Surgery for Obesity and Related Diseases ] (2015) 00–00
Original article
Laparoscopic Roux-en-Y gastric bypass for failed gastric banding: outcomes in 642 patients Pierre Fournier, M.D.a,b, Daniel Gero, M.D.b, Anna Dayer-Jankechova, M.D.a,c, Pierre Allemann, M.D.a, Nicolas Demartines, M.D.a, Jean-Pierre Marmuse, M.D.b, Michel Suter, M.D.a,c,* a
Department of Visceral Surgery, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland b Department of Digestive Surgery, Bichat-Claude-Bernard University Hospital, Paris, France c Department of Surgery, Riviera-Chablais Hospital, Aigle-Monthey, Switzerland Received March 16, 2015; accepted April 14, 2015
Abstract
Background: Laparoscopic adjustable gastric banding (LAGB) is a well-tolerated procedure but has high long-term complication and failure rates. Laparoscopic conversion to Roux-en-Y gastric bypass (LRYGB) is one of the rescue strategies. Objectives: To analyze short- and long-term results of reoperative LRYGB after failed LAGB. Setting: Three European expert bariatric center (2 university hospitals and 1 regional hospital). Methods: A retrospective review of prospectively collected data, including all consecutive patients submitted to revisional LRYGB for failed LAGB between 1999 and 2013, was performed. Complications were classified according to the Dindo-Clavien system. Long-term results in terms of weight loss were analyzed in a subgroup of patients. Results: A total of 642 patients (569 women and 73 men) were included. Mean ⫾ standard deviation operating time was 188 ⫾ 43 minutes. There was no mortality and an overall complication rate of 9.7%, including 3.6% major complications, with no difference between the 1- or 2-step approaches. Follow-up rate was 88% at 10 years for the Swiss patient cohort. The mean excess body mass index loss was between 65% and 70% throughout the study period, and the mean total weight loss was between 28% and 30% based on the maximum weight. The mean body mass index decreased from 44.7 kg/m2 before LAGB to 31.6, 32.2, and 32.5 kg/m2 at 1, 5, and 10 years after revision. Conclusions: Revisional LRYGB is well tolerated and feasible after failed LAGB. A 1-step approach, in cases without erosion, does not increase operative morbidity. Results up to 10 years after revision are comparable to those reported after primary LRYGB. (Surg Obes Relat Dis 2015;]:00–00.) r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Roux-en-Y gastric bypass; Gastric banding; Revisional surgery; Complications; Long-term
Worldwide obesity has nearly doubled since 1980, with more than 1.4 billion adults estimated to be overweight in *
Correspondence: Michel Suter, M.D., Department of Surgery, Hôpital Riviera-Chablais, 1860 Aigle, Switzerland. E-mail: [email protected]
2008 [1]. This epidemic of obesity represents an enormous burden for the healthcare systems in many countries by generating both direct and indirect costs. In morbid obesity, bariatric surgery is the only form of treatment that provides sustained long-term weight loss and its associated benefits in severely obese patients. It induces not only long-term
http://dx.doi.org/10.1016/j.soard.2015.04.007 1550-7289/r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
2
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
weight loss but also remission or improvement of obesityrelated co-morbidities, enhancement of quality of life, and an increase in long-term survival [2,3]. Because of its relative simplicity, gastric banding (LAGB) has been the first bariatric procedure performed by laparoscopy [4]. As such, and because of its very low operative morbidity, it has largely contributed to the boom in the demand for bariatric surgery throughout the world and has remained, until very recently, 1 of the 2 most commonly performed procedures, along with Roux-en-Y gastric bypass (RYGB). Early reports of LAGB were very enthusiastic, and most groups recognize that a small proportion of patients do extremely well even years after LAGB. Some authors still consider GB as the bariatric procedure of choice [5], but several others have shown increasing proportions of patients presenting with insufficient weight loss and/or severe long-term anatomic or functional complications requiring reoperation [6–9]. Many complications eventually lead to partial or total band deflation, and usually to weight regain, even with close follow-up [5,6]. As a consequence, an increasing number of patients who initially had LAGB eventually require major revisional surgery under general anesthesia. Options include simple band removal (almost inevitably associated with important weight regain), band revision (repositioning, replacement), or conversion to another bariatric procedure. Almost 20 years ago, conversion to RYGB was found to be the procedure of choice for the failed vertical banded gastroplasty, leading to better weight loss and fewer further reoperations than revision of the gastroplasty alone [10,11]. By analogy, and despite the fact that vertical banded gastroplasty is different from LAGB, the authors have used RYGB as the preferred revisional procedure in patients with failed LAGB, like many other authors [12–36]. The purpose of the present study was to analyze a multicenter experience with laparoscopic conversion of LAGB to LRYGB with regard to both short- and longterm results. Patients and methods Patients who underwent LRYGB in 1 or 2 steps after LAGB in the authors’ 3 expert bariatric centers between January 1999 and December 2013 were reviewed for the purpose of this study on the basis of 2 prospective databases, 1 of them incorporating all patients who underwent surgery performed by or under the direct supervision of the same surgeon in 2 hospitals. Patients were included independently from the fact that the initial procedure (LAGB) was performed in one of the authors’ centers or elsewhere. Before surgery, a careful and complete evaluation was performed by a multidisciplinary team, including a specialized endocrinologist, dieticians, psychologists, and a bariatric surgeon. Upper gastrointestinal contract studies
were performed routinely. In patients with the band still in place, these studies were used to check the position of the band, the size and shape of the pouch, and the size and motility of the esophagus. All patients were also submitted to esophagogastroscopy to rule out band erosion and to assess any other pathologic finding. Only a few patients had additional investigations, such as esophageal manometry or 24-hour pH study. In an attempt to overcome the complications or side effects of LAGB and at the same time maintain or achieve further weight loss, provided there was no obvious reason to rapidly remove the band, such as a band erosion or acute pouch dilation, all patients were first included in a therapeutic regimen consisting of intensified follow-up with repeated band adjustments, together with dietary and psychological counseling. Patients with former LAGB were eligible for conversion to LRYGB if, despite all these measures, they developed severe band-related complications causing a major negative effect on their quality of life or interference with the purpose of the procedure (i.e., maintenance of satisfactory weight loss and/or resolution or improvement of co-morbidities). All patients gave informed consent before reoperation, which included consent for the procedure as well as for follow-up, inclusion in institution databases, and deidentified outcome-data reporting. Because this is a retrospective study, no ethics committee was sought. Data collection included anthropometric data (age, sex, height, weight both before LAGB and at the time of LRYGB, and the minimum weight achieved since LAGB), perioperative data (duration of surgery, length of stay, operative morbidity and mortality), long-term data (weight, body mass index [BMI], excess BMI loss [EBMIL], total weight loss [TBWL]), and late complications, excluding micronutrient deficiencies. Early complications were classified according to the Dindo-Clavien system [37]. Life-threatening complications or those requiring reoperation (Dindo-Clavien IIIb-V) were considered major complications. Band removal and conversion to LRYGB was performed in 1 or 2 steps, based on the operating surgeon’s decision. Patients with band erosion, and those who consulted after having had their band removed elsewhere, were all treated in a 2-step fashion. Surgical technique All procedures were done by laparoscopy. Antibiotic prophylaxis and thromboembolic prophylaxis using low molecular weight heparin were used routinely. In patients operated on in 1 stage, the procedure usually began with the removal of the port. Access to the abdominal cavity was obtained using either a Verres needle or an open technique according to the surgeon’s preference. Adhesions between the left lobe of the liver, the lesser omentum, the stomach, and the band were divided carefully.
