UPPER GI ANZJSurg.com
Laparoscopic sleeve gastrectomy: review of 500 cases in single surgeon Australian practice Simon C. Gibson, Philip A. Le Page and Craig J. Taylor UGI Surgery, Concord Hospital, Sydney, New South Wales, Australia
Key words bariatric surgery, gastrectomy, laparoscopic surgery, obesity, sleeve gastrectomy. Correspondence Mr Philip A. Le Page, OClinic, 21 Gillies Street, Crows Nest, Sydney, NSW 2065, Australia. Email: [email protected] S. C. Gibson MD, FRCS; P. A. Le Page MBBS, FRACS; C. J. Taylor MBBS, FRACS. Accepted for publication 26 October 2013. doi: 10.1111/ans.12483
Abstract Introduction: Reported results and techniques of laparoscopic sleeve gastrectomy (LSG) are variable. Our objective was to assess results of weight loss, complications and reflux in a large consecutive series of LSG, describing technical detail which contributed to outcomes. Methods: Retrospective review of prospectively collected data of 500 consecutive patients undergoing LSG. Patient demographics, weight loss, complications and functional outcomes were analysed and operative technique described. Results: Five hundred patients underwent LSG over 3 years (37 revisional). Mean (range) preoperative body mass index was 45 kg/m2 (35–76 kg/m2). Mean follow-up and length of hospital stay were 14 months (1–34) and 3.8 days (3–12), respectively. All-cause 30-day readmission rate 1.2%. Mean excess weight loss (interquartile range, available patient data) was 43% (22–65%, 423 patients), 58% (45–70%, 352 patients), 76% (52–84%, 258 patients), 71% (51–87%, 102 patients) and 73% (55–86%, 13 patients) at 3, 6, 12, 24, 36 months, respectively. There was no mortality. Intraoperative complications occurred in two (0.4%) – splenic bleeding; bougie related oesophageal injury. Early surgical complications in four (1.2%) patients (one staple line leak and three post-operative bleeds). Other early complications occurred in three (0.6%) patients (one pseudomembranous colitis; one central line sepsis; one portal venous thrombosis) and late in four (0.8%) patients (three port-site incisional hernias; midsleeve stricture requiring endoscopic dilatation). Gastro-oesophageal reflux symptoms decreased from 45 to 6%. Conclusion: With attention to detail, LSG can lead to good excess weight loss with minimal complications. Tenants to success include repair of hiatal laxity, generous width at angula incisura and complete resection of posterior fundus.
Introduction Although initially introduced as a staged procedure prior to definitive malabsorptive surgery in the form of biliary-pancreatic diversion, the laparoscopic sleeve gastrectomy (LSG) has since developed into a stand-alone bariatric operation associated with good, mediumterm weight loss and satisfactory complication rates when conducted in experienced hands.1–4 Meta-analysis and systematic review supports the overall efficacy of the procedure.5,6 Despite the supposed simplicity of the procedure and excellent results in selected series,2–4,7 the procedure has been associated with morbidity in the form of staple line leaks, reintervention for dysphagia, gastro-oesophageal reflux disease (GORD) and rarely mortality (albeit usually in high-risk groups).2,4,5,8–11 The long-term weight loss © 2013 Royal Australasian College of Surgeons
efficacy has also been questioned and some of these issues may have been underreported due to publication bias. Although addressed by a recent consensus document,12 there are multiple controversies regarding the technique of LSG, and this may in part be what has led to the variable published results. Our study objective was to ascertain efficacy of weight loss and complication rates in our experience of 500 consecutive cases of LSG in a single surgeon practice.
Methods A retrospective review was performed from a prospectively collected database (LapBase Data Manager for Bariatric Surgery, Cremorne, Australia) of the first 500 patients who underwent LSG by single ANZ J Surg 85 (2015) 673–677
674
bariatric surgeon (CJT) at a tertiary public referral hospital and two major private hospitals. No exclusion criteria were applied for our study. Local ethics guidelines were followed in the conduct of the review. Patient selection for LSG had been by collaborative decisionmaking between the patient and our multidisciplinary bariatric team. This comprised review by the operating surgeon, psychologist, dietician and other medical specialists when indicated. Alternative treatments were offered to patients, particularly laparoscopic adjustable gastric banding. Criteria for consideration of either surgery included age over 17 years in the absence of any medical cause for weight gain and a body mass index (BMI) > 40 kg/m2 or >35 kg/m2 in the presence of obesity-related co-morbidities. Specifically, LSG has been preferred over laparoscopic adjustable gastric banding in patients with a high BMI (>45), entrenched emotional eating patterns, high consumption of high calorie liquids and soft foods, and where geographical issues may impact upon ability to participate in regular aftercare. Data had been collected by the bariatric surgeon (CJT) and Practice staff trained in database management. Data items analysed were age, gender, preoperative BMI, symptomatic preoperative GORD, date of operation and follow-up. Data items used for outcome analysis included post-operative BMI, complications (intraoperative, post-operative, early and late) and post-operative symptomatic GORD (at 6 months). Symptomatic GORD was diagnosed when patients suffered episodes of retrosternal heartburn relieved by antacid, nocturnal pooling of pharyngeal fluid or postural regurgitation were experienced at least once per week. Follow-up was at the bariatric practice at 2 weeks, then every three months for a year followed by annual visits. Statistical analysis was performed with GraphPad Software (La Jolla, CA, USA). P value of less than 0.05 was deemed significant.
