Surgery for Obesity and Related Diseases 10 (2014) 106–111
Original article
The early use of Roux limb as surgical treatment for proximal postsleeve gastrectomy leaks Mohomad Chour, M.D.a,*, Ramzi S. Alami, M.D.b, Fadi Sleilaty, M.D.a, Raja Wakim, M.D.a a
Department of Surgery, Mount Lebanon Hospital, Hazmieh, Lebanon b American University of Beirut Medical Center, Beirut, Lebanon Received February 27, 2013; accepted August 19, 2013
Abstract
Background: Laparoscopic sleeve gastrectomy (LSG) is thought to be a simpler and safer operation compared with malabsorptive operations that include an enteric anastomosis. Leakage along the staple line at the gastroesophageal junction (GEJ) is difficult to treat and is a known complication of sleeve gastrectomy. Nonsurgical treatment methods often fail to heal the leaks and patients often require conversion to other procedures for definitive treatment. We report our experience with conversion to Roux-en-Y anastomosis over the leak site as a treatment option, comparing patients who had early treatment to late intervention. The purpose of the study is to stress the medical and social benefits of early surgical reintervention with conversion to Roux-en-Y anastomosis over the leak site. Methods: Six patients underwent Roux limb placement over the leak site. Four of the patients had delayed surgery (group A), and the other 2 had early intervention (group B). Results: Patients in group A had a median increase of all medical cost by 500%, whereas the 2 patients who underwent early intervention (group B) had an increase by 200%. The mean time until complete recovery (removal of all drains, adequate oral intake, and return to normal daily activity) in group A was 131.25 days (range 99–165) versus 38 days (range 28–48) in group B. Conclusions: Roux-en-Y gastrojejunostomy over the leak site is an effective technique to treat refractory staple line leakage and can be adopted as early treatment in selected patients after stabilization, thereby reducing the cost and length of hospital stays. (Surg Obes Relat Dis 2014;10:106–111.) r 2014 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Laparoscopic sleeve gastrectomy; gastroesophageal junction leak; Roux limb anastomosis; Roux-en-Y gastrojejunostomy
Laparoscopic sleeve gastrectomy (LSG) is thought to be a simpler and safer operation compared with malabsorptive operations that include an enteric anastomosis. Leakage along the staple line at the gastroesophageal junction (GEJ) is difficult to treat and is a known complication of sleeve gastrectomy [1], with a reported incidence ranging between .7% to 5% (mean 2.3%) [2]. Leaks can be classified as acute, early, late, and chronic, according to the time of diagnosis (within 7 days, within 1–6 weeks, after 6 weeks, and after 12 weeks, respectively) [3]. * Correspondence: Mohomad Chour, M.D., Mount Lebanon Hospital, Blvd. Camille Chamoun, P.O.Box 470, Hazmieh, Lebanon. E-mail: [email protected]
The ideal treatment for leaks is yet to be established. Some nonsurgical treatment options are available and include the use of different products, such as surgical glues under endoscopy [4,5] and fibrin sealant (Tissucol, Baxter Healthcare, Deerfield, IL), as well as endoscopic clip application or stent placement [4,6]. However, there is no consensus or agreement about the efficacy of these options. In recent years, many authors have supported the use of flexible coated stents as a second step [4,7,8]. Tolerance to stents is variable. Nausea, vomiting, drooling, early satiety, and retrosternal discomfort are the most common symptoms after their placement [7]. Most authors recommend a period of 6 to 8 weeks as the optimal time to withdraw the stent [9]. This procedure is not easy. Adhesions due to the stent
1550-7289/14/$ – see front matter r 2014 American Society for Metabolic and Bariatric Surgery. All rights reserved. http://dx.doi.org/10.1016/j.soard.2013.08.017
Postsleeve Gastrectomy Leak / Surgery for Obesity and Related Diseases 10 (2014) 106–111
are frequent, and mucosal tears can occur [10]. Stent migration is one of the main concerns when these are used, because this occurs in up to one third of patients [7,9] and often requires replacement or removal of the stent. Patients who do not respond to any of these procedures or those with a persistent leak are candidates for surgical intervention, such as conversion to Roux-en-Y gastric bypass (RYGB), gastric resection with Roux-en-Y esophago-jejunostomy, Roux-en-Y esophago-jejunostomy without gastric resection, and Roux-en-Y gastrojejunostomy anastomosis over the leak area. The use of a Roux-en-Y gastrojejunostomy anastomosed over the site of a leak after sleeve gastrectomy was first described in the literature by Baltasar et al. [11]. The same group published a follow-up paper describing their experience and outcomes for 3 patients treated with this same technique [12]. The study reported here describes an experience with conversion to Roux-en-Y anastomosis over the leak site as a treatment option and compares patients who had early treatment to patients who had late intervention. Materials and methods Study population From March 2007 to December 2012, 524 patients underwent LSG in our surgical unit. Retrospective analysis of the database was carried out by querying all the LSG cases managed in this period; 6 cases of proximal postsleeve gastrectomy leaks (3 were referred from other institutions and 3 were originally operated in our unit) formed the study group. The sleeve gastrectomy procedure Four to 5 trocars were placed. The division of the vascular supply of the gastric greater curvature was carried out with LigaSure Vessel Sealing device (Valleylab, Boulder, CO) or ultrasonic shears and started at 6–8 cm from the pylorus, proceeding upward to the angle of His. The gastric pouch was created by using a linear stapler (EndoGIA, U.S. Surgical, Norwalk, CT), with 5–6 sequential 4.8/60-mm loads. The stapler was applied alongside a 36-Fr calibrating bougie tightly positioned against the lesser curve to obtain an 80–120-mL gastric pouch. No oversewing or buttress materials were used to reinforce the staple line. The resected stomach was extracted through the 15-mm port-site incision. A methylene blue dye test was routinely performed through a nasogastric tube. A corrugated rubber drain was routinely placed at the side of the staple line, and the nasogastric tube was removed at the end of the procedure. Unless there was a high clinical suspicion, no routine tests were performed postoperatively looking for a leak.