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00
Understanding the anatomy represented a crucial part of the procedure: the hiatus, the esophagogastric junction, the angle of His, and the exact location of the band in relation to the latter must be identified. The gastrogastric sutures were cut and the entire gastrogastric tunnel was opened. Once exposure of the complete anterior aspect of the band was achieved, the band was opened and removed, facilitating further dissection, especially at the level of the angle of His. Deciding where to create the pouch depended on the individual anatomy of the patient and on the former location of the band. In cases of slippage and/or pouch dilation, the stomach was divided above the band. If the band had slipped upward, division was made below the band. In most cases, however, the band was still approximately in the adequate location, and division of the stomach was performed, at least partially, at the level of the band. In these cases, the fibrous capsule that had formed between the band and the stomach was removed on the anterior stomach wall and up to the angle of His. If the band had been placed correctly above the lesser sac, the posterior capsule was located above the level at which the stomach was divided and did not need to be removed. Dissection usually started on the lesser curvature about 2 cm below the band, until the lesser sac was opened, and moved upward until the posterior peritoneal reflexion line of the lesser sac was identified. An attempt was always made to create a gastric pouch that appeared similar to the one created in a primary gastric bypass. The stomach was divided using a linear stapler and a first 45-mm cartridge placed horizontally about 3 cm below the esophagogastric junction, followed by 1 or 2 vertical cartridges aiming at the angle of His. Staple line reinforcement material was used for the vertical portion of the staple line depending on the surgeon’s preference. Once the pouch was created, the bypass was completed in a standard fashion, as in primary cases, using a retrocolic and retrogastric 100–150 cm Roux limb, a 21-mm circular stapled gastrojejunostomy, and a linear-stapled jejunojejunostomy. All mesenteric defects were closed. In patients operated with a 2-step approach, great care must be taken during the second step to avoid stapling multiple layers of the stomach. This was easy if the gastrogastric tunnel had been taken down during band removal but could be tricky in other cases. Recognizing the former tunnel during the second stage can be challenging; therefore, careful dissection was emphasized. Intraoperative endoscopy can be useful to help identifying important landmarks, especially the esophagogastric junction. A water-soluble contrast swallow was performed on the first postoperative day. If it proved negative, patients were slowly advanced on a semi-liquid diet that was maintained for 1 month, after which patients were allowed to progressively return to normal food. Follow-up included a first visit after 1 month, quarterly visits the first year, 2 visits the second year, and an annual control later on.
3
Follow-up rates between patients who underwent surgery in France and in Switzerland differed considerably, mostly because many patients in Paris were from out of the country or lived at a distance from Paris. For these reasons, they were usually followed-up by their general practitioner and did not want to travel for follow-up visits. According to current guidelines in Switzerland, however, patients have to be followed-up by the bariatric team. To provide meaningful data, the authors decided to analyze long-term results only in the Swiss cohort. Patients not seen for more than 1 year before a given time interval were considered as lost to follow-up. Statistical analysis Statistical analysis was performed using the SPSS 17.0 software package (Chicago, SPSS Inc.). The paired t test was used for 2 paired samples of continuous variables, and the categorical variables were analyzed by the χ2 test or the Fisher’s exact test. A value of P o .05 was considered statistically significant. Results A total of 642 patients were operated on during the study period. There were 569 women (88%) and 73 men (12%), with a mean age of 43 years. Table 1 displays the patients’ anthropometric characteristics at reoperation. A 1-step approach was used in 387 patients (60.2%), whereas 255 patients (39.8%) underwent LRYGB after previous removal of the band in a 2-step approach. The preoperative characteristics of the patients with respect to age, BMI, and sex ratio did not differ between these 2 groups. The mean duration of surgery was 188 ⫾ 41 minutes (range, 90–335 min) and was unrelated to whether surgery was performed in 1 or 2 steps. Overall early postoperative morbidity was 9.7% (Table 2). Minor complications developed in 39 (6.1%) patients, and major complications developed in 23 (3.6%) patients. There was no difference in the complication rates between patients operated on using a 1-step or a 2-step approach: 6% versus 6.3% (P ¼ .89) for minor complications, 4.2 versus 2.8 (P ¼ .33) for major complications. Learning curve played a role, and there were significantly fewer complications among the second half of patients compared with the first, earlier, half (3.7% versus 14.6%, P o .001, Fisher’s exact test). There was also a trend Table 1 Patients’ characteristics before conversion to Roux-en-Y gastric bypass Characteristic
1-step
2-step
P value
Sex (women/men) Body mass index at redo (kg/m2) Age, yr
340/37 44 (8) 43 (10)
219/36 43.6 (9.2) 43.5 (10)
.47 .60 .55
Data are presented as number or mean (SD) as appropriate.
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
4 Table 2 Early complications Type of complication
Overall complications Minor complications (Dindo-Clavien I–IIIa) Surgical site infection Fever of unknown origin Leak (gastrojejunostomy) Pulmonary embolism Hemorrhage Others Major complications (Dindo-Clavien IIIb–IV) Leak Hemorrhage Severe pneumonia Intestinal occlusion Left pleural empyema Anastomotic stricture Intraabdominal abscess Fever with sepsis of unknown origin Bile leak
1-step, n (%)
2-step, n (%)
P value
39 (10.2) 23 (6)
23 (9.1) 16 (6.3)
.68 .86
14 (3.6) 2 (.5) 1 (.2) 2 (.5) 3 (.8) 0 16 (4.2)
12 (4.7) 1 (.4) 0 0 0 3 (1.2) 7 (2.8)
.54 1 1 .52 .28 .06 .51
7 2 1 2 0 1 1 1
(.2) (.2) (.2)
3 2 1 0 1 0 0 0
.74 .65 1 .52 .39 1 1 1
1 (.2)
0
(1.8) (.5) (.2) (.5)
(1.2) (.8) (.4) (.4)
follow-up in this group, none of them directly related to the LRYGB (3 patients who committed suicide, 1 patient with multimetastatic carcinoma, and 1 with myocardial infarction). Fig. 1 shows the number of eligible patients at each time interval and the percentage of follow-up, which was 88% after 10 years. Twenty-two patients (12.4%) developed long-term complications, of whom 17 (9.6%) required reoperation, including 1 complete reversal for chronic abdominal pain of unknown origin and 3 requiring conversion to distal gastric bypass for insufficient weight loss or massive weight regain. The mean BMI decreased initially and remained relatively stable after the first year (Fig. 2), and around 75% of the patients maintained a BMI o35 kg/m2 throughout the study period (Fig. 3). Fig. 4 shows the percentage of EBMIL compared with prebanding weight and Fig. 5 the percentage of TBWL, taking into account the maximal weight of the patients, either at LAGB or at LRYGB. Whatever the reference value, the mean EBMIL remained between 65% and 70% up to 10 years after conversion to RYGB, and the mean TBWL remained between 28% and 30%.
1
Discussion toward fewer major complications in the most recent half (2.5% versus 4.7%, P ¼ .2, Fisher’s exact test). There was no 30-day mortality. The mean duration of hospital stay was 5.9 ⫾ 4.1 days (range 3–29). As previously mentioned, long-term results are provided for patients from the Swiss cohort only (177 patients). At the time of LRYGB, 135 of these patients had maintained some weight loss compared with the pre-LAGB weight (from –1 to –83 kg), 4 had the same weight as before LAGB, and 38 had gained between 1 and 46 kg compared with their pre-LAGB weight. Five deaths occurred during
This study has 3 major findings. First, conversion to LRYGB after failed LAGB can be done by a laparoscopic approach with a positive safety profile. Second, in the absence of band erosion, a 1-step approach does not increase perioperative morbidity compared with a 2-step approach (band removal first, followed some time later by LRYGB). Third, results after conversion to LRYGB are very satisfactory up to 10 years after redo surgery, with a mean EBMIL between 65% and 70% in relation to the preband weight, corresponding to roughly 75% of the patients maintaining a BMI o35 kg/m2.