Gibson et al.
Fig. 1. ‘Critical view’ of left crus after full fundal mobilization.
hiatus hernia is present, then the phreno-oesophageal ligament is incised, the distal esophagus circumferentially mobilized into the abdominal cavity and the hiatus repaired anteriorly and posteriorly. Greater curve mobilization The greater curve is mobilized using the Ligasure device (Covidien, North Haven, CT, USA). The lesser sac is entered at the junction of body and antrum, adjacent to the gastric wall, and mobilized proximally until the peri-hiatal dissection area is reached. Any significant vessels are dealt with by a double application of the Ligasure. During this phase, all posterior adhesions to the pancreas and posterior short gastric vessels are divided, completely mobilizing the posterior fundus, preserving only the left gastric vascular pedicle. This exposes the left crus in its entirety, giving the ‘critical view’ of the LSG (Fig. 1). The greater curve mobilization is then carried distally along the antrum towards the pylorus, terminating 2 cm from the pylorus.
Surgical technique Positioning and ports Patients are placed in a supine, feet together 30-degree antiTrendelenburg position and restrained using lower limb strap and foot support. Access is gained to the abdominal cavity using an optical 12 mm port (Xcel; Johnson and Johnson Medical, North Ryde, NSW, Australia) in the left upper quadrant. Further ports are placed in a radial arc 20 cm below the xiphisternum (15 mm right and para-median, 5 mm left flank). The Nathanson liver retractor (Cook Medical Inc., Brisbane, QLD, Australia) is inserted subxiphisternally. Hiatal dissection Prior to dissection, a 32Fr wide bore hollow suction bougie (Allergan Limited, Buckinghamshire, UK) is passed into the stomach to empty it and thereby facilitate dissection. It is then temporarily withdrawn into the oesophagus. The gastro-oesophageal fat pad is mobilized with hook diathermy allowing for subsequent accurate staple line placement. Peritoneum over the angle of His is then divided, facilitating subsequent posterior fundal mobilization. The hiatus is assessed for laxity and, if present, 0 Ethibond (Johnson and Johnson Medical, New Brunswick, NJ, USA) suture is placed anteriorly to reapproximate the left and right crura. If an obvious
Staple technique In preparation for stapling, the hollow 32Fr bougie is then passed along the lesser curve into the distal antrum. Placing this on suction ensures, it holds its position thereby consistently maintaining the correct calibration and preventing a wider gastric tube. Key landmarks are used to guide correct stapler position and achieve distinct horizontal and vertical stapling components. Transection commences 2 cm proximal to the pylorus, and typically involves two horizontal firings of the stapler from the far right-sided laparoscopic port. The focus during the second firing is to terminate diagonally opposite the angula incisura, identified by the ‘pes anserinus’ vascular pattern. A deliberate drift away from the bougie should occur here so as to ensure the angula incisura becomes the widest point of the sleeve (Fig. 2). The transection is then continued in a vertical direction towards the angle of His, using the left inferior phrenic vessels as a directional alignment guide. These vessels, which lie on the left crus approximately 1 cm lateral to the cardio-oesophageal junction, are a surprisingly consistent anatomical marker and are of particular assistance when view is obscured by excessive fat. Typically, this vertical component of the transection requires four 60 mm stapler cartridges, via the midline laparoscopic port, articulating as necessary. The shaft © 2013 Royal Australasian College of Surgeons
Review 500 laparoscopic sleeve gastrectomies
675
in the latter part, we have buttressed the most proximal staple firing with Seamguard (Gore Medical, Belrose, NSW, Australia). We do not place gastropexy sutures. Fibrin sealant (Baxter, CA, USA) or fibrillar (Ethicon, Norderstedt, Germany) has been used throughout for haemostasis along the staple line.
Fig. 2. Staple line at angula incisura.
Extraction The specimen is removed via the 15 mm right-sided port. In our first 300 cases, we did not routinely suture repair the extraction port fascia. Following a small number of incisional hernias (see Results), we now routinely close this fascial defect using an 0 Vicryl suture (Johnson and Johnson Medical) placed under laparoscopic guidance using an Endoclose device (Covidien). Post-operatively Patients routinely remain in hospital till day 3 and allowed only sips of fluid with IV supplementation. On day 3, they are discharged and follow a liquid only diet for 2 weeks with a limit of 100 mL per hour. Patients then progress onto a puree diet, supplemented with multivitamins and soluble fibre, for a further 2 weeks prior to recommencing normal diet.