107
The Roux-en-Y gastrojejunostomy over the leak site This operation was performed by 2 surgeons who had at least 10 years of experience with bariatric and advanced upper and lower gastrointestinal surgeries. A supraumbilical midline laparotomy and lysis of adhesions were performed until identification of the left lateral segment of the liver, the left and the right crura. This was followed by dissection of the whole gastric tube and the distal esophagus, and finally, the leak site was identified using methylene blue dye test. The edges of the gastric opening were refreshed and widened (1–2 cm) and an antecolic Roux limb of 70 cm in length was brought up. A side-to-side gastrojejunostomy was then fashioned, using single layered, full thickness, interrupted, 3-0 Vicryl suture. A methylene blue dye test was performed through a nasogastric tube to check for anastomotic integrity after clamping the distal stomach. A corrugated rubber drain was routinely placed alongside the anastomosis.
Results During the study period, 524 patients underwent LSG for the treatment of morbid obesity. Mean body mass index (BMI) was 41.2 kg/m2; 332 (63.3%) patients were female. Mean patient age was 32.5 years (range 16–69 yr). In 21 patients (4%), LSG was revisional surgery for insufficient weight loss: after adjustable gastric banding in 15 cases and after vertical banded gastroplasty (VBG) in 6 cases. Staple line leaks occurred in 3 patients (.57%). Three patients were referred to our institution for management of their leaks. All patients with leak were female, mean BMI was 40.5 ⫾ 3.5 kg/m2, and mean age was 33.6 years (range 22–47 yr). Two of them had undergone previous bariatric procedures (adjustable gastric band in one and VBG in the other). The patient characteristics are shown in Table 1. In our 3 cases, no intraoperative complications had occurred, and all patients experienced failure of the methylene blue test to detect a leak. Two patients experienced early postoperative complications. One experienced staple line bleeding that required transfusion of 4 units of packed red blood cells for stabilization. The other had intestinal obstruction that resolved with nonsurgical management. Leaks were diagnosed at a mean of 10.3 days (range, 2– 14 days) postoperatively: 1 (16.6 %) acute (within 7 days) and 5 (83.4 %) early (within 1 to 6 weeks). In all cases, the leakage developed at the GEJ (proximal leak). Three leaks were diagnosed by UGI series, and the remainder were definitely diagnosed by CT (computed tomography) scan. Four patients had delayed Roux limb anastomosis over the leak site (group A), and the other 2 had the Roux limb within 3–4 weeks (group B). Within group A, 3 patients underwent CT-guided drainage followed by surgical drainage, and only 1 patient underwent surgical drainage (the patient with a leak
108
M. Chour et al. / Surgery for Obesity and Related Diseases 10 (2014) 106–111
Table 1 Summary of cases (time from the original laparoscopic sleeve gastrectomy) Gender/Age
BMI kg/m2
Notes
Fistula diagnosed
Drainage
Endoscopic treatment
Roux-en-Y done
Case 1
F/23
46
10 days
F/44
40
3 weeks: Clips 3 months: Polyflex 3 months: Polyflex 2
Case 3 (ref) Case 4 (ref) Case 5 (ref) Case 6
F/22
33
–
2 days
10 days, radiologic 2 weeks surgical 2 weeks, radiologic 3 month surgical 2 days surgical
4 months
Case 2
Transfusion of 4 units PRBCs postsleeve LGB 10 years ago
F/35
40
–
12 days
F/47
40
F/31
42
VBG 6 years ago, -postsleeve intestinal obstruction –
2 weeks
5 months
3 months
2 weeks
12 days, radiologic 18 days surgical 2 weeks surgical
2 months: Ultraflex then Polyflex –
4 months
–
4 weeks
10 days
12 days radiologic
–
3 weeks
BMI ¼ body mass index; F ¼ female; LGB ¼ laparoscopic gastric band; PRBCs ¼ packed red blood cells; ref ¼ referred case; VBG ¼ vertical banded gastroplasty.
diagnosed within 2 days). Within group B, 1 patient underwent CT guided drainage only, and 1 patient underwent surgical drainage only (Table 1). Surgical drainage consisted of laparoscopic exploration and a thorough washout and drainage of the abdomen. Leaksite suture was attempted in 1 case (the case with the acute leak) and feeding jejunostomy was placed in 2 patients. Three patients of group A had an endoscopic treatment, whereas the last patient and the 2 patients of group B did not have endoscopic treatment. Endoscopic treatment consisted of endogastric clipping of the leak site in 1 patient 3 weeks after leak diagnosis, followed by insertion of a selfexpanding stent (Polyflex, Boston Scientific, Natick, MA) 3 months after leak diagnosis. The second patient had a Polyflex stent inserted at 3 months of diagnosis; then at 4
months of diagnosis and due to migration, it was replaced by another Polyflex stent. The third patient had a coated metallic Ultraflex Esophageal NG Stent System inserted at 2 months of leak diagnosis, but it failed to treat the leak; to decrease mucosal adhesions, a Polyflex stent was inserted on top of the Ultraflex stent and then both stents were removed after 1 month (Table 1). Self-expanding stents were placed under fluoroscopic and endoscopic guidance. Plain x-rays were performed weekly to check the correct stent position. The procedures were all performed by an experienced gastroenterologist. The 3 patients had stent migration at some point in time, and the stents were removed. All 3 stented patients had persistence of the leak after 4, 5, and 4 months, respectively, documented by continuous drain output (450 cc/d) (Fig. 1).