Fig. 1. Number of patients eligible for follow-up in the Swiss cohort (numbers below columns) and follow-up rates over time. LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00
5
Fig. 2. Evolution of the mean body mass index (⫾ SD) over time in the Swiss patient cohort. BMI ¼ body mass index; LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
LAGB is a relatively simple, adjustable, and reversible low-risk laparoscopic procedure that preserves the digestive tract. After promising early results, LAGB progressively lost its popularity in many European countries and in North America, due both to poor long-term outcomes in terms of weight loss and high long-term complication and reoperation rates. Common indications for reoperation after failed LAGB include insufficient weight loss or weight regain, complications associated with the implants (erosion, slippage, port and tubing problems), or anatomic or functional problems such as pouch dilation, progressive late food intolerance, intractable gastroesophageal reflux disease, and esophageal dilation and/or dysmotility. Several recent studies reported long-term reoperations in 31%–80% of patients after LAGB [5–9,13]. Reoperation for failed LAGB can involve simple revision (i.e., band change or repositioning) or conversion to another bariatric operation. Although some authors suggest that simple revision can be effective in patients who had
satisfactory weight loss after LAGB before they developed a complication [12,23,38], others support systematic conversion to another procedure, especially in patients with primary inadequate weight loss [12,24,25]. Simple revision further exposes the patient to all potential long-term complications of LAGB, with their inherent risks of further need for redo surgery. Revisional bariatric surgery is technically demanding and is usually associated with higher intraoperative and perioperative risks than primary procedures, notably of gastrointestinal complications [39–41]. According to a systematic review by the American Society for Metabolic and Bariatric Surgery Revision Task Force, these procedures should only be done by experienced bariatric surgeons in centers with sufficient resources [39]. To the best of the authors’ knowledge, the present study is the largest reporting on revisional LRYGB after failed LAGB. It indicates that conversion to LRYGB after failed LAGB can be performed with acceptable perioperative
Fig. 3. Distribution of body mass index over time. BMI ¼ body mass index; LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
6
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
Fig. 4. Percentage of excess body mass index lost before gastric bypass and over time in reference to the initial body mass index (before gastric banding). EBMIL ¼ excess body mass index lost; LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
morbidity and no mortality when performed by experienced surgeons. The overall complication rate of 9.6% lies in the lower range of that reported among studies including at least 50 patients undergoing conversion from band to bypass (7%–46.3%), and the major morbidity of 3.6% is within usually reported rates (1.6%–13.4%) [26–36,40,42,43]. In fact, the complication rate reported herein is similar to that of many studies of primary RYGB [44,45], including the authors’ own experience. As stated by others [36], experience plays a role in postoperative morbidity. This is confirmed in the present study, in which morbidity was much lower among the second, later, half of patients. Leak at the gastrojejunostomy or on the gastric pouch is a complication intuitively expected to be more common in revisional gastric bypass than in primary cases, because stapling on the stomach is done in fibrous or inflammatory tissue. The leak rate of 1.7% lies within the range reported for primary gastric bypass in many large series [44]. Attention to technical details is important to prevent leaks. Whenever possible, stapling should be done away from band-related scar tissue. If necessary, the fibrous capsule formed around the band must be removed before stapling.
Finally, staple height must be adjusted to tissue thickness, and larger staples (green or black loads) are routinely used for the vertical portion of the gastric pouch in revisional cases. Staple-line reinforcement material may also be of help, but was used rarely in the present study. Whether conversion from failed LAGB to LRYGB should be performed in 1 or 2 steps remains controversial. A 1-step approach has clear advantages, such as decreased overall operative room, anesthesia, and hospital time; decreased patient waiting time; and better allocation of resources. It may also be beneficial if there is concern for reimbursement. Finally, in a 1-stage approach, the band can be used as a guide for dissection and recognition of the anatomy. In contrast, proponents of a 2-step approach argue that the first stage (band removal) is relatively easy and straightforward with low morbidity. Time between the first and second steps is usually associated with weight regain, but can allow for local inflammation to subside, potentially leading to reduced morbidity. In the present study, where 60% of the conversions were done in 1 step, the complication rate was not increased with this approach and was similar to the 9% overall and 4% major complication rates
Fig. 5. Evolution of the percentage of total weight loss ⫾ SD in reference to the maximum weight achieved by the patients either before gastric banding or at gastric bypass. %TWL ¼ total weight loss; RYGB ¼ Roux-en-Y gastric bypass.
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00
recently reported by Aarts et al. with the same 1-step approach [26]. In their study, Tran et al. also failed to show any difference in this respect [46], but in another small study, Van Nieuwenhove et al. found a trend toward fewer complications with a 2-stage approach, at the expense of increased operation time and hospital stay [47]. The currently available literature and the present results suggest that most patients can be converted in 1 step, except those with band erosion, infection, or very active inflammation at the time of band removal. If a 2-step approach is chosen, the authors strongly advise not only removal of the band during the first step, but also complete removal of the gastrogastric sutures and opening of the tunnel because it may be very difficult to recognize these structures later on. Results in terms of weight loss after conversion from LAGB to RYGB are poorly reported, because duration of follow-up is limited to 2–3 years or even less in most studies. Furthermore, many studies show results based on weight before redo surgery and not initial weight at the time of LAGB. Because the mean weight at redo is generally inferior to that before primary surgery, results are difficult to interpret, and a meaningful comparison with results of primary RYGB is hard to establish. In a small group of patients, Perathoner et al. reported a mean BMI of 28.8 kg/m2 5 years after conversion in a small group of patients [20]. Aarts et al. found a mean EWL of 53% compared with prebanding weight 5 years after conversion in a group of 54 patients [26]. In the present study, with a very high followup rate, a mean EBMIL (related to the pre-LAGB BMI) of 65%–70% up to 10 years after conversion is shown, corresponding to a mean BMI of 32–33 kg/m2 as of the first year after revisional LRYGB. This corresponds to 75%–80% of the patients maintaining a BMI o35 for up to 10 years, but also means that a minority of patients fails to achieve satisfactory weight despite 2 bariatric procedures. It is worth noting, however, that 6 of the 9 patients whose BMI was 435 kg/m2 10 years after redo surgery were in the super-obese category at the time of GB and that these 9 patients had an average of 9.3 (3.6–20.4) BMI units lost at 10 years, compared with the initial figure. In the authors’ experience, these results are similar to those obtained after primary LRYGB (unpublished data). Despite the multicenter setting and the large size of the cohort, the present study has some limitations. First, in patients who presented with the band still in place, the decision to proceed in 1 or 2 steps was biased by the surgeon’s discretion. Therefore, caution should be applied when interpreting comparisons between these 2 approaches. Second, only a fraction of the patients who underwent surgery were included in the long-term analysis regarding weight loss. Although the latter group had an excellent follow-up rate of 88% at 10 years, it may not represent the entire study population, especially with respect to the fact that the average pre-LRYGB weight in the French group was significantly higher than in the Swiss cohort. Third,
7
results regarding the effect of weight loss on co-morbidities could not be analyzed because of lack of data, especially before LAGB. On the other hand, strengths of the present study are the large sample size, which allows for some subgroup comparisons; the long duration of follow-up; and the high follow-up rate. Conclusions Laparoscopic conversion of a failed gastric band into RYGB is a relatively difficult but safe procedure. It is associated with morbidity similar to that of primary LRYGB and can be done without any mortality. In experienced hands, a 1-step approach is safe in most cases and does not increase the complication rate. Up to 10 years after redo surgery, results in terms of weight loss are comparable to those observed after primary LRYGB. Further studies, ideally randomized, will be necessary to draw definitive conclusions about the safety of a routine 1-step approach. More studies are also needed to assess long-term results, especially about resolution or improvement of co-morbidities. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] WHO Fact sheet No 311. Obesity and overweight [homepage on the Internet]. World Health Organization [cited 2014 Sep 20]. Available from: http://www.who.int/mediacentre/factsheets/fs311/en/. [2] Schauer PR, Bhatt DL, Kashyap SR. Bariatric surgery versus intensive medical therapy for diabetes. N Engl J Med 2014;371 (7):682. [3] Colquitt JL, Picot J, Loveman E, Clegg AJ. Surgery for obesity. Cochrane Database Syst Rev 2009;2. CD003641. [4] Belachew M, Legrand MJ, Defechereux TH, Burtheret MP, Jacquet N. Laparoscopic adjustable silicone gastric banding in the treatment of morbid obesity. A preliminary report. Surg Endosc 1994;8 (11):1354–6. [5] O’Brien PE, MacDonald L, Anderson M, Brennan L, Brown WA. Long-term outcome after bariatric surgery: fifteen-year follow-up of adjustable gastric banding and a systematic review of the surgical bariatric literature. Ann Surg 2013;257(1):87–94. [6] Gero D, Dayer-Jankechova A, Worreth M, Giusti V, Suter M. Laparoscopic gastric banding outcomes do not depend on device or technique. Long-term results of a prospective randomized study comparing the Lapbands and the SAGBs. Obes Surg 2014;24 (1):114–22. [7] Aarts EO, Dogan K, Koehestanie P, Anfenacker TJ, Janssen IMC, Berends FJ. Long-term results after laparoscopic adjustable gastric banding: a mean fourteen year follow-up study. Surg Obes Relat Dis 2014;10(4):633–40. [8] Naef M, Mouton WG, Naef U, Kummer O, Muggli B, Wagner HE. Graft survival and complications after laparoscopic gastric banding for morbid obesity—lessons learned from a 12-year experience. Obes Surg 2010;20(9):1206–14.