Results
Fig. 3. Complete fundal excision.
of a blunt grasper entering through the left flank port is used during the last two firings to elevate the mobilized posterior fundus laterally, allowing the stapler cartridge to be positioned next to the bougie, ensuring no posterior fundus is retained (Fig. 3). No lateral traction is applied to the specimen during the actual stapling phase. Initially, we had used the Echilon Flex stapler (Johnson and Johnson Medical) for the first 350 cases with ‘Green’ (60 × 4.1 mm) cartridges for all parts of the stomach transection. After trialing, from February 2012, the Covidien Tristaple (Covidien) with ‘Black’ cartridges was used. Each individual firing is assessed for integrity and any excess staples are removed from the crotch prior to further firing to avoid fouling of the blade. Consecutive stapling is placed such that the central blade of the stapler lies within the crotch of the previous stapling. Staple line reinforcement Our preferred management of the proximal staple line has evolved over time. Initially, omentum was sutured to the most proximal 10 cm of staple line. Subsequently, suturing to the staple line directly was not done but either the partially mobilized fat pad or omentum was used to cover the angle of His by suturing to the left crus using interrupted polydiaxone sutures (Johnson and Johnson Medical). In the later part of the series, omentum only was used to suture to the left crus, thereby covering the staple line and occupy any ‘deadspace’ that may be present adjacent to the stapled apex. Additionally, © 2013 Royal Australasian College of Surgeons
Between November 2009 and August 2012, 500 consecutive patients underwent LSG. Female patients comprised 68%, the mean (range) BMI (kg/m2) was 45 (35–76), and mean (range) patient age was 41 (17–73) years. In 37 patients (7%), LSG was performed as a revision procedure following unsuccessful (34) or complicated (two erosions, one recurrent slippage) gastric banding. In all cases, conversion from gastric band to sleeve was performed as a two-stage procedure with minimum interval of 3 months. Mean (range) follow-up for the whole group was 14 (1–34) months. All operations were successfully completed laparoscopically without need for open conversion. Operative time steadily decreased with increasing experience, reducing from a mean (range) of 81 (42–180) minutes in the first 100 primary cases to 47 (36–115) in the last 100 cases (P < 0.001). Mean length of hospital stay was 3.8 days (range 3–12). The all-cause 30-day readmission rate was 1.2%. Mean loss of excess weight was 43% (interquartile range (IQR) 22–65%, 423 patients), 58% (IQR 45–70%, 352 patients), 76% (IQR 52–84%, 258 patients), 71% (IQR 51–87%, 102 patients) and 73% (IQR 55–86%, 13 patients) at 3, 6, 12, 24 and 36 months, respectively. This is depicted graphically in Figure 4. From an initial mean BMI 45, the mean and IQR for postoperative BMI at 3, 6, 12, 24 and 36 months was 36.4 (IQR 40.6–32, 423 patients), 33.4 (IQR 36–31, 352 patients), 29.8 (IQR 34.6–28.2, 258 patients), 30.8 (IQR 34.8, 27.6, 102 patients) and 30.4 (IQR 34–27.8, 13 patients). There was no mortality. Complications are summarized in Table 1. Intraoperative complications occurred in two (0.4%) patients. One patient who had extensive gastrosplenic adhesions from previous gastric banding developed troublesome splenic bleeding mandating laparoscopic splenectomy. One patient had a perforation of their distal oesophagus by the anaesthetist passing the bougie, and was repaired primarily without complication. Early
676
Gibson et al.
cation. In contrast, post-sleeve reflux symptoms were present in just 30 (6%) patients, manifest as simple heartburn responding to acid lowering medication in 26 (5.2%) patients, significant belching in two (0.4%) patients or volume reflux in two (0.4%) of the patients. During sleeve gastrectomy, anterior repair of hiatal laxity was performed in 265 (53%) patients, while formal repair of hiatus hernia was required in a further 30 (6%) patients.
100 90 80 70
%EWL
60 50
Discussion 40 30 25th percentile
20
Mean 75th percentile
10 0 0 (500)
3 (423)
6 (352)
12 (258)
24 (102)
36 (13)
Post-operative months (Number of patients reaching each time-period milestone at census with available data in parentheses)
Fig. 4. Loss of excess weight (EWL) over time (mean and interquartile range).