Fig. 1. Gastric leak. Gastrografin swallow showed persistent leak (thick arrow) even after stent placement (light arrow).
Postsleeve Gastrectomy Leak / Surgery for Obesity and Related Diseases 10 (2014) 106–111
109
Table 2 Summary of the postoperative recovery period
Case Case Case Case Case Case
1 2 3 4 5 6
Time of Roux limb anastomosis from original LSG
Discharge day after Roux limb anastomosis
Days until complete recovery from original LSG*
Follow-up month
% excess BMI loss
UGI series flow
4 5 4 3 4 3
12 17 11 8 20 7
129 165 132 99 48 28
44 35 19 13 32 22
61.5% 45% 53.8% 80% 45% 68.1%
SþL SþL L SþL S S
months months months months weeks weeks
BMI ¼ body mass index; L ¼ small bowel limb; LSG ¼ laparoscopic sleeve gastrectomy; S ¼ stomach; UGI ¼ upper gastrointestinal. Complete recovery indicates return to normal activity without drains.
*
The Roux-en-Y gastrojejunostomy was performed in all 6 patients as described. The mean operative time during the Roux limb anastomosis was 260 minutes in group A versus 180 minutes in group B. No intraoperative incident was noted, no transfusion was necessary, and no mortality was encountered. Three patients developed a small leak at the anastomosis, which healed spontaneously after a mean time of 11 days. The length of hospital stay after the Roux limb anastomosis was 12.75 days (range 8–17 d) for group A versus 13.5 days for group B (7–20 d). The mean time until complete recovery (removal of all drains and return to normal daily activity) in group A was 131.25 days (range 99–165 d) and 38 days (28–48 d) for group B. The 4 patients in whom the Roux limb was delayed (group A) had a median increase of all medical cost by 500%, whereas the 2 patients who underwent early Roux limb (group B) had an increase by 200% (Table 2). A swallow study was performed approximately 3 months after surgery showing that both the sleeve and the Roux-enY limb were functioning in 3 patients, only the sleeve was functioning in 2 patients, and the Roux limb was the only functioning pathway in 1 patient. At a mean follow-up of 27.5 months (range from 13–44 mo), all 6 patients were tolerating the procedure well. Mean percentage of excess weight loss was 58.8%. Three patients had vitamin B12 deficiency, and 2 patients had iron deficiency. Discussion Leaks after LSG are reported to occur in between .7% to 5% (mean 2.3%) of cases [2]. Clinically, they may range from mild micro leaks that present from weeks to months after surgery as the cause of perisleeve abscesses and chronic fistula to an abdominal catastrophe with sepsis, hemodynamic instability, multisystem organ failure, and rarely, patient demise [2,4,13]. Management options vary and depend on the timing and clinical presentation of the leak. Immediate reoperation is the preferred course of action for the unstable patient, usually with washout, irrigation of the abdominal cavity, wide drainage, and an attempt at suturing of the leak if the tissue condition allows it [2]. Stable patients and leaks presenting
later in the postoperative course pose a yet unanswered dilemma regarding the best treatment algorithm. The methods used to date in attempts at leak closure have been percutaneous, endoscopic, or surgical [14]. According to the International Sleeve Gastrectomy Expert Panel (Consensus Statement 2011), the surgeon should wait until 12 weeks after conservative therapy to allow for healing and avoid thick adhesions during reoperation. The operation that most experts would perform is conversion to RYGB [3]. In our experience, all attempts for conservative management failed, and the 6 cases were eventually treated surgically. We prefer the Roux limb gastrojejunal anastomosis as described by Baltasar et al. [11]. This approach should be attempted by experienced surgeons who can deal with any operative complications that could lead to change in the plan. In this study, the 2 surgeons had at least 10 years of experience with bariatric and advanced upper and lower gastrointestinal surgeries. All were open surgeries, to avoid an increase in operative time and co-morbidities, but the technique is feasible laparoscopically, and we may attempt that approach in the future. In a recent article, Van de Vrande et al. described successful laparoscopic Roux limb placement in 11 patients with chronic postLSG fistula [15]. The addition of a Roux limb aims at transforming a highpressure system, constituted by the sleeve construction, into a low-pressure system. A low-pressure system reportedly facilitates fistula healing in foregut surgery [16,17]. The concept is the same as the concept behind performing a RYGB. However, RYGB requires transection of the stomach, a step that can be very hazardous because of the tissue changes caused by the chronic inflammation. Compared with total gastrectomy or RYGB conversion, Roux limb anastomosis spares the patient the metabolic consequences and the decrease in quality of life that can be associated with a RYGB or a total gastrectomy [15]. The factors that may contribute to nonoperative management failure in our series are the inexperience of the gastroenterologist performing the procedures, the use of stents 2 months after the leak was diagnosed, and the fact that these prostheses, in principle, are designed for use in esophageal disease (stenosis). When used to treat leaks, the usual placement site is at the end of the esophagus and the
110
M. Chour et al. / Surgery for Obesity and Related Diseases 10 (2014) 106–111