8
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
[9] Himpens J, Cadière GB, Bazi M, Vouche M, Cadière B, Dapri G. Long-term outcomes of laparoscopic adjustable gastric banding. Arch Surg 2011;146(7):802–7. [10] Van Gemert WG, Van Wersch MM, Greve JWM, Soeters PB. Revisional surgery after failed vertical banded gastroplasty: restoration of vertical banded gastroplasty or conversion to gastric bypass. Obes Surg 2008;8(1):21–8. [11] Behrns KE, Smith CD, Kelly KA, Sarr MG. Reoperative bariatric surgery. Lessons learned to improved patient selection and results. Ann Surg 1993;218(6):646–53. [12] Ardestani A, Lautz DB, Tavakkolizadeh A. Band revision versus Roux-en-Y gastric bypass conversion as salvage operation after laparoscopic adjustable gastric banding. Surg Obes Relat Dis 2011;7 (1):33–7. [13] 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(4):391–8. [14] Park JY, Song D, Kim YD. Causes and outcomes of revisional bariatric surgery: initial experience at a single center. Ann Surg Treat Res 2014;86(6):295–301. [15] Kruger RS, Pricolo VE, Streeter TT, Colacchio DA, Andrade UA. A bariatric surgery center of excellence: operative trends and longterm outcomes. J Am Coll Surg 2014;218(6):1163–74. [16] Stroh C, Weiner R, Wolff S, Knoll C, Manger T. Arbeitsgruppe Adipositaschirurgie, Kompetenznetz Adipositas. Revisional surgery and reoperations in obesity and metabolic surgery: data analysis of the German bariatric surgery registry 2005–2012 [in German]. Chirurg 2015;86(4):346–54. [17] Fernando Santos B, Wallaert JB, Trus TL. Band removal and conversion to sleeve or bypass: are they equally safe? Surg Endosc 2014;28(11):3086–91 [18] Coblijn UK, Verveld CJ, van Wagensveld BA, Lagarde SM. Laparoscopic Roux-en-Y gastric bypass or laparoscopic sleeve gastrectomy as revisional procedure after adjustable gastric band—a systematic review. Obes Surg 2013;23(11):1899–914. [19] Van Nieuwenhove Y, Ceelen W, Van Renterghem K, Van de Putte D, Henckens T, Pattyn P. Conversion from band to bypass in two steps reduces the risk for anastomotic strictures. Obes Surg 2011;21 (4):501–5. [20] Perathoner A, Zitt M, Lanthaler M, Pratschke J, Biebl M, Mittermair R. Long-term follow-up evaluation of revisional gastric bypass after failed adjustable gastric banding. Surg Endosc 2013;27(11):4305–12. [21] Marin-Perez P, Betancourt A, Lamota M, Lo Menzo E, Szomstein S, Rosenthal R. Outcomes after laparoscopic conversion of failed adjustable gastric banding to sleeve gastrectomy or Roux-en-Y gastric bypass. Br J Surg 2014;101(2):254–6. [22] Elnahas A, Graybiel K, Farrokhyar F, Gmora S, Anvari M, Hong D. Revisional surgery after failed laparoscopic adjustable gastric banding: a systematic review. Surg Endosc 2013;27(3):740–5. [23] Bueter M, Thalheimer A, Wierlemann A, Fein M. Reoperations after gastric banding: replacement or alternative procedures. Surg Endosc 2009;23(2):334–40. [24] Lanthaler M, Mittermair R, Erne B, Weiss H, Aigner F, Nehoda H. Laparoscopic gastric re-banding versus laparoscopic gastric bypass as a rescue operation for patients with pouch dilatation. Obes Surg 2006;16(4):484–7. [25] Weber M, Muller MK, Michel JM, et al. Laparoscopic Roux-en-Y gastric bypass, but not rebanding, should be proposed as rescue procedure for patients with failed laparoscopic gastric banding. Ann Surg 2003;238(6):827–34. [26] Aarts E, Koehestanie P, Dogan K, Berends F, Janssen I. Revisional surgery after failed gastric banding: results of one-stage conversion to RYGB in 195 patients. Surg Obes Relat Dis 2014;10(6):1077–84.
[27] Deylgat B, D’Hondt M, Pottel H, Vansteenkiste F, Van Rooy F, Devriendt D. Indications, safety, and feasibility of conversion of failed bariatric surgery to Roux-en-Y gastric bypass: a retrospective comparative study with primary laparoscopic Roux-en-Y gastric bypass. Surg Endosc 2012;26(7):1997–2002. [28] Jennings NA, Boyle M, Mahawar K, Balupuri S, Small PK. Revisional laparoscopic Roux-en-Y gastric bypass following failed adjustable gastric banding. Obes Surg 2013;23(7):947–52. [29] Robert M, Poncet G, Boulez J, Mion F, Espalieu P. Laparoscopic gastric bypass for failure of adjustable gastric banding: a review of 85 cases. Obes Surg 2011;21(10):1513–9. [30] Slegtenhorst BR, Van der Harst E, Demirkiran A, de Korte J, Schelfhout LJ, Klaassen RA. Effect of primary versus revisional Roux-en-Y gastric bypass: inferior weight loss of revisional surgery after gastric banding. Surg Obes Relat Dis 2013;9(2):253–9. [31] te Riele W, Sze YK, Wiezer MJ, van Ramhorst B. Conversion of failed laparoscopic gastric banding to gastric bypass as safe and effective as primary gastric bypass in morbidly obese patients. Surg Obes Relat Dis 2008;4(6):735–9. [32] Thereaux J, Veyrie N, Barsamian C, et al. Similar postoperative safety between primary and revisional gastric bypass for failed gastric banding. JAMA Surg 2014;149(8):780–6. [33] Cadière GB, Himpens J, Bazi M, et al. Are laparoscopic gastric bypass after gastroplasty and primary gastric bypass similar in terms of results? Obes Surg 2011;21(6):692–8. [34] 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(10):1349–53. [35] Hii MW, Lake AC, Kenfield C, Hopkins GH. Laparoscopic conversion of failed gastric banding to Roux-en-Y gastric bypass. Shortterm follow-up and technical considerations. Obes Surg 2012;22 (7):1022–8. [36] Alpers JA, Wens C, Van Vlodrop V, et al. Perioperative outcomes of revisional laparoscopic gastric bypass after failed adjustable gastric banding and after vertical banded gastroplasty: experience with 107 cases and subgroup analysis. Surg Endosc 2013;27(2):558–64. [37] Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–13. [38] Shouten R, Japink D, Meesters B, Nelemans PJ, Greve JWM. Systematic literature review of reoperations after gastric banding: is a stepwise approach justified? Surg Obes Relat Dis 2011;7(1):99–109. [39] Brethauer SA, Kothari S, Sudan R, et al. Systematic review on reoperative bariatric surgery: American Society for Metabolic and Bariatric Surgery Revision Task Force. Surg Obes Relat Dis 2014;10(5):952–72. [40] Worni M, Østbye T, Shah A, et al. High risks for adverse outcomes after gastric bypass surgery following failed gastric banding—a population-based trend analysis of the United States. Ann Surg 2013;257(2):279–86. [41] Stroh C, Benedix D, Weiner R, et al. Is a one-step sleeve gastrectomy indicated as a revision procedure after gastric banding? Data analysis from a quality assurance study of the surgical treatment of obesity in Germany. Obes Surg 2014;24(1):9–14. [42] Carr WR, Jennings NA, Boyle M, Mahawar K, Balupuri S, Small PK. A retrospective comparison of early results of conversion of failed gastric banding to sleeve gastrectomy or gastric bypass. Surg Obes Relat Dis. Epub 2014 Aug 26. [43] Carandina S, Maldonado PS, Tabbara M, et al. Two-step conversion surgery after failed laparoscopic adjustable gastric banding. Comparison between laparoscopic Roux-en-Y gastric bypass and laparoscopic gastric sleeve. Surg Obes Relat Dis 2014;10(6):1085–9. [44] Zellmer JD, Mathiason MA, Kallies KJ, Kothari SN. Is laparoscopic sleeve gastrectomy a lower risk bariatric procedure compared with
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00 laparoscopic Roux-en-Y gastric bypass? A meta-analysis. Am J Surg 2014;208(6):903–10. [45] Carlin AM, Zeni TM, English WJ, et al. The comparative effectiveness of sleeve gastrectomy, gastric bypass, and adjustable gastric banding procedures for the treatment of morbid obesity. Ann Surg 2013;257(5):791–7.
9
[46] Tran TT, Pauli E, Lyn-Sue JR, Haluck R, Rogers AM. Revisional weight loss surgery after failed laparoscopic gastric banding: an institutional experience. Surg Endosc 2013;27(11):4087–93. [47] Van Nieuwenhove Y, Ceelen W, Van Renterghem K, Van de Putte D, Henckens T, Pattyn P. Conversion from band to bypass in two steps reduces the risk for anastomotic strictures. Obes Surg 2011;21(4):501–5.