Table 1 Post-operative complications Complication
%
Number
Oesophageal bougie injury
0.2%
1
Splenic injury
0.2%
1
Staple line abscess/leak
0.2%
1
Post-operative haemorrhage 0.6%
1 2
Portal vein thrombosis
0.2%
1
DVT or PE† Late port-site hernia
0 0.6%
0 3
Stricture Wound infection Other minor Total
0.2% 0 0.8% 3%
1 0 4 15
Management Laparoscopic repair (during case) Laparoscopic splenectomy (during case) Drainage and antibiotics Conservative Transfusion/ laparoscopic washout Anticoagulation for 3 months Laparoscopic mesh repair Endoscopic dilatation
†Deep venous thrombosis; pulmonary embolus.
surgical complications occurred in four (1.2%) patients. There was only one proximal staple line leak, managed by laparoscopic washout and drainage allowing control of a chronic low volume fistula. Other complications occurred early in three (0.6%) patients, one of which had developed portal venous thrombosis as diagnosed on computed tomography. Late complications occurred in four (0.8%) patients, including three 15 mm port-site incisional hernias in the period prior to suture closure. Two hundred twenty-five (45%) patients reported GORD symptoms preoperatively, with 130 (26%) already on acid lowering medi-
With the increased use of LSG as a definitive bariatric operation, it has become increasingly important to ascertain its safety and efficacy. This series has demonstrated that the sleeve gastrectomy can lead to significant medium-term excess weight loss (EWL) with a low complication rate and an excellent functional result with minimum reintervention required. Our results of EWL showed that they peaked around 12 months post-operatively and were maintained to 36 months, favourable in comparison with other studies.13 Like Himpens,4 we believe complete fundal excision is required to achieve such results and also to prevent GORD. Critically, complete posterior fundal mobilization is required and achieved by division of both anterior and posterior short gastric vessels with exposure of the entire left crus and lateral edge of the phreno-oesophageal ligament. We also believe in the importance of a complete sleeve gastrectomy (as distinct from the ‘vertical’ sleeve gastrectomy, popular in North America and elsewhere) with stapling initiated very close to the pylorus (within 1–2 cm). This variant results in distinct horizontal (antral) and vertical (body and fundus) staple lines, and may have several advantages over the more common antral-preserving vertical sleeve gastrectomy. There is a greater reduction in gastric volume, and possibly an increased rate of gastric emptying, particularly to liquids, an effect which may not occur to the same extent with antral sparing.14 This latter point may explain the dumping-like effect frequently reported by patients following the consumption of liquids high in sugar or fat such as milk shakes, helping to reduce the desire for such substances. There may also be beneficial diabetic effects from earlier gastric emptying with nutrients reaching the distal small bowel sooner.15 Another benefit of distinct horizontal and vertical components is that it naturally allows for a greater margin at the angula incisura. The importance of a wider margin at the angula must be emphasized; a natural change in plane occurs between the gastric body and antrum of approximately 30 degrees. The sleeve may twist and narrow at this point if this is not allowed for, increasing the risk of complications including GORD, food intolerance and even proximal leak.16 Our series has demonstrated a low requirement for reintervention due to food intolerance (one patient) and we believe that this is due to the respect we have for the angula incisura. We did not perform gastropexy sutures to control the final position of the sleeve staple line, and feel they are not required provided the angula is respected, and the stomach is not pulled or distorted during the stapling phase. Overall complication rates in our series also compare favourably13 and importantly no complication had led to any mortality. The single leak (0.2%) that we experienced presented as a non-communicating abscess adjacent to the proximal staple line. The majority of reported © 2013 Royal Australasian College of Surgeons
Review 500 laparoscopic sleeve gastrectomies
leaks from the staple line occur proximally. It is unclear whether this is a pressure, technical or ischaemic phenomenon. Significant controversy exists as to the best method of leak prevention. Musella et al.16 have shown that over sewing the staple line itself does not reduce staple line leaks but increases stenosis of the sleeve, whereas Ser et al. have demonstrated a significantly reduced leak rate (10% versus 0%) when the staple line was reinforced.16,17 Although patching the proximal staple line appears effective in protecting the proximal staple line, buttressing with Seamguard saves time and avoids the risk of hematoma albeit at slightly increased cost.18 We have used two techniques in this series to reinforce the proximal staple line: a simple fat pad/omental patch over the proximal corner, and commercial buttressed staple line reinforcement (Seamguard). While both techniques appear to be equally effective, we have evolved our technique to use Seamguard and omental patching, avoiding use of the fat pad after the only leak occurred. Additionally, in prevention, we believe respecting a 1 cm margin between the staple line and the lateral side of the oesophagus is important. The potential for chronicity of leaks after sleeve gastrectomy remains a key challenge for the procedure. We have learnt other lessons during our experience. Following our experience with a bougie perforation, our technique quite deliberately was to align the lesser curve to the oesophagus thereby facilitating passage. The splenic bleed we experienced was significantly because of the dense adhesion from a previous band. Given that, we also had three patients with post-operative haemorrhage, we have become quite deliberate in removing any tension from the short gastrics while the sealing device was activated. Post-operative anticoagulation may contribute but must be balanced against the risk of thromboembolism. While we did not experience any deep venous thromboses, one patient developed a portal vein thrombosis. Finally, reassuringly, port-site hernia was abolished after placing closure suture to the 15 mm lateral port. Unlike in some series, symptoms of GORD improved in our cohort.4,19 We feel this is due to our aggressive approach to the hiatus with routine assessment and repair of the hiatus for laxity or hernia (required in more than half of patients). This approach is facilitated by commencing the procedure at the hiatus thus clarifying whether hiatal repair is needed, identifying the precise position of the oesophagus and opening the angle of His to aid fundal mobilization. Limitations of our study primarily relate to its retrospective nature. Exemplifying this is that we have not controlled the study during changes in technique thereby preventing direct comparison. However, the large numbers and consistency of a single surgeon practice do make the results significant. In conclusion, the sleeve gastrectomy is a safe bariatric procedure with a low risk of complications. Tenants to successful sleeve gastrectomy include repair of hiatal laxity or hernia, generous width at the angula incisura, avoidance of twists and complete resection of the posterior fundus.