proximal part of the gastric sleeve, which frequently leads to migration. Nonoperative management of a chronic leak after LSG requires a multimodal approach and a range of repetitive endoscopic therapies, which become very expensive, often with unrewarding results. There is currently no evidencebased algorithm for the treatment of leaks, including guidance for which modality to try first [18]. The mean operative time during the Roux limb anastomosis was 260 minutes in group A versus 180 minutes in group B, which could be explained by the greater fibrosis and adhesions that formed after stent placement. In our study, all conservative management failed, and the 6 patients required Roux limb surgery. When comparing the 2 groups, the 2 patients who underwent early Roux limb anastomosis (group B) had a decrease of all medical cost and a reduced time until complete recovery (removal of all drains, adequate oral intake, and return to normal daily activity; Table 2). According to our experience, especially in the last 2 cases, we now believe that early intervention is a good option for leaks that did not close after 4 weeks of conservative management. After a leak is diagnosed, adequate drainage is of paramount importance and a mandatory adjunct for any endoscopic interventions. This can be achieved with a drain placed during initial surgery or with additional imaging-guided drains. Percutaneous imaging-guided drainage and distal enteral or parenteral hyperalimentation together with systemic broad-spectrum antibiotics are the mainstay of the nonoperative management. The International Sleeve Gastrectomy Expert Panel stated that the use of a stent is a valid treatment option for an acute proximal leak (o7 d); after 30 days, the likelihood of a leak to seal by exclusion using a stent is very low [3]. We add that if the leak output does not decrease in 4 weeks, early Roux limb anastomosis should be considered. Roux-en-Y gastrojejunostomy over the leak site has many advantages. It is an effective technique to treat staple line leak, can be adopted as early treatment (in selected patients and after drainage), reduces the cost and length of hospital stay, and reduces the delay in leak healing, all of which should decrease the stress on the patient and the family. The limitations of the present study are the small number of cases and the need to be compared with postsleeve leak cases that were treated by other surgical interventions (conversion to RYGB or Roux-en-Y esophago-jejunostomy anastomosis). Longer follow-up is needed to determine whether any other secondary effects (gastroesophageal reflux disease, increasing gastric acidity, anastomotic ulcers, and metabolic perturbation) will occur. Conclusions The gastroesophageal junction is a potential site of leakage after sleeve gastrectomy that is usually difficult to treat conservatively. Roux-en-Y gastrojejunostomy is an effective
technique to treat refractory staple line leakage and can be adopted as early treatment in selected patients after stabilization, thereby reducing the cost and length of hospital stays. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] Dapri G, Vaz C, Cadière GB, Himpens J. A prospective randomized study comparing two different techniques for laparoscopic sleeve gastrectomy. Obes Surg 2007;17:1435–41. [2] Burgos A, Braghetto I, Csendes A, et al. Gastric leak after laparoscopicsleeve gastrectomy for obesity. Obes Surg 2009;19:1672–7. [3] Rosenthal RJ, International Sleeve Gastrectomy Expert Panel, Diaz AA, Arvidsson D, et al. International Sleeve Gastrectomy Expert Panel Consensus Statement: best practice guidelines based on experience of 412,000 cases. Surg Obes Relat Dis 2012;8:8–19. [4] Casella G, Soricelli E, Rizello M, et al. Nonsurgical treatment of staple line leaks after laparoscopic sleeve gastrectomy. Obes Surg 2009;19:821–6. [5] Papavramidis TS, Kotzampassi K, Kotidis E, Eleftheriadis EE, Papavramidis ST. Endoscopic fibrin sealing of gastrocutaneous fistulas after sleeve gastrectomy and biliopancreatic diversion with duodenal switch. J Gastroenterol Hepatol 2008;23:1802–5. [6] Akhras J, Tobi M, Zagnoon A. Endoscopic fibrin sealant injection with application of hemostatic clips: a novel method of closing a refractory gastrocutaneous fistula. Dig Dis Sci 2005;50:1872–4. [7] Serra C, Baltasar A, Andreo L, et al. Treatment of gastric leaks with coated self-expanding stents after sleeve gastrectomy. Obes Surg 2007;17:866–72. [8] Blackmon SH, Santora R, Schwarz P, Barroso A, Dunkin BJ. Utility of removable esophageal covered self-expanding metal stents for leak and fistula management. Ann Thorac Surg 2010;89:931–7. [9] Fukumoto R, Orlina J, McGinty J, Teixeira J. Use of polyflex stents in treatment of acute esophageal and gastric leaks after bariatric surgery. Surg Obes Relat Dis 2006;2:570–2. [10] Marquez MF, Ayza MF, Lozano RB, Morales Mdel M, Diez JM, Poujoulet RB. Gastric leak after laparoscopic sleeve gastrectomy. Obes Surg 2010;20:1306–11. [11] Baltasar A, Bou R, Bengochea M, Serra C, Cipagauta L. Use of a Roux Limb to correct esophagogastric junction fistulas after sleeve gastrectomy. Obes Surg 2007;17:1408–10. [12] Baltasar A, Serra C, Bengochea M, Bou R, Andreo L. Use of Roux limb as remedial surgery for sleeve gastrectomy fistulas. Surg Obes Relat Dis 2008;4:759–63. [13] Cottam D, Qureshi FG, Mattar SG, et al. Laparoscopic sleeve gastrectomy as an initial weight loss procedure for high risk patients with morbid obesity. Surg Endosc 2006;20:859–63. [14] Sakran N, Goitein D, Raziel A, et al. Gastric leaks after sleeve gastrectomy: a multicenter experience with 2,834 patients. Surg Endosc 2013;27:240–5. [15] Van de Vrande S, Himpens J, El Mourad H, Debaerdemaeker R, Leman G. Management of chronic proximal fistulas after sleeve gastrectomy by laparoscopic Roux-limb placement. Surg Obes Relat Dis. Epub 2013 Jan 16. [16] Gagner M. Leaks after sleeve gastrectomy are associated with smaller bougies: prevention and treatment strategies. Surg Laparosc Endosc Percutan Tech 2010;20:166–9. [17] Yehoshua RT, Eidelman LA, Stein M, et al. Laparoscopic sleeve gastrectomy—volume and pressure assessment. Obes Surg 2008;18:1083–8. [18] Beitner M, Cohen J, Kurian M. Endoscopic management of persistent leak after laparoscopic sleeve gastrectomy. Bariatric Times 2012;9:22–4.
