surgery in high-risk, superobese patients6,11 . In the present series, one of 20 patients undergoing LSG after failed LAGB developed a leak from the staple line at the level of the oesophagogastric junction. At 24 months, the overall (combining nonresponders and band complication groups) mean %EWL was 28(25) per cent, which is comparable to earlier published results for LSG as a revisional option7,20,22 . The weight loss data are at 24-month follow-up, so the results should be interpreted with caution, as longer follow-up 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
might show some weight regain. These results are worse than those for LRYGB, where %EWL of up to 60 per cent is achievable22 . In the present series there was a statistically significant difference in %EWL at 24 months for nonresponders converted to LRYGB and those converted to LSG (55(22) versus 26(31) per cent respectively). Based on the results presented, LSG does not seem to be the ideal operation when converting a failed LAGB. In the present series, 39 (66 per cent) of the 59 patients with a failed LAGB were converted to LRYGB, which is a technically more challenging procedure than LSG owing to the need to dissect both the angle of His and the lesser curve – the area of the greatest scarring and adhesions. Conversion to LRYGB is associated with a complication rate of less than 19 per cent22,23 . In the present study, only two (5 per cent) major postoperative complications occurred. The mean operating time was longer than for the average primary LRYGB, which is usually 110 min22 . This observation was noted in two other studies22,23 . With a %EWL of 55 per cent at 24 months, the present results are comparable to those already published23 . The overall complication rate in the present series was 22 per cent (13 of 59) with no deaths, which is in keeping with the literature (13–34 per cent)22,23 . The leak rate was 2 per cent (1 of 59), in a patient in whom the capsule was not taken down and 3·5-mm staples (blue cartridge linear staplers) were used. The necrosis of the gastric fundus was, in the authors’ opinion, most likely due to lack of blood supply and/or trauma. Patients who fail to lose weight or who regain weight after gastric banding (non-responders) with no apparent band complications are considered to have failure of restrictive procedures and the rescue procedure should include a malabsorptive component. LSG is probably a less desirable option in this scenario. Most bariatric surgeons do not consider converting a failed restrictive procedure such as LAGB to another restrictive procedure such as LSG to be a good strategy. Patients with bandassociated complications, such as slippage, pouch dilatation or band erosion, that occurred after successful weight loss can undergo conversion to any of the rescue procedures mentioned. In order to choose the most successful strategy, it is important to review the overall condition and comorbidities of the patient. Symptomatic reflux and diabetes are considered an indication for LRYGB, unless the patient refuses the procedure. Another alternative for revision of failed LAGB is BPD–DS, which is a more complex conversion with higher morbidity and mortality rates. Some authors have reported the safety of BPDDS as a reoperative intervention10 , but others11 have found higher morbidity and mortality rates (62 and 8 per cent respectively). However, a small study13 directly www.bjs.co.uk
BJS 2014; 101: 254–260
Laparoscopic conversion of failed adjustable gastric banding
comparing the standard revisional procedures of LRYGB and BPD–DS failed to demonstrate any superiority for one procedure over the other. BPD–DS, as either a primary or revisional procedure, is not routinely offered in the present authors’ practice. In their centre, the present authors perform both LRYGB and LSG as primary and revisional bariatric procedures. They routinely recommend a single-stage approach, and also recommend LRYGB. Patients have a choice of procedure provided they do not have severe gastro-oesophageal reflux disease, when the authors will perform only LRYGB, not LSG. In patients with eroded bands, the authors always undertake a two-stage approach. The overall complication rate for LSG as a revision of LAGB is higher than that of LSG as a primary procedure (13–34 per cent)18 – 20 . The different complication rates in primary versus revisional procedures are probably explained by tissue trauma and ischaemia. In addition, a compromised vascular supply of the superior aspect of the staple line, due to dissection of the left crus, may be a factor. Surgeons should balance the possibility of increased cost when managing a complicated conversion of LAGB to LSG performed in a single stage in comparison with an uncomplicated two-stage approach. Several weaknesses and strengths of this study need to be addressed. This series is relatively small and retrospective, with an intermediate follow-up of 24 months. Poor followup in this institution is related to many factors, including the fact that patients come from other countries or states to undergo reoperation. Additionally, in the USA, owing to the current economic situation, patients tend to move out of state, which can make it difficult to attain meaningful data on long-term success. Disclosure