References 1. Omana JJ, Nguyen SQ, Herron D, Kini S. Comparison of comorbidity resolution and improvement between laparoscopic sleeve gastrectomy
© 2013 Royal Australasian College of Surgeons
677
2.
3.
4. 5.
6. 7.
8.
9.
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
and laparoscopic adjustable gastric banding. Surg. Endosc. 2010; 24: 2513–7. Bellanger DE, Greenway FL. Laparoscopic sleeve gastrectomy, 529 cases without a leak: short-term results and technical considerations. Obes. Surg. 2011; 21: 146–50. Gadiot RP, Biter LU, Zengerink HJ et al. Laparoscopic sleeve gastrectomy with an extensive posterior mobilization: technique and preliminary results. Obes. Surg. 2012; 22: 320–9. Himpens J, Dobbeleir J, Peeters G. Long-term results of laparoscopic sleeve gastrectomy for obesity. Ann. Surg. 2010; 252: 319–24. Brethauer SA, Hammel JP, Schauer PR. Systematic review of sleeve gastrectomy as staging and primary bariatric procedure. Surg. Obes. Relat. Dis. 2009; 5: 469–75. Aggarwal S, Kini SU, Herron DM. Laparoscopic sleeve gastrectomy for morbid obesity: a review. Surg. Obes. Relat. Dis. 2007; 3: 189–94. Lalor PF, Tucker ON, Szomstein S, Rosenthal RJ. Complications after laparoscopic sleeve gastrectomy. Surg. Obes. Relat. Dis. 2008; 4: 33– 8. Kelogrigoris M, Sotiropoulou E, Stathopoulos K, Georgiadou V, Philippousis P, Thanos L. CT-guided percutaneous drainage of infected collections due to gastric leak after sleeve gastrectomy for morbid obesity: initial experience. Cardiovasc. Intervent. Radiol. 2011; 34: 585–9. Stroh C, Birk D, Flade-Kuthe R et al. Results of sleeve gastrectomy-data from a nationwide survey on bariatric surgery in Germany. Obes. Surg. 2009; 19: 632–40. 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. Keidar A, Appelbaum L, Schweiger C, Elazary R, Baltasar A. Dilated upper sleeve can be associated with severe postoperative gastroesophageal dysmotility and reflux. Obes. Surg. 2010; 20: 140–7. Rosenthal RJ, International Sleeve Gastrectomy Expert P, Diaz AA et al. International sleeve gastrectomy expert panel consensus statement: best practice guidelines based on experience of >12 000 cases. Surg. Obes. Relat. Dis. 2012; 8: 8–19. Shi X, Karmali S, Sharma AM, Birch DW. A review of laparoscopic sleeve gastrectomy for morbid obesity. Obes. Surg 2010; 20: 1171–7. Bernstine H, Tzioni-Yehoshua R, Groshar D et al. Gastric emptying is not affected by sleeve gastrectomy–scintigraphic evaluation of gastric emptying after sleeve gastrectomy without removal of the gastric antrum. Obes. Surg. 2009; 19: 293–8. Shah S, Shah P, Todkar J, Gagner M, Sonar S, Solav S. Prospective controlled study of effect of laparoscopic sleeve gastrectomy on small bowel transit time and gastric emptying half-time in morbidly obese patients with type 2 diabetes mellitus. Surg. Obes. Relat. Dis. 2010; 6: 152–7. Musella M, Milone M, Bellini M, Leongito M, Guarino R, Milone F. Laparoscopic sleeve gastrectomy. Do we need to oversew the staple line? Ann. Ital. Chir. 2011; 82: 273–7. Ser KH, Lee WJ, Lee YC, Chen JC, Su YH, Chen SC. Experience in laparoscopic sleeve gastrectomy for morbidly obese Taiwanese: stapleline reinforcement is important for preventing leakage. Surg. Endosc. 2010; 24: 2253–9. Dapri G, Cadiere GB, Himpens J. Reinforcing the staple line during laparoscopic sleeve gastrectomy: prospective randomized clinical study comparing three different techniques. Obes. Surg. 2010; 20: 462–7. Chiu S, Birch DW, Shi X, Sharma AM, Karmali S. Effect of sleeve gastrectomy on gastroesophageal reflux disease: a systematic review. Surg. Obes. Relat. Dis. 2011; 7: 510–5.