Original article
The early use of Roux limb as surgical treatment for proximal postsleeve gastrectomy leaks Mohomad Chour, M.D.a,*, Ramzi S. Alami, M.D.b, Fadi Sleilaty, M.D.a, Raja Wakim, M.D.a a
Department of Surgery, Mount Lebanon Hospital, Hazmieh, Lebanon b American University of Beirut Medical Center, Beirut, Lebanon Received February 27, 2013; accepted August 19, 2013
Abstract
Background: Laparoscopic sleeve gastrectomy (LSG) is thought to be a simpler and safer operation compared with malabsorptive operations that include an enteric anastomosis. Leakage along the staple line at the gastroesophageal junction (GEJ) is difficult to treat and is a known complication of sleeve gastrectomy. Nonsurgical treatment methods often fail to heal the leaks and patients often require conversion to other procedures for definitive treatment. We report our experience with conversion to Roux-en-Y anastomosis over the leak site as a treatment option, comparing patients who had early treatment to late intervention. The purpose of the study is to stress the medical and social benefits of early surgical reintervention with conversion to Roux-en-Y anastomosis over the leak site. Methods: Six patients underwent Roux limb placement over the leak site. Four of the patients had delayed surgery (group A), and the other 2 had early intervention (group B). Results: Patients in group A had a median increase of all medical cost by 500%, whereas the 2 patients who underwent early intervention (group B) had an increase by 200%. The mean time until complete recovery (removal of all drains, adequate oral intake, and return to normal daily activity) in group A was 131.25 days (range 99–165) versus 38 days (range 28–48) in group B. Conclusions: Roux-en-Y gastrojejunostomy over the leak site is an effective technique to treat refractory staple line leakage and can be adopted as early treatment in selected patients after stabilization, thereby reducing the cost and length of hospital stays. (Surg Obes Relat Dis 2014;10:106–111.) r 2014 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Laparoscopic sleeve gastrectomy; gastroesophageal junction leak; Roux limb anastomosis; Roux-en-Y gastrojejunostomy
Laparoscopic sleeve gastrectomy (LSG) is thought to be a simpler and safer operation compared with malabsorptive operations that include an enteric anastomosis. Leakage along the staple line at the gastroesophageal junction (GEJ) is difficult to treat and is a known complication of sleeve gastrectomy [1], with a reported incidence ranging between .7% to 5% (mean 2.3%) [2]. Leaks can be classified as acute, early, late, and chronic, according to the time of diagnosis (within 7 days, within 1–6 weeks, after 6 weeks, and after 12 weeks, respectively) [3]. * Correspondence: Mohomad Chour, M.D., Mount Lebanon Hospital, Blvd. Camille Chamoun, P.O.Box 470, Hazmieh, Lebanon. E-mail: [email protected]
The ideal treatment for leaks is yet to be established. Some nonsurgical treatment options are available and include the use of different products, such as surgical glues under endoscopy [4,5] and fibrin sealant (Tissucol, Baxter Healthcare, Deerfield, IL), as well as endoscopic clip application or stent placement [4,6]. However, there is no consensus or agreement about the efficacy of these options. In recent years, many authors have supported the use of flexible coated stents as a second step [4,7,8]. Tolerance to stents is variable. Nausea, vomiting, drooling, early satiety, and retrosternal discomfort are the most common symptoms after their placement [7]. Most authors recommend a period of 6 to 8 weeks as the optimal time to withdraw the stent [9]. This procedure is not easy. Adhesions due to the stent
1550-7289/14/$ – see front matter r 2014 American Society for Metabolic and Bariatric Surgery. All rights reserved. http://dx.doi.org/10.1016/j.soard.2013.08.017
Postsleeve Gastrectomy Leak / Surgery for Obesity and Related Diseases 10 (2014) 106–111
are frequent, and mucosal tears can occur [10]. Stent migration is one of the main concerns when these are used, because this occurs in up to one third of patients [7,9] and often requires replacement or removal of the stent. Patients who do not respond to any of these procedures or those with a persistent leak are candidates for surgical intervention, such as conversion to Roux-en-Y gastric bypass (RYGB), gastric resection with Roux-en-Y esophago-jejunostomy, Roux-en-Y esophago-jejunostomy without gastric resection, and Roux-en-Y gastrojejunostomy anastomosis over the leak area. The use of a Roux-en-Y gastrojejunostomy anastomosed over the site of a leak after sleeve gastrectomy was first described in the literature by Baltasar et al. [11]. The same group published a follow-up paper describing their experience and outcomes for 3 patients treated with this same technique [12]. The study reported here describes an experience with conversion to Roux-en-Y anastomosis over the leak site as a treatment option and compares patients who had early treatment to patients who had late intervention. Materials and methods Study population From March 2007 to December 2012, 524 patients underwent LSG in our surgical unit. Retrospective analysis of the database was carried out by querying all the LSG cases managed in this period; 6 cases of proximal postsleeve gastrectomy leaks (3 were referred from other institutions and 3 were originally operated in our unit) formed the study group. The sleeve gastrectomy procedure Four to 5 trocars were placed. The division of the vascular supply of the gastric greater curvature was carried out with LigaSure Vessel Sealing device (Valleylab, Boulder, CO) or ultrasonic shears and started at 6–8 cm from the pylorus, proceeding upward to the angle of His. The gastric pouch was created by using a linear stapler (EndoGIA, U.S. Surgical, Norwalk, CT), with 5–6 sequential 4.8/60-mm loads. The stapler was applied alongside a 36-Fr calibrating bougie tightly positioned against the lesser curve to obtain an 80–120-mL gastric pouch. No oversewing or buttress materials were used to reinforce the staple line. The resected stomach was extracted through the 15-mm port-site incision. A methylene blue dye test was routinely performed through a nasogastric tube. A corrugated rubber drain was routinely placed at the side of the staple line, and the nasogastric tube was removed at the end of the procedure. Unless there was a high clinical suspicion, no routine tests were performed postoperatively looking for a leak.