Original article
Laparoscopic Roux-en-Y gastric bypass for failed gastric banding: outcomes in 642 patients Pierre Fournier, M.D.a,b, Daniel Gero, M.D.b, Anna Dayer-Jankechova, M.D.a,c, Pierre Allemann, M.D.a, Nicolas Demartines, M.D.a, Jean-Pierre Marmuse, M.D.b, Michel Suter, M.D.a,c,* a
Department of Visceral Surgery, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland b Department of Digestive Surgery, Bichat-Claude-Bernard University Hospital, Paris, France c Department of Surgery, Riviera-Chablais Hospital, Aigle-Monthey, Switzerland Received March 16, 2015; accepted April 14, 2015
Abstract
Background: Laparoscopic adjustable gastric banding (LAGB) is a well-tolerated procedure but has high long-term complication and failure rates. Laparoscopic conversion to Roux-en-Y gastric bypass (LRYGB) is one of the rescue strategies. Objectives: To analyze short- and long-term results of reoperative LRYGB after failed LAGB. Setting: Three European expert bariatric center (2 university hospitals and 1 regional hospital). Methods: A retrospective review of prospectively collected data, including all consecutive patients submitted to revisional LRYGB for failed LAGB between 1999 and 2013, was performed. Complications were classified according to the Dindo-Clavien system. Long-term results in terms of weight loss were analyzed in a subgroup of patients. Results: A total of 642 patients (569 women and 73 men) were included. Mean ⫾ standard deviation operating time was 188 ⫾ 43 minutes. There was no mortality and an overall complication rate of 9.7%, including 3.6% major complications, with no difference between the 1- or 2-step approaches. Follow-up rate was 88% at 10 years for the Swiss patient cohort. The mean excess body mass index loss was between 65% and 70% throughout the study period, and the mean total weight loss was between 28% and 30% based on the maximum weight. The mean body mass index decreased from 44.7 kg/m2 before LAGB to 31.6, 32.2, and 32.5 kg/m2 at 1, 5, and 10 years after revision. Conclusions: Revisional LRYGB is well tolerated and feasible after failed LAGB. A 1-step approach, in cases without erosion, does not increase operative morbidity. Results up to 10 years after revision are comparable to those reported after primary LRYGB. (Surg Obes Relat Dis 2015;]:00–00.) r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Roux-en-Y gastric bypass; Gastric banding; Revisional surgery; Complications; Long-term
Worldwide obesity has nearly doubled since 1980, with more than 1.4 billion adults estimated to be overweight in *
Correspondence: Michel Suter, M.D., Department of Surgery, Hôpital Riviera-Chablais, 1860 Aigle, Switzerland. E-mail: [email protected]
2008 [1]. This epidemic of obesity represents an enormous burden for the healthcare systems in many countries by generating both direct and indirect costs. In morbid obesity, bariatric surgery is the only form of treatment that provides sustained long-term weight loss and its associated benefits in severely obese patients. It induces not only long-term
http://dx.doi.org/10.1016/j.soard.2015.04.007 1550-7289/r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
2
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
weight loss but also remission or improvement of obesityrelated co-morbidities, enhancement of quality of life, and an increase in long-term survival [2,3]. Because of its relative simplicity, gastric banding (LAGB) has been the first bariatric procedure performed by laparoscopy [4]. As such, and because of its very low operative morbidity, it has largely contributed to the boom in the demand for bariatric surgery throughout the world and has remained, until very recently, 1 of the 2 most commonly performed procedures, along with Roux-en-Y gastric bypass (RYGB). Early reports of LAGB were very enthusiastic, and most groups recognize that a small proportion of patients do extremely well even years after LAGB. Some authors still consider GB as the bariatric procedure of choice [5], but several others have shown increasing proportions of patients presenting with insufficient weight loss and/or severe long-term anatomic or functional complications requiring reoperation [6–9]. Many complications eventually lead to partial or total band deflation, and usually to weight regain, even with close follow-up [5,6]. As a consequence, an increasing number of patients who initially had LAGB eventually require major revisional surgery under general anesthesia. Options include simple band removal (almost inevitably associated with important weight regain), band revision (repositioning, replacement), or conversion to another bariatric procedure. Almost 20 years ago, conversion to RYGB was found to be the procedure of choice for the failed vertical banded gastroplasty, leading to better weight loss and fewer further reoperations than revision of the gastroplasty alone [10,11]. By analogy, and despite the fact that vertical banded gastroplasty is different from LAGB, the authors have used RYGB as the preferred revisional procedure in patients with failed LAGB, like many other authors [12–36]. The purpose of the present study was to analyze a multicenter experience with laparoscopic conversion of LAGB to LRYGB with regard to both short- and longterm results. Patients and methods Patients who underwent LRYGB in 1 or 2 steps after LAGB in the authors’ 3 expert bariatric centers between January 1999 and December 2013 were reviewed for the purpose of this study on the basis of 2 prospective databases, 1 of them incorporating all patients who underwent surgery performed by or under the direct supervision of the same surgeon in 2 hospitals. Patients were included independently from the fact that the initial procedure (LAGB) was performed in one of the authors’ centers or elsewhere. Before surgery, a careful and complete evaluation was performed by a multidisciplinary team, including a specialized endocrinologist, dieticians, psychologists, and a bariatric surgeon. Upper gastrointestinal contract studies
were performed routinely. In patients with the band still in place, these studies were used to check the position of the band, the size and shape of the pouch, and the size and motility of the esophagus. All patients were also submitted to esophagogastroscopy to rule out band erosion and to assess any other pathologic finding. Only a few patients had additional investigations, such as esophageal manometry or 24-hour pH study. In an attempt to overcome the complications or side effects of LAGB and at the same time maintain or achieve further weight loss, provided there was no obvious reason to rapidly remove the band, such as a band erosion or acute pouch dilation, all patients were first included in a therapeutic regimen consisting of intensified follow-up with repeated band adjustments, together with dietary and psychological counseling. Patients with former LAGB were eligible for conversion to LRYGB if, despite all these measures, they developed severe band-related complications causing a major negative effect on their quality of life or interference with the purpose of the procedure (i.e., maintenance of satisfactory weight loss and/or resolution or improvement of co-morbidities). All patients gave informed consent before reoperation, which included consent for the procedure as well as for follow-up, inclusion in institution databases, and deidentified outcome-data reporting. Because this is a retrospective study, no ethics committee was sought. Data collection included anthropometric data (age, sex, height, weight both before LAGB and at the time of LRYGB, and the minimum weight achieved since LAGB), perioperative data (duration of surgery, length of stay, operative morbidity and mortality), long-term data (weight, body mass index [BMI], excess BMI loss [EBMIL], total weight loss [TBWL]), and late complications, excluding micronutrient deficiencies. Early complications were classified according to the Dindo-Clavien system [37]. Life-threatening complications or those requiring reoperation (Dindo-Clavien IIIb-V) were considered major complications. Band removal and conversion to LRYGB was performed in 1 or 2 steps, based on the operating surgeon’s decision. Patients with band erosion, and those who consulted after having had their band removed elsewhere, were all treated in a 2-step fashion. Surgical technique All procedures were done by laparoscopy. Antibiotic prophylaxis and thromboembolic prophylaxis using low molecular weight heparin were used routinely. In patients operated on in 1 stage, the procedure usually began with the removal of the port. Access to the abdominal cavity was obtained using either a Verres needle or an open technique according to the surgeon’s preference. Adhesions between the left lobe of the liver, the lesser omentum, the stomach, and the band were divided carefully.