Laparoscopic sleeve gastrectomy: review of 500 cases in single surgeon Australian practice Simon C. Gibson, Philip A. Le Page and Craig J. Taylor UGI Surgery, Concord Hospital, Sydney, New South Wales, Australia
Key words bariatric surgery, gastrectomy, laparoscopic surgery, obesity, sleeve gastrectomy. Correspondence Mr Philip A. Le Page, OClinic, 21 Gillies Street, Crows Nest, Sydney, NSW 2065, Australia. Email: [email protected] S. C. Gibson MD, FRCS; P. A. Le Page MBBS, FRACS; C. J. Taylor MBBS, FRACS. Accepted for publication 26 October 2013. doi: 10.1111/ans.12483
Abstract Introduction: Reported results and techniques of laparoscopic sleeve gastrectomy (LSG) are variable. Our objective was to assess results of weight loss, complications and reflux in a large consecutive series of LSG, describing technical detail which contributed to outcomes. Methods: Retrospective review of prospectively collected data of 500 consecutive patients undergoing LSG. Patient demographics, weight loss, complications and functional outcomes were analysed and operative technique described. Results: Five hundred patients underwent LSG over 3 years (37 revisional). Mean (range) preoperative body mass index was 45 kg/m2 (35–76 kg/m2). Mean follow-up and length of hospital stay were 14 months (1–34) and 3.8 days (3–12), respectively. All-cause 30-day readmission rate 1.2%. Mean excess weight loss (interquartile range, available patient data) was 43% (22–65%, 423 patients), 58% (45–70%, 352 patients), 76% (52–84%, 258 patients), 71% (51–87%, 102 patients) and 73% (55–86%, 13 patients) at 3, 6, 12, 24, 36 months, respectively. There was no mortality. Intraoperative complications occurred in two (0.4%) – splenic bleeding; bougie related oesophageal injury. Early surgical complications in four (1.2%) patients (one staple line leak and three post-operative bleeds). Other early complications occurred in three (0.6%) patients (one pseudomembranous colitis; one central line sepsis; one portal venous thrombosis) and late in four (0.8%) patients (three port-site incisional hernias; midsleeve stricture requiring endoscopic dilatation). Gastro-oesophageal reflux symptoms decreased from 45 to 6%. Conclusion: With attention to detail, LSG can lead to good excess weight loss with minimal complications. Tenants to success include repair of hiatal laxity, generous width at angula incisura and complete resection of posterior fundus.
Introduction Although initially introduced as a staged procedure prior to definitive malabsorptive surgery in the form of biliary-pancreatic diversion, the laparoscopic sleeve gastrectomy (LSG) has since developed into a stand-alone bariatric operation associated with good, mediumterm weight loss and satisfactory complication rates when conducted in experienced hands.1–4 Meta-analysis and systematic review supports the overall efficacy of the procedure.5,6 Despite the supposed simplicity of the procedure and excellent results in selected series,2–4,7 the procedure has been associated with morbidity in the form of staple line leaks, reintervention for dysphagia, gastro-oesophageal reflux disease (GORD) and rarely mortality (albeit usually in high-risk groups).2,4,5,8–11 The long-term weight loss © 2013 Royal Australasian College of Surgeons
efficacy has also been questioned and some of these issues may have been underreported due to publication bias. Although addressed by a recent consensus document,12 there are multiple controversies regarding the technique of LSG, and this may in part be what has led to the variable published results. Our study objective was to ascertain efficacy of weight loss and complication rates in our experience of 500 consecutive cases of LSG in a single surgeon practice.