107
The Roux-en-Y gastrojejunostomy over the leak site This operation was performed by 2 surgeons who had at least 10 years of experience with bariatric and advanced upper and lower gastrointestinal surgeries. A supraumbilical midline laparotomy and lysis of adhesions were performed until identification of the left lateral segment of the liver, the left and the right crura. This was followed by dissection of the whole gastric tube and the distal esophagus, and finally, the leak site was identified using methylene blue dye test. The edges of the gastric opening were refreshed and widened (1–2 cm) and an antecolic Roux limb of 70 cm in length was brought up. A side-to-side gastrojejunostomy was then fashioned, using single layered, full thickness, interrupted, 3-0 Vicryl suture. A methylene blue dye test was performed through a nasogastric tube to check for anastomotic integrity after clamping the distal stomach. A corrugated rubber drain was routinely placed alongside the anastomosis.
Results During the study period, 524 patients underwent LSG for the treatment of morbid obesity. Mean body mass index (BMI) was 41.2 kg/m2; 332 (63.3%) patients were female. Mean patient age was 32.5 years (range 16–69 yr). In 21 patients (4%), LSG was revisional surgery for insufficient weight loss: after adjustable gastric banding in 15 cases and after vertical banded gastroplasty (VBG) in 6 cases. Staple line leaks occurred in 3 patients (.57%). Three patients were referred to our institution for management of their leaks. All patients with leak were female, mean BMI was 40.5 ⫾ 3.5 kg/m2, and mean age was 33.6 years (range 22–47 yr). Two of them had undergone previous bariatric procedures (adjustable gastric band in one and VBG in the other). The patient characteristics are shown in Table 1. In our 3 cases, no intraoperative complications had occurred, and all patients experienced failure of the methylene blue test to detect a leak. Two patients experienced early postoperative complications. One experienced staple line bleeding that required transfusion of 4 units of packed red blood cells for stabilization. The other had intestinal obstruction that resolved with nonsurgical management. Leaks were diagnosed at a mean of 10.3 days (range, 2– 14 days) postoperatively: 1 (16.6 %) acute (within 7 days) and 5 (83.4 %) early (within 1 to 6 weeks). In all cases, the leakage developed at the GEJ (proximal leak). Three leaks were diagnosed by UGI series, and the remainder were definitely diagnosed by CT (computed tomography) scan. Four patients had delayed Roux limb anastomosis over the leak site (group A), and the other 2 had the Roux limb within 3–4 weeks (group B). Within group A, 3 patients underwent CT-guided drainage followed by surgical drainage, and only 1 patient underwent surgical drainage (the patient with a leak
108
M. Chour et al. / Surgery for Obesity and Related Diseases 10 (2014) 106–111
Table 1 Summary of cases (time from the original laparoscopic sleeve gastrectomy) Gender/Age
BMI kg/m2
Notes
Fistula diagnosed
Drainage
Endoscopic treatment
Roux-en-Y done
Case 1
F/23
46
10 days
F/44
40
3 weeks: Clips 3 months: Polyflex 3 months: Polyflex 2
Case 3 (ref) Case 4 (ref) Case 5 (ref) Case 6
F/22
33
–
2 days
10 days, radiologic 2 weeks surgical 2 weeks, radiologic 3 month surgical 2 days surgical
4 months
Case 2
Transfusion of 4 units PRBCs postsleeve LGB 10 years ago
F/35
40
–
12 days
F/47
40
F/31
42
VBG 6 years ago, -postsleeve intestinal obstruction –
2 weeks
5 months
3 months
2 weeks
12 days, radiologic 18 days surgical 2 weeks surgical
2 months: Ultraflex then Polyflex –
4 months
–
4 weeks
10 days
12 days radiologic
–
3 weeks
BMI ¼ body mass index; F ¼ female; LGB ¼ laparoscopic gastric band; PRBCs ¼ packed red blood cells; ref ¼ referred case; VBG ¼ vertical banded gastroplasty.
diagnosed within 2 days). Within group B, 1 patient underwent CT guided drainage only, and 1 patient underwent surgical drainage only (Table 1). Surgical drainage consisted of laparoscopic exploration and a thorough washout and drainage of the abdomen. Leaksite suture was attempted in 1 case (the case with the acute leak) and feeding jejunostomy was placed in 2 patients. Three patients of group A had an endoscopic treatment, whereas the last patient and the 2 patients of group B did not have endoscopic treatment. Endoscopic treatment consisted of endogastric clipping of the leak site in 1 patient 3 weeks after leak diagnosis, followed by insertion of a selfexpanding stent (Polyflex, Boston Scientific, Natick, MA) 3 months after leak diagnosis. The second patient had a Polyflex stent inserted at 3 months of diagnosis; then at 4
months of diagnosis and due to migration, it was replaced by another Polyflex stent. The third patient had a coated metallic Ultraflex Esophageal NG Stent System inserted at 2 months of leak diagnosis, but it failed to treat the leak; to decrease mucosal adhesions, a Polyflex stent was inserted on top of the Ultraflex stent and then both stents were removed after 1 month (Table 1). Self-expanding stents were placed under fluoroscopic and endoscopic guidance. Plain x-rays were performed weekly to check the correct stent position. The procedures were all performed by an experienced gastroenterologist. The 3 patients had stent migration at some point in time, and the stents were removed. All 3 stented patients had persistence of the leak after 4, 5, and 4 months, respectively, documented by continuous drain output (450 cc/d) (Fig. 1).