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00
Understanding the anatomy represented a crucial part of the procedure: the hiatus, the esophagogastric junction, the angle of His, and the exact location of the band in relation to the latter must be identified. The gastrogastric sutures were cut and the entire gastrogastric tunnel was opened. Once exposure of the complete anterior aspect of the band was achieved, the band was opened and removed, facilitating further dissection, especially at the level of the angle of His. Deciding where to create the pouch depended on the individual anatomy of the patient and on the former location of the band. In cases of slippage and/or pouch dilation, the stomach was divided above the band. If the band had slipped upward, division was made below the band. In most cases, however, the band was still approximately in the adequate location, and division of the stomach was performed, at least partially, at the level of the band. In these cases, the fibrous capsule that had formed between the band and the stomach was removed on the anterior stomach wall and up to the angle of His. If the band had been placed correctly above the lesser sac, the posterior capsule was located above the level at which the stomach was divided and did not need to be removed. Dissection usually started on the lesser curvature about 2 cm below the band, until the lesser sac was opened, and moved upward until the posterior peritoneal reflexion line of the lesser sac was identified. An attempt was always made to create a gastric pouch that appeared similar to the one created in a primary gastric bypass. The stomach was divided using a linear stapler and a first 45-mm cartridge placed horizontally about 3 cm below the esophagogastric junction, followed by 1 or 2 vertical cartridges aiming at the angle of His. Staple line reinforcement material was used for the vertical portion of the staple line depending on the surgeon’s preference. Once the pouch was created, the bypass was completed in a standard fashion, as in primary cases, using a retrocolic and retrogastric 100–150 cm Roux limb, a 21-mm circular stapled gastrojejunostomy, and a linear-stapled jejunojejunostomy. All mesenteric defects were closed. In patients operated with a 2-step approach, great care must be taken during the second step to avoid stapling multiple layers of the stomach. This was easy if the gastrogastric tunnel had been taken down during band removal but could be tricky in other cases. Recognizing the former tunnel during the second stage can be challenging; therefore, careful dissection was emphasized. Intraoperative endoscopy can be useful to help identifying important landmarks, especially the esophagogastric junction. A water-soluble contrast swallow was performed on the first postoperative day. If it proved negative, patients were slowly advanced on a semi-liquid diet that was maintained for 1 month, after which patients were allowed to progressively return to normal food. Follow-up included a first visit after 1 month, quarterly visits the first year, 2 visits the second year, and an annual control later on.
3
Follow-up rates between patients who underwent surgery in France and in Switzerland differed considerably, mostly because many patients in Paris were from out of the country or lived at a distance from Paris. For these reasons, they were usually followed-up by their general practitioner and did not want to travel for follow-up visits. According to current guidelines in Switzerland, however, patients have to be followed-up by the bariatric team. To provide meaningful data, the authors decided to analyze long-term results only in the Swiss cohort. Patients not seen for more than 1 year before a given time interval were considered as lost to follow-up. Statistical analysis Statistical analysis was performed using the SPSS 17.0 software package (Chicago, SPSS Inc.). The paired t test was used for 2 paired samples of continuous variables, and the categorical variables were analyzed by the χ2 test or the Fisher’s exact test. A value of P o .05 was considered statistically significant. Results A total of 642 patients were operated on during the study period. There were 569 women (88%) and 73 men (12%), with a mean age of 43 years. Table 1 displays the patients’ anthropometric characteristics at reoperation. A 1-step approach was used in 387 patients (60.2%), whereas 255 patients (39.8%) underwent LRYGB after previous removal of the band in a 2-step approach. The preoperative characteristics of the patients with respect to age, BMI, and sex ratio did not differ between these 2 groups. The mean duration of surgery was 188 ⫾ 41 minutes (range, 90–335 min) and was unrelated to whether surgery was performed in 1 or 2 steps. Overall early postoperative morbidity was 9.7% (Table 2). Minor complications developed in 39 (6.1%) patients, and major complications developed in 23 (3.6%) patients. There was no difference in the complication rates between patients operated on using a 1-step or a 2-step approach: 6% versus 6.3% (P ¼ .89) for minor complications, 4.2 versus 2.8 (P ¼ .33) for major complications. Learning curve played a role, and there were significantly fewer complications among the second half of patients compared with the first, earlier, half (3.7% versus 14.6%, P o .001, Fisher’s exact test). There was also a trend Table 1 Patients’ characteristics before conversion to Roux-en-Y gastric bypass Characteristic
1-step
2-step
P value
Sex (women/men) Body mass index at redo (kg/m2) Age, yr
340/37 44 (8) 43 (10)
219/36 43.6 (9.2) 43.5 (10)
.47 .60 .55
Data are presented as number or mean (SD) as appropriate.
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
4 Table 2 Early complications Type of complication
Overall complications Minor complications (Dindo-Clavien I–IIIa) Surgical site infection Fever of unknown origin Leak (gastrojejunostomy) Pulmonary embolism Hemorrhage Others Major complications (Dindo-Clavien IIIb–IV) Leak Hemorrhage Severe pneumonia Intestinal occlusion Left pleural empyema Anastomotic stricture Intraabdominal abscess Fever with sepsis of unknown origin Bile leak
1-step, n (%)
2-step, n (%)
P value
39 (10.2) 23 (6)
23 (9.1) 16 (6.3)
.68 .86
14 (3.6) 2 (.5) 1 (.2) 2 (.5) 3 (.8) 0 16 (4.2)
12 (4.7) 1 (.4) 0 0 0 3 (1.2) 7 (2.8)
.54 1 1 .52 .28 .06 .51
7 2 1 2 0 1 1 1
(.2) (.2) (.2)
3 2 1 0 1 0 0 0
.74 .65 1 .52 .39 1 1 1
1 (.2)
0
(1.8) (.5) (.2) (.5)
(1.2) (.8) (.4) (.4)
follow-up in this group, none of them directly related to the LRYGB (3 patients who committed suicide, 1 patient with multimetastatic carcinoma, and 1 with myocardial infarction). Fig. 1 shows the number of eligible patients at each time interval and the percentage of follow-up, which was 88% after 10 years. Twenty-two patients (12.4%) developed long-term complications, of whom 17 (9.6%) required reoperation, including 1 complete reversal for chronic abdominal pain of unknown origin and 3 requiring conversion to distal gastric bypass for insufficient weight loss or massive weight regain. The mean BMI decreased initially and remained relatively stable after the first year (Fig. 2), and around 75% of the patients maintained a BMI o35 kg/m2 throughout the study period (Fig. 3). Fig. 4 shows the percentage of EBMIL compared with prebanding weight and Fig. 5 the percentage of TBWL, taking into account the maximal weight of the patients, either at LAGB or at LRYGB. Whatever the reference value, the mean EBMIL remained between 65% and 70% up to 10 years after conversion to RYGB, and the mean TBWL remained between 28% and 30%.
1
Discussion toward fewer major complications in the most recent half (2.5% versus 4.7%, P ¼ .2, Fisher’s exact test). There was no 30-day mortality. The mean duration of hospital stay was 5.9 ⫾ 4.1 days (range 3–29). As previously mentioned, long-term results are provided for patients from the Swiss cohort only (177 patients). At the time of LRYGB, 135 of these patients had maintained some weight loss compared with the pre-LAGB weight (from –1 to –83 kg), 4 had the same weight as before LAGB, and 38 had gained between 1 and 46 kg compared with their pre-LAGB weight. Five deaths occurred during
This study has 3 major findings. First, conversion to LRYGB after failed LAGB can be done by a laparoscopic approach with a positive safety profile. Second, in the absence of band erosion, a 1-step approach does not increase perioperative morbidity compared with a 2-step approach (band removal first, followed some time later by LRYGB). Third, results after conversion to LRYGB are very satisfactory up to 10 years after redo surgery, with a mean EBMIL between 65% and 70% in relation to the preband weight, corresponding to roughly 75% of the patients maintaining a BMI o35 kg/m2.
Fig. 1. Number of patients eligible for follow-up in the Swiss cohort (numbers below columns) and follow-up rates over time. LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00
5
Fig. 2. Evolution of the mean body mass index (⫾ SD) over time in the Swiss patient cohort. BMI ¼ body mass index; LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
LAGB is a relatively simple, adjustable, and reversible low-risk laparoscopic procedure that preserves the digestive tract. After promising early results, LAGB progressively lost its popularity in many European countries and in North America, due both to poor long-term outcomes in terms of weight loss and high long-term complication and reoperation rates. Common indications for reoperation after failed LAGB include insufficient weight loss or weight regain, complications associated with the implants (erosion, slippage, port and tubing problems), or anatomic or functional problems such as pouch dilation, progressive late food intolerance, intractable gastroesophageal reflux disease, and esophageal dilation and/or dysmotility. Several recent studies reported long-term reoperations in 31%–80% of patients after LAGB [5–9,13]. Reoperation for failed LAGB can involve simple revision (i.e., band change or repositioning) or conversion to another bariatric operation. Although some authors suggest that simple revision can be effective in patients who had
satisfactory weight loss after LAGB before they developed a complication [12,23,38], others support systematic conversion to another procedure, especially in patients with primary inadequate weight loss [12,24,25]. Simple revision further exposes the patient to all potential long-term complications of LAGB, with their inherent risks of further need for redo surgery. Revisional bariatric surgery is technically demanding and is usually associated with higher intraoperative and perioperative risks than primary procedures, notably of gastrointestinal complications [39–41]. According to a systematic review by the American Society for Metabolic and Bariatric Surgery Revision Task Force, these procedures should only be done by experienced bariatric surgeons in centers with sufficient resources [39]. To the best of the authors’ knowledge, the present study is the largest reporting on revisional LRYGB after failed LAGB. It indicates that conversion to LRYGB after failed LAGB can be performed with acceptable perioperative
Fig. 3. Distribution of body mass index over time. BMI ¼ body mass index; LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
6
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
Fig. 4. Percentage of excess body mass index lost before gastric bypass and over time in reference to the initial body mass index (before gastric banding). EBMIL ¼ excess body mass index lost; LAGB ¼ laparoscopic adjustable gastric banding; RYGB ¼ Roux-en-Y gastric bypass.