Methods A retrospective review was performed from a prospectively collected database (LapBase Data Manager for Bariatric Surgery, Cremorne, Australia) of the first 500 patients who underwent LSG by single ANZ J Surg 85 (2015) 673–677
674
bariatric surgeon (CJT) at a tertiary public referral hospital and two major private hospitals. No exclusion criteria were applied for our study. Local ethics guidelines were followed in the conduct of the review. Patient selection for LSG had been by collaborative decisionmaking between the patient and our multidisciplinary bariatric team. This comprised review by the operating surgeon, psychologist, dietician and other medical specialists when indicated. Alternative treatments were offered to patients, particularly laparoscopic adjustable gastric banding. Criteria for consideration of either surgery included age over 17 years in the absence of any medical cause for weight gain and a body mass index (BMI) > 40 kg/m2 or >35 kg/m2 in the presence of obesity-related co-morbidities. Specifically, LSG has been preferred over laparoscopic adjustable gastric banding in patients with a high BMI (>45), entrenched emotional eating patterns, high consumption of high calorie liquids and soft foods, and where geographical issues may impact upon ability to participate in regular aftercare. Data had been collected by the bariatric surgeon (CJT) and Practice staff trained in database management. Data items analysed were age, gender, preoperative BMI, symptomatic preoperative GORD, date of operation and follow-up. Data items used for outcome analysis included post-operative BMI, complications (intraoperative, post-operative, early and late) and post-operative symptomatic GORD (at 6 months). Symptomatic GORD was diagnosed when patients suffered episodes of retrosternal heartburn relieved by antacid, nocturnal pooling of pharyngeal fluid or postural regurgitation were experienced at least once per week. Follow-up was at the bariatric practice at 2 weeks, then every three months for a year followed by annual visits. Statistical analysis was performed with GraphPad Software (La Jolla, CA, USA). P value of less than 0.05 was deemed significant.
Gibson et al.
Fig. 1. ‘Critical view’ of left crus after full fundal mobilization.
hiatus hernia is present, then the phreno-oesophageal ligament is incised, the distal esophagus circumferentially mobilized into the abdominal cavity and the hiatus repaired anteriorly and posteriorly. Greater curve mobilization The greater curve is mobilized using the Ligasure device (Covidien, North Haven, CT, USA). The lesser sac is entered at the junction of body and antrum, adjacent to the gastric wall, and mobilized proximally until the peri-hiatal dissection area is reached. Any significant vessels are dealt with by a double application of the Ligasure. During this phase, all posterior adhesions to the pancreas and posterior short gastric vessels are divided, completely mobilizing the posterior fundus, preserving only the left gastric vascular pedicle. This exposes the left crus in its entirety, giving the ‘critical view’ of the LSG (Fig. 1). The greater curve mobilization is then carried distally along the antrum towards the pylorus, terminating 2 cm from the pylorus.
Surgical technique Positioning and ports Patients are placed in a supine, feet together 30-degree antiTrendelenburg position and restrained using lower limb strap and foot support. Access is gained to the abdominal cavity using an optical 12 mm port (Xcel; Johnson and Johnson Medical, North Ryde, NSW, Australia) in the left upper quadrant. Further ports are placed in a radial arc 20 cm below the xiphisternum (15 mm right and para-median, 5 mm left flank). The Nathanson liver retractor (Cook Medical Inc., Brisbane, QLD, Australia) is inserted subxiphisternally. Hiatal dissection Prior to dissection, a 32Fr wide bore hollow suction bougie (Allergan Limited, Buckinghamshire, UK) is passed into the stomach to empty it and thereby facilitate dissection. It is then temporarily withdrawn into the oesophagus. The gastro-oesophageal fat pad is mobilized with hook diathermy allowing for subsequent accurate staple line placement. Peritoneum over the angle of His is then divided, facilitating subsequent posterior fundal mobilization. The hiatus is assessed for laxity and, if present, 0 Ethibond (Johnson and Johnson Medical, New Brunswick, NJ, USA) suture is placed anteriorly to reapproximate the left and right crura. If an obvious
Staple technique In preparation for stapling, the hollow 32Fr bougie is then passed along the lesser curve into the distal antrum. Placing this on suction ensures, it holds its position thereby consistently maintaining the correct calibration and preventing a wider gastric tube. Key landmarks are used to guide correct stapler position and achieve distinct horizontal and vertical stapling components. Transection commences 2 cm proximal to the pylorus, and typically involves two horizontal firings of the stapler from the far right-sided laparoscopic port. The focus during the second firing is to terminate diagonally opposite the angula incisura, identified by the ‘pes anserinus’ vascular pattern. A deliberate drift away from the bougie should occur here so as to ensure the angula incisura becomes the widest point of the sleeve (Fig. 2). The transection is then continued in a vertical direction towards the angle of His, using the left inferior phrenic vessels as a directional alignment guide. These vessels, which lie on the left crus approximately 1 cm lateral to the cardio-oesophageal junction, are a surprisingly consistent anatomical marker and are of particular assistance when view is obscured by excessive fat. Typically, this vertical component of the transection requires four 60 mm stapler cartridges, via the midline laparoscopic port, articulating as necessary. The shaft © 2013 Royal Australasian College of Surgeons
Review 500 laparoscopic sleeve gastrectomies
675
in the latter part, we have buttressed the most proximal staple firing with Seamguard (Gore Medical, Belrose, NSW, Australia). We do not place gastropexy sutures. Fibrin sealant (Baxter, CA, USA) or fibrillar (Ethicon, Norderstedt, Germany) has been used throughout for haemostasis along the staple line.
Fig. 2. Staple line at angula incisura.