Fig. 1. Gastric leak. Gastrografin swallow showed persistent leak (thick arrow) even after stent placement (light arrow).
Postsleeve Gastrectomy Leak / Surgery for Obesity and Related Diseases 10 (2014) 106–111
109
Table 2 Summary of the postoperative recovery period
Case Case Case Case Case Case
1 2 3 4 5 6
Time of Roux limb anastomosis from original LSG
Discharge day after Roux limb anastomosis
Days until complete recovery from original LSG*
Follow-up month
% excess BMI loss
UGI series flow
4 5 4 3 4 3
12 17 11 8 20 7
129 165 132 99 48 28
44 35 19 13 32 22
61.5% 45% 53.8% 80% 45% 68.1%
SþL SþL L SþL S S
months months months months weeks weeks
BMI ¼ body mass index; L ¼ small bowel limb; LSG ¼ laparoscopic sleeve gastrectomy; S ¼ stomach; UGI ¼ upper gastrointestinal. Complete recovery indicates return to normal activity without drains.
*
The Roux-en-Y gastrojejunostomy was performed in all 6 patients as described. The mean operative time during the Roux limb anastomosis was 260 minutes in group A versus 180 minutes in group B. No intraoperative incident was noted, no transfusion was necessary, and no mortality was encountered. Three patients developed a small leak at the anastomosis, which healed spontaneously after a mean time of 11 days. The length of hospital stay after the Roux limb anastomosis was 12.75 days (range 8–17 d) for group A versus 13.5 days for group B (7–20 d). The mean time until complete recovery (removal of all drains and return to normal daily activity) in group A was 131.25 days (range 99–165 d) and 38 days (28–48 d) for group B. The 4 patients in whom the Roux limb was delayed (group A) had a median increase of all medical cost by 500%, whereas the 2 patients who underwent early Roux limb (group B) had an increase by 200% (Table 2). A swallow study was performed approximately 3 months after surgery showing that both the sleeve and the Roux-enY limb were functioning in 3 patients, only the sleeve was functioning in 2 patients, and the Roux limb was the only functioning pathway in 1 patient. At a mean follow-up of 27.5 months (range from 13–44 mo), all 6 patients were tolerating the procedure well. Mean percentage of excess weight loss was 58.8%. Three patients had vitamin B12 deficiency, and 2 patients had iron deficiency. Discussion Leaks after LSG are reported to occur in between .7% to 5% (mean 2.3%) of cases [2]. Clinically, they may range from mild micro leaks that present from weeks to months after surgery as the cause of perisleeve abscesses and chronic fistula to an abdominal catastrophe with sepsis, hemodynamic instability, multisystem organ failure, and rarely, patient demise [2,4,13]. Management options vary and depend on the timing and clinical presentation of the leak. Immediate reoperation is the preferred course of action for the unstable patient, usually with washout, irrigation of the abdominal cavity, wide drainage, and an attempt at suturing of the leak if the tissue condition allows it [2]. Stable patients and leaks presenting
later in the postoperative course pose a yet unanswered dilemma regarding the best treatment algorithm. The methods used to date in attempts at leak closure have been percutaneous, endoscopic, or surgical [14]. According to the International Sleeve Gastrectomy Expert Panel (Consensus Statement 2011), the surgeon should wait until 12 weeks after conservative therapy to allow for healing and avoid thick adhesions during reoperation. The operation that most experts would perform is conversion to RYGB [3]. In our experience, all attempts for conservative management failed, and the 6 cases were eventually treated surgically. We prefer the Roux limb gastrojejunal anastomosis as described by Baltasar et al. [11]. This approach should be attempted by experienced surgeons who can deal with any operative complications that could lead to change in the plan. In this study, the 2 surgeons had at least 10 years of experience with bariatric and advanced upper and lower gastrointestinal surgeries. All were open surgeries, to avoid an increase in operative time and co-morbidities, but the technique is feasible laparoscopically, and we may attempt that approach in the future. In a recent article, Van de Vrande et al. described successful laparoscopic Roux limb placement in 11 patients with chronic postLSG fistula [15]. The addition of a Roux limb aims at transforming a highpressure system, constituted by the sleeve construction, into a low-pressure system. A low-pressure system reportedly facilitates fistula healing in foregut surgery [16,17]. The concept is the same as the concept behind performing a RYGB. However, RYGB requires transection of the stomach, a step that can be very hazardous because of the tissue changes caused by the chronic inflammation. Compared with total gastrectomy or RYGB conversion, Roux limb anastomosis spares the patient the metabolic consequences and the decrease in quality of life that can be associated with a RYGB or a total gastrectomy [15]. The factors that may contribute to nonoperative management failure in our series are the inexperience of the gastroenterologist performing the procedures, the use of stents 2 months after the leak was diagnosed, and the fact that these prostheses, in principle, are designed for use in esophageal disease (stenosis). When used to treat leaks, the usual placement site is at the end of the esophagus and the
110
M. Chour et al. / Surgery for Obesity and Related Diseases 10 (2014) 106–111