morbidity and no mortality when performed by experienced surgeons. The overall complication rate of 9.6% lies in the lower range of that reported among studies including at least 50 patients undergoing conversion from band to bypass (7%–46.3%), and the major morbidity of 3.6% is within usually reported rates (1.6%–13.4%) [26–36,40,42,43]. In fact, the complication rate reported herein is similar to that of many studies of primary RYGB [44,45], including the authors’ own experience. As stated by others [36], experience plays a role in postoperative morbidity. This is confirmed in the present study, in which morbidity was much lower among the second, later, half of patients. Leak at the gastrojejunostomy or on the gastric pouch is a complication intuitively expected to be more common in revisional gastric bypass than in primary cases, because stapling on the stomach is done in fibrous or inflammatory tissue. The leak rate of 1.7% lies within the range reported for primary gastric bypass in many large series [44]. Attention to technical details is important to prevent leaks. Whenever possible, stapling should be done away from band-related scar tissue. If necessary, the fibrous capsule formed around the band must be removed before stapling.
Finally, staple height must be adjusted to tissue thickness, and larger staples (green or black loads) are routinely used for the vertical portion of the gastric pouch in revisional cases. Staple-line reinforcement material may also be of help, but was used rarely in the present study. Whether conversion from failed LAGB to LRYGB should be performed in 1 or 2 steps remains controversial. A 1-step approach has clear advantages, such as decreased overall operative room, anesthesia, and hospital time; decreased patient waiting time; and better allocation of resources. It may also be beneficial if there is concern for reimbursement. Finally, in a 1-stage approach, the band can be used as a guide for dissection and recognition of the anatomy. In contrast, proponents of a 2-step approach argue that the first stage (band removal) is relatively easy and straightforward with low morbidity. Time between the first and second steps is usually associated with weight regain, but can allow for local inflammation to subside, potentially leading to reduced morbidity. In the present study, where 60% of the conversions were done in 1 step, the complication rate was not increased with this approach and was similar to the 9% overall and 4% major complication rates
Fig. 5. Evolution of the percentage of total weight loss ⫾ SD in reference to the maximum weight achieved by the patients either before gastric banding or at gastric bypass. %TWL ¼ total weight loss; RYGB ¼ Roux-en-Y gastric bypass.
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00
recently reported by Aarts et al. with the same 1-step approach [26]. In their study, Tran et al. also failed to show any difference in this respect [46], but in another small study, Van Nieuwenhove et al. found a trend toward fewer complications with a 2-stage approach, at the expense of increased operation time and hospital stay [47]. The currently available literature and the present results suggest that most patients can be converted in 1 step, except those with band erosion, infection, or very active inflammation at the time of band removal. If a 2-step approach is chosen, the authors strongly advise not only removal of the band during the first step, but also complete removal of the gastrogastric sutures and opening of the tunnel because it may be very difficult to recognize these structures later on. Results in terms of weight loss after conversion from LAGB to RYGB are poorly reported, because duration of follow-up is limited to 2–3 years or even less in most studies. Furthermore, many studies show results based on weight before redo surgery and not initial weight at the time of LAGB. Because the mean weight at redo is generally inferior to that before primary surgery, results are difficult to interpret, and a meaningful comparison with results of primary RYGB is hard to establish. In a small group of patients, Perathoner et al. reported a mean BMI of 28.8 kg/m2 5 years after conversion in a small group of patients [20]. Aarts et al. found a mean EWL of 53% compared with prebanding weight 5 years after conversion in a group of 54 patients [26]. In the present study, with a very high followup rate, a mean EBMIL (related to the pre-LAGB BMI) of 65%–70% up to 10 years after conversion is shown, corresponding to a mean BMI of 32–33 kg/m2 as of the first year after revisional LRYGB. This corresponds to 75%–80% of the patients maintaining a BMI o35 for up to 10 years, but also means that a minority of patients fails to achieve satisfactory weight despite 2 bariatric procedures. It is worth noting, however, that 6 of the 9 patients whose BMI was 435 kg/m2 10 years after redo surgery were in the super-obese category at the time of GB and that these 9 patients had an average of 9.3 (3.6–20.4) BMI units lost at 10 years, compared with the initial figure. In the authors’ experience, these results are similar to those obtained after primary LRYGB (unpublished data). Despite the multicenter setting and the large size of the cohort, the present study has some limitations. First, in patients who presented with the band still in place, the decision to proceed in 1 or 2 steps was biased by the surgeon’s discretion. Therefore, caution should be applied when interpreting comparisons between these 2 approaches. Second, only a fraction of the patients who underwent surgery were included in the long-term analysis regarding weight loss. Although the latter group had an excellent follow-up rate of 88% at 10 years, it may not represent the entire study population, especially with respect to the fact that the average pre-LRYGB weight in the French group was significantly higher than in the Swiss cohort. Third,
7
results regarding the effect of weight loss on co-morbidities could not be analyzed because of lack of data, especially before LAGB. On the other hand, strengths of the present study are the large sample size, which allows for some subgroup comparisons; the long duration of follow-up; and the high follow-up rate. Conclusions Laparoscopic conversion of a failed gastric band into RYGB is a relatively difficult but safe procedure. It is associated with morbidity similar to that of primary LRYGB and can be done without any mortality. In experienced hands, a 1-step approach is safe in most cases and does not increase the complication rate. Up to 10 years after redo surgery, results in terms of weight loss are comparable to those observed after primary LRYGB. Further studies, ideally randomized, will be necessary to draw definitive conclusions about the safety of a routine 1-step approach. More studies are also needed to assess long-term results, especially about resolution or improvement of co-morbidities. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] WHO Fact sheet No 311. Obesity and overweight [homepage on the Internet]. World Health Organization [cited 2014 Sep 20]. Available from: http://www.who.int/mediacentre/factsheets/fs311/en/. [2] Schauer PR, Bhatt DL, Kashyap SR. Bariatric surgery versus intensive medical therapy for diabetes. N Engl J Med 2014;371 (7):682. [3] Colquitt JL, Picot J, Loveman E, Clegg AJ. Surgery for obesity. Cochrane Database Syst Rev 2009;2. CD003641. [4] Belachew M, Legrand MJ, Defechereux TH, Burtheret MP, Jacquet N. Laparoscopic adjustable silicone gastric banding in the treatment of morbid obesity. A preliminary report. Surg Endosc 1994;8 (11):1354–6. [5] O’Brien PE, MacDonald L, Anderson M, Brennan L, Brown WA. Long-term outcome after bariatric surgery: fifteen-year follow-up of adjustable gastric banding and a systematic review of the surgical bariatric literature. Ann Surg 2013;257(1):87–94. [6] Gero D, Dayer-Jankechova A, Worreth M, Giusti V, Suter M. Laparoscopic gastric banding outcomes do not depend on device or technique. Long-term results of a prospective randomized study comparing the Lapbands and the SAGBs. Obes Surg 2014;24 (1):114–22. [7] Aarts EO, Dogan K, Koehestanie P, Anfenacker TJ, Janssen IMC, Berends FJ. Long-term results after laparoscopic adjustable gastric banding: a mean fourteen year follow-up study. Surg Obes Relat Dis 2014;10(4):633–40. [8] Naef M, Mouton WG, Naef U, Kummer O, Muggli B, Wagner HE. Graft survival and complications after laparoscopic gastric banding for morbid obesity—lessons learned from a 12-year experience. Obes Surg 2010;20(9):1206–14.