The authors declare no conflict of interest. References 1 Suter M, Calmes JM, Paroz A, Giusti V. A 10-year experience with laparoscopic gastric banding for morbid obesity: high long-term complication and failure rates. Obes Surg 2006; 16: 829–835. 2 van Gemert WG, van Wersch MM, Greve JW, Soeters PB. Revisional surgery after vertical banded gastroplasty: restoration of vertical banded gastroplasty or conversion to gastric bypass. Obes Surg 1998; 8: 21–28. 3 Favretti F, Segato G, Ashton D, Busetto L, De Luca M, Mazza M et al. Laparoscopic adjustable gastric banding in 1791 consecutive obese patients: 12-year results. Obes Surg 2007; 17: 168–175. 2014 BJS Society Ltd Published by John Wiley & Sons Ltd
259
4 Gagner M, Gumbs AA. Gastric banding: conversion to sleeve, bypass, or DS. Surg Endosc 2007; 21: 1931–1935. 5 Cunneen SA. Review of meta-analytic comparisons of bariatric surgery with a focus on laparoscopic adjustable gastric banding. Surg Obes Relat Dis 2008; 4(Suppl): S47–S55. 6 Foletto M, Bernante P, Busetto L, Pomerri F, Vecchiato G, Prevedello L et al. Laparoscopic gastric rebanding for slippage with pouch dilation: results on 29 consecutive patients. Obes Surg 2008; 18: 1099 –1103. 7 Biertho L, Steffen R, Branson R, Potoczna N, Ricklin T, Piec G et al. Management of failed adjustable gastric banding. Surgery 2005; 137: 33–41. 8 Dapri G, Cadi`ere GB, Himpens J. Feasibility and technique of laparoscopic conversion of adjustable gastric banding to sleeve gastrectomy. Surg Obes Relat Dis 2009; 5: 72– 76. 9 Dargent J. Surgical treatment of morbid obesity by adjustable gastric band: the case for a conservative strategy in the case of failure – a 9-year series. Obes Surg 2004; 14: 986–990. 10 Topart P, Becouarn G, Ritz P. Biliopancreatic diversion with duodenal switch or gastric bypass for failed gastric banding: retrospective study from two institutions with preliminary results. Surg Obes Relat Dis 2007; 3: 521–525. 11 Spivak H, Beltran OR, Slavchev P, Wilson EB. Laparoscopic revision from LAP-BAND to gastric bypass. Surg Endosc 2007; 21: 1388–1392. 12 van Wageningen B, Berends FJ, Van Ramshorst B, Janssen IF. Revision of failed laparoscopic adjustable gastric banding to Roux-en-Y gastric bypass. Obes Surg 2006; 16: 137–141. 13 Di Betta E, Mittempergher F, Di Fabio F, Casella C, Terraroli C, Salerni B. Duodenal switch without gastric resection after failed gastric restrictive surgery for morbid obesity. Obes Surg 2006; 16: 258–261. 14 Patel S, Szomstein S, Rosenthal RJ. Reasons and outcomes of reoperative bariatric surgery for failed and complicated procedures (excluding adjustable gastric banding). Obes Surg 2011; 21: 1209–1219. 15 Patel S, Eckstein J, Acholonu E, Abu-Jaish W, Szomstein S, Rosenthal RJ. Reasons and outcomes of laparoscopic revisional surgery after laparoscopic adjustable gastric banding for morbid obesity. Surg Obes Relat Dis 2010; 6: 391–398. 16 Abu-Jaish W, Rosenthal RJ. Sleeve gastrectomy: a new surgical approach for morbid obesity. Expert Rev Gastroenterol Hepatol 2010; 4: 101–119. 17 Acholonu E, McBean E, Court I, Bellorin O, Szomstein S, Rosenthal RJ. Safety and short-term outcomes of laparoscopic sleeve gastrectomy as a revisional approach for failed laparoscopic adjustable gastric banding in the treatment of morbid obesity. Obes Surg 2009; 19: 1612–1616. 18 Buchwald H, Oien DM. Metabolic/bariatric surgery worldwide 2008. Obes Surg 2009; 19: 1605–1611.
www.bjs.co.uk
BJS 2014; 101: 254–260
260
P. Marin-Perez, A. Betancourt, M. Lamota, E. Lo Menzo, S. Szomstein and R. Rosenthal
19 Deitel M, Crosby RD, Gagner M. The First International Consensus Summit for Sleeve Gastrectomy (SG), New York City, October 25–27, 2007. Obes Surg 2008; 18: 487–496. 20 Rubin M, Yehoshua RT, Stein M, Lederfein D, Fichman S, Bernstine H et al. Laparoscopic sleeve gastrectomy with minimal morbidity. Early results in 120 morbidly obese patients. Obes Surg 2008; 18: 1567–1570. 21 Frezza EE, Jaramillo-de la Torre EJ, Enriquez CC, Gee L, Wachtel MS, Lopez Corvala JA. Laparoscopic sleeve
gastrectomy after gastric banding removal: a feasibility study. Surg Innov 2009; 16: 68–72. 22 Tucker O, Sucandy I, Szomstein S, Rosenthal RJ. Revisional surgery after failed laparoscopic adjustable gastric banding. Surg Obes Relat Dis 2008; 4: 740–747. 23 Mognol P, Chosidow D, Marmuse JP. Laparoscopic conversion of laparoscopic gastric banding to Roux-en-Y gastric bypass: a review of 70 patients. Obes Surg 2004; 14: 1349–1353.
Snapshot quiz
Snapshot quiz 14/2 Answer: Anastomotic leak at the enteroenterostomy. During the initial procedure, the integrity of the gastrojejunostomy was assessed using methylene blue dye. At reoperation, an anastomotic leak secondary to staple-line failure at the enteroenterostomy was identified. The blue discoloration of the skin resulted from leakage of the dye at this site and subsequent extravasation through the port site. In addition to the physiological deterioration on day 1, the bright blue ‘bruising’ was a further cutaneous indication of an underlying anastomotic leak.
2014 BJS Society Ltd Published by John Wiley & Sons Ltd
www.bjs.co.uk
BJS 2014; 101: 254–260