proximal part of the gastric sleeve, which frequently leads to migration. Nonoperative management of a chronic leak after LSG requires a multimodal approach and a range of repetitive endoscopic therapies, which become very expensive, often with unrewarding results. There is currently no evidencebased algorithm for the treatment of leaks, including guidance for which modality to try first [18]. The mean operative time during the Roux limb anastomosis was 260 minutes in group A versus 180 minutes in group B, which could be explained by the greater fibrosis and adhesions that formed after stent placement. In our study, all conservative management failed, and the 6 patients required Roux limb surgery. When comparing the 2 groups, the 2 patients who underwent early Roux limb anastomosis (group B) had a decrease of all medical cost and a reduced time until complete recovery (removal of all drains, adequate oral intake, and return to normal daily activity; Table 2). According to our experience, especially in the last 2 cases, we now believe that early intervention is a good option for leaks that did not close after 4 weeks of conservative management. After a leak is diagnosed, adequate drainage is of paramount importance and a mandatory adjunct for any endoscopic interventions. This can be achieved with a drain placed during initial surgery or with additional imaging-guided drains. Percutaneous imaging-guided drainage and distal enteral or parenteral hyperalimentation together with systemic broad-spectrum antibiotics are the mainstay of the nonoperative management. The International Sleeve Gastrectomy Expert Panel stated that the use of a stent is a valid treatment option for an acute proximal leak (o7 d); after 30 days, the likelihood of a leak to seal by exclusion using a stent is very low [3]. We add that if the leak output does not decrease in 4 weeks, early Roux limb anastomosis should be considered. Roux-en-Y gastrojejunostomy over the leak site has many advantages. It is an effective technique to treat staple line leak, can be adopted as early treatment (in selected patients and after drainage), reduces the cost and length of hospital stay, and reduces the delay in leak healing, all of which should decrease the stress on the patient and the family. The limitations of the present study are the small number of cases and the need to be compared with postsleeve leak cases that were treated by other surgical interventions (conversion to RYGB or Roux-en-Y esophago-jejunostomy anastomosis). Longer follow-up is needed to determine whether any other secondary effects (gastroesophageal reflux disease, increasing gastric acidity, anastomotic ulcers, and metabolic perturbation) will occur. Conclusions The gastroesophageal junction is a potential site of leakage after sleeve gastrectomy that is usually difficult to treat conservatively. Roux-en-Y gastrojejunostomy is an effective
technique to treat refractory staple line leakage and can be adopted as early treatment in selected patients after stabilization, thereby reducing the cost and length of hospital stays. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. References [1] Dapri G, Vaz C, Cadière GB, Himpens J. A prospective randomized study comparing two different techniques for laparoscopic sleeve gastrectomy. Obes Surg 2007;17:1435–41. [2] Burgos A, Braghetto I, Csendes A, et al. Gastric leak after laparoscopicsleeve gastrectomy for obesity. Obes Surg 2009;19:1672–7. [3] Rosenthal RJ, International Sleeve Gastrectomy Expert Panel, Diaz AA, Arvidsson D, et al. International Sleeve Gastrectomy Expert Panel Consensus Statement: best practice guidelines based on experience of 412,000 cases. Surg Obes Relat Dis 2012;8:8–19. [4] Casella G, Soricelli E, Rizello M, et al. Nonsurgical treatment of staple line leaks after laparoscopic sleeve gastrectomy. Obes Surg 2009;19:821–6. [5] Papavramidis TS, Kotzampassi K, Kotidis E, Eleftheriadis EE, Papavramidis ST. Endoscopic fibrin sealing of gastrocutaneous fistulas after sleeve gastrectomy and biliopancreatic diversion with duodenal switch. J Gastroenterol Hepatol 2008;23:1802–5. [6] Akhras J, Tobi M, Zagnoon A. Endoscopic fibrin sealant injection with application of hemostatic clips: a novel method of closing a refractory gastrocutaneous fistula. Dig Dis Sci 2005;50:1872–4. [7] Serra C, Baltasar A, Andreo L, et al. Treatment of gastric leaks with coated self-expanding stents after sleeve gastrectomy. Obes Surg 2007;17:866–72. [8] Blackmon SH, Santora R, Schwarz P, Barroso A, Dunkin BJ. Utility of removable esophageal covered self-expanding metal stents for leak and fistula management. Ann Thorac Surg 2010;89:931–7. [9] Fukumoto R, Orlina J, McGinty J, Teixeira J. Use of polyflex stents in treatment of acute esophageal and gastric leaks after bariatric surgery. Surg Obes Relat Dis 2006;2:570–2. [10] Marquez MF, Ayza MF, Lozano RB, Morales Mdel M, Diez JM, Poujoulet RB. Gastric leak after laparoscopic sleeve gastrectomy. Obes Surg 2010;20:1306–11. [11] Baltasar A, Bou R, Bengochea M, Serra C, Cipagauta L. Use of a Roux Limb to correct esophagogastric junction fistulas after sleeve gastrectomy. Obes Surg 2007;17:1408–10. [12] Baltasar A, Serra C, Bengochea M, Bou R, Andreo L. Use of Roux limb as remedial surgery for sleeve gastrectomy fistulas. Surg Obes Relat Dis 2008;4:759–63. [13] Cottam D, Qureshi FG, Mattar SG, et al. Laparoscopic sleeve gastrectomy as an initial weight loss procedure for high risk patients with morbid obesity. Surg Endosc 2006;20:859–63. [14] Sakran N, Goitein D, Raziel A, et al. Gastric leaks after sleeve gastrectomy: a multicenter experience with 2,834 patients. Surg Endosc 2013;27:240–5. [15] Van de Vrande S, Himpens J, El Mourad H, Debaerdemaeker R, Leman G. Management of chronic proximal fistulas after sleeve gastrectomy by laparoscopic Roux-limb placement. Surg Obes Relat Dis. Epub 2013 Jan 16. [16] Gagner M. Leaks after sleeve gastrectomy are associated with smaller bougies: prevention and treatment strategies. Surg Laparosc Endosc Percutan Tech 2010;20:166–9. [17] Yehoshua RT, Eidelman LA, Stein M, et al. Laparoscopic sleeve gastrectomy—volume and pressure assessment. Obes Surg 2008;18:1083–8. [18] Beitner M, Cohen J, Kurian M. Endoscopic management of persistent leak after laparoscopic sleeve gastrectomy. Bariatric Times 2012;9:22–4.