8
P. Fournier et al. / Surgery for Obesity and Related Diseases ] (2015) 00–00
[9] Himpens J, Cadière GB, Bazi M, Vouche M, Cadière B, Dapri G. Long-term outcomes of laparoscopic adjustable gastric banding. Arch Surg 2011;146(7):802–7. [10] Van Gemert WG, Van Wersch MM, Greve JWM, Soeters PB. Revisional surgery after failed vertical banded gastroplasty: restoration of vertical banded gastroplasty or conversion to gastric bypass. Obes Surg 2008;8(1):21–8. [11] Behrns KE, Smith CD, Kelly KA, Sarr MG. Reoperative bariatric surgery. Lessons learned to improved patient selection and results. Ann Surg 1993;218(6):646–53. [12] Ardestani A, Lautz DB, Tavakkolizadeh A. Band revision versus Roux-en-Y gastric bypass conversion as salvage operation after laparoscopic adjustable gastric banding. Surg Obes Relat Dis 2011;7 (1):33–7. [13] 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(4):391–8. [14] Park JY, Song D, Kim YD. Causes and outcomes of revisional bariatric surgery: initial experience at a single center. Ann Surg Treat Res 2014;86(6):295–301. [15] Kruger RS, Pricolo VE, Streeter TT, Colacchio DA, Andrade UA. A bariatric surgery center of excellence: operative trends and longterm outcomes. J Am Coll Surg 2014;218(6):1163–74. [16] Stroh C, Weiner R, Wolff S, Knoll C, Manger T. Arbeitsgruppe Adipositaschirurgie, Kompetenznetz Adipositas. Revisional surgery and reoperations in obesity and metabolic surgery: data analysis of the German bariatric surgery registry 2005–2012 [in German]. Chirurg 2015;86(4):346–54. [17] Fernando Santos B, Wallaert JB, Trus TL. Band removal and conversion to sleeve or bypass: are they equally safe? Surg Endosc 2014;28(11):3086–91 [18] Coblijn UK, Verveld CJ, van Wagensveld BA, Lagarde SM. Laparoscopic Roux-en-Y gastric bypass or laparoscopic sleeve gastrectomy as revisional procedure after adjustable gastric band—a systematic review. Obes Surg 2013;23(11):1899–914. [19] Van Nieuwenhove Y, Ceelen W, Van Renterghem K, Van de Putte D, Henckens T, Pattyn P. Conversion from band to bypass in two steps reduces the risk for anastomotic strictures. Obes Surg 2011;21 (4):501–5. [20] Perathoner A, Zitt M, Lanthaler M, Pratschke J, Biebl M, Mittermair R. Long-term follow-up evaluation of revisional gastric bypass after failed adjustable gastric banding. Surg Endosc 2013;27(11):4305–12. [21] Marin-Perez P, Betancourt A, Lamota M, Lo Menzo E, Szomstein S, Rosenthal R. Outcomes after laparoscopic conversion of failed adjustable gastric banding to sleeve gastrectomy or Roux-en-Y gastric bypass. Br J Surg 2014;101(2):254–6. [22] Elnahas A, Graybiel K, Farrokhyar F, Gmora S, Anvari M, Hong D. Revisional surgery after failed laparoscopic adjustable gastric banding: a systematic review. Surg Endosc 2013;27(3):740–5. [23] Bueter M, Thalheimer A, Wierlemann A, Fein M. Reoperations after gastric banding: replacement or alternative procedures. Surg Endosc 2009;23(2):334–40. [24] Lanthaler M, Mittermair R, Erne B, Weiss H, Aigner F, Nehoda H. Laparoscopic gastric re-banding versus laparoscopic gastric bypass as a rescue operation for patients with pouch dilatation. Obes Surg 2006;16(4):484–7. [25] Weber M, Muller MK, Michel JM, et al. Laparoscopic Roux-en-Y gastric bypass, but not rebanding, should be proposed as rescue procedure for patients with failed laparoscopic gastric banding. Ann Surg 2003;238(6):827–34. [26] Aarts E, Koehestanie P, Dogan K, Berends F, Janssen I. Revisional surgery after failed gastric banding: results of one-stage conversion to RYGB in 195 patients. Surg Obes Relat Dis 2014;10(6):1077–84.
[27] Deylgat B, D’Hondt M, Pottel H, Vansteenkiste F, Van Rooy F, Devriendt D. Indications, safety, and feasibility of conversion of failed bariatric surgery to Roux-en-Y gastric bypass: a retrospective comparative study with primary laparoscopic Roux-en-Y gastric bypass. Surg Endosc 2012;26(7):1997–2002. [28] Jennings NA, Boyle M, Mahawar K, Balupuri S, Small PK. Revisional laparoscopic Roux-en-Y gastric bypass following failed adjustable gastric banding. Obes Surg 2013;23(7):947–52. [29] Robert M, Poncet G, Boulez J, Mion F, Espalieu P. Laparoscopic gastric bypass for failure of adjustable gastric banding: a review of 85 cases. Obes Surg 2011;21(10):1513–9. [30] Slegtenhorst BR, Van der Harst E, Demirkiran A, de Korte J, Schelfhout LJ, Klaassen RA. Effect of primary versus revisional Roux-en-Y gastric bypass: inferior weight loss of revisional surgery after gastric banding. Surg Obes Relat Dis 2013;9(2):253–9. [31] te Riele W, Sze YK, Wiezer MJ, van Ramhorst B. Conversion of failed laparoscopic gastric banding to gastric bypass as safe and effective as primary gastric bypass in morbidly obese patients. Surg Obes Relat Dis 2008;4(6):735–9. [32] Thereaux J, Veyrie N, Barsamian C, et al. Similar postoperative safety between primary and revisional gastric bypass for failed gastric banding. JAMA Surg 2014;149(8):780–6. [33] Cadière GB, Himpens J, Bazi M, et al. Are laparoscopic gastric bypass after gastroplasty and primary gastric bypass similar in terms of results? Obes Surg 2011;21(6):692–8. [34] 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(10):1349–53. [35] Hii MW, Lake AC, Kenfield C, Hopkins GH. Laparoscopic conversion of failed gastric banding to Roux-en-Y gastric bypass. Shortterm follow-up and technical considerations. Obes Surg 2012;22 (7):1022–8. [36] Alpers JA, Wens C, Van Vlodrop V, et al. Perioperative outcomes of revisional laparoscopic gastric bypass after failed adjustable gastric banding and after vertical banded gastroplasty: experience with 107 cases and subgroup analysis. Surg Endosc 2013;27(2):558–64. [37] Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–13. [38] Shouten R, Japink D, Meesters B, Nelemans PJ, Greve JWM. Systematic literature review of reoperations after gastric banding: is a stepwise approach justified? Surg Obes Relat Dis 2011;7(1):99–109. [39] Brethauer SA, Kothari S, Sudan R, et al. Systematic review on reoperative bariatric surgery: American Society for Metabolic and Bariatric Surgery Revision Task Force. Surg Obes Relat Dis 2014;10(5):952–72. [40] Worni M, Østbye T, Shah A, et al. High risks for adverse outcomes after gastric bypass surgery following failed gastric banding—a population-based trend analysis of the United States. Ann Surg 2013;257(2):279–86. [41] Stroh C, Benedix D, Weiner R, et al. Is a one-step sleeve gastrectomy indicated as a revision procedure after gastric banding? Data analysis from a quality assurance study of the surgical treatment of obesity in Germany. Obes Surg 2014;24(1):9–14. [42] Carr WR, Jennings NA, Boyle M, Mahawar K, Balupuri S, Small PK. A retrospective comparison of early results of conversion of failed gastric banding to sleeve gastrectomy or gastric bypass. Surg Obes Relat Dis. Epub 2014 Aug 26. [43] Carandina S, Maldonado PS, Tabbara M, et al. Two-step conversion surgery after failed laparoscopic adjustable gastric banding. Comparison between laparoscopic Roux-en-Y gastric bypass and laparoscopic gastric sleeve. Surg Obes Relat Dis 2014;10(6):1085–9. [44] Zellmer JD, Mathiason MA, Kallies KJ, Kothari SN. Is laparoscopic sleeve gastrectomy a lower risk bariatric procedure compared with
Roux-en-Y Gastric Bypass for Failed Gastric Banding / Surgery for Obesity and Related Diseases ] (2015) 00–00 laparoscopic Roux-en-Y gastric bypass? A meta-analysis. Am J Surg 2014;208(6):903–10. [45] Carlin AM, Zeni TM, English WJ, et al. The comparative effectiveness of sleeve gastrectomy, gastric bypass, and adjustable gastric banding procedures for the treatment of morbid obesity. Ann Surg 2013;257(5):791–7.
9
[46] Tran TT, Pauli E, Lyn-Sue JR, Haluck R, Rogers AM. Revisional weight loss surgery after failed laparoscopic gastric banding: an institutional experience. Surg Endosc 2013;27(11):4087–93. [47] Van Nieuwenhove Y, Ceelen W, Van Renterghem K, Van de Putte D, Henckens T, Pattyn P. Conversion from band to bypass in two steps reduces the risk for anastomotic strictures. Obes Surg 2011;21(4):501–5.