Extraction The specimen is removed via the 15 mm right-sided port. In our first 300 cases, we did not routinely suture repair the extraction port fascia. Following a small number of incisional hernias (see Results), we now routinely close this fascial defect using an 0 Vicryl suture (Johnson and Johnson Medical) placed under laparoscopic guidance using an Endoclose device (Covidien). Post-operatively Patients routinely remain in hospital till day 3 and allowed only sips of fluid with IV supplementation. On day 3, they are discharged and follow a liquid only diet for 2 weeks with a limit of 100 mL per hour. Patients then progress onto a puree diet, supplemented with multivitamins and soluble fibre, for a further 2 weeks prior to recommencing normal diet.
Results
Fig. 3. Complete fundal excision.
of a blunt grasper entering through the left flank port is used during the last two firings to elevate the mobilized posterior fundus laterally, allowing the stapler cartridge to be positioned next to the bougie, ensuring no posterior fundus is retained (Fig. 3). No lateral traction is applied to the specimen during the actual stapling phase. Initially, we had used the Echilon Flex stapler (Johnson and Johnson Medical) for the first 350 cases with ‘Green’ (60 × 4.1 mm) cartridges for all parts of the stomach transection. After trialing, from February 2012, the Covidien Tristaple (Covidien) with ‘Black’ cartridges was used. Each individual firing is assessed for integrity and any excess staples are removed from the crotch prior to further firing to avoid fouling of the blade. Consecutive stapling is placed such that the central blade of the stapler lies within the crotch of the previous stapling. Staple line reinforcement Our preferred management of the proximal staple line has evolved over time. Initially, omentum was sutured to the most proximal 10 cm of staple line. Subsequently, suturing to the staple line directly was not done but either the partially mobilized fat pad or omentum was used to cover the angle of His by suturing to the left crus using interrupted polydiaxone sutures (Johnson and Johnson Medical). In the later part of the series, omentum only was used to suture to the left crus, thereby covering the staple line and occupy any ‘deadspace’ that may be present adjacent to the stapled apex. Additionally, © 2013 Royal Australasian College of Surgeons
Between November 2009 and August 2012, 500 consecutive patients underwent LSG. Female patients comprised 68%, the mean (range) BMI (kg/m2) was 45 (35–76), and mean (range) patient age was 41 (17–73) years. In 37 patients (7%), LSG was performed as a revision procedure following unsuccessful (34) or complicated (two erosions, one recurrent slippage) gastric banding. In all cases, conversion from gastric band to sleeve was performed as a two-stage procedure with minimum interval of 3 months. Mean (range) follow-up for the whole group was 14 (1–34) months. All operations were successfully completed laparoscopically without need for open conversion. Operative time steadily decreased with increasing experience, reducing from a mean (range) of 81 (42–180) minutes in the first 100 primary cases to 47 (36–115) in the last 100 cases (P < 0.001). Mean length of hospital stay was 3.8 days (range 3–12). The all-cause 30-day readmission rate was 1.2%. Mean loss of excess weight was 43% (interquartile range (IQR) 22–65%, 423 patients), 58% (IQR 45–70%, 352 patients), 76% (IQR 52–84%, 258 patients), 71% (IQR 51–87%, 102 patients) and 73% (IQR 55–86%, 13 patients) at 3, 6, 12, 24 and 36 months, respectively. This is depicted graphically in Figure 4. From an initial mean BMI 45, the mean and IQR for postoperative BMI at 3, 6, 12, 24 and 36 months was 36.4 (IQR 40.6–32, 423 patients), 33.4 (IQR 36–31, 352 patients), 29.8 (IQR 34.6–28.2, 258 patients), 30.8 (IQR 34.8, 27.6, 102 patients) and 30.4 (IQR 34–27.8, 13 patients). There was no mortality. Complications are summarized in Table 1. Intraoperative complications occurred in two (0.4%) patients. One patient who had extensive gastrosplenic adhesions from previous gastric banding developed troublesome splenic bleeding mandating laparoscopic splenectomy. One patient had a perforation of their distal oesophagus by the anaesthetist passing the bougie, and was repaired primarily without complication. Early
676
Gibson et al.
cation. In contrast, post-sleeve reflux symptoms were present in just 30 (6%) patients, manifest as simple heartburn responding to acid lowering medication in 26 (5.2%) patients, significant belching in two (0.4%) patients or volume reflux in two (0.4%) of the patients. During sleeve gastrectomy, anterior repair of hiatal laxity was performed in 265 (53%) patients, while formal repair of hiatus hernia was required in a further 30 (6%) patients.
100 90 80 70
%EWL
60 50
Discussion 40 30 25th percentile
20
Mean 75th percentile
10 0 0 (500)
3 (423)
6 (352)
12 (258)
24 (102)
36 (13)
Post-operative months (Number of patients reaching each time-period milestone at census with available data in parentheses)
Fig. 4. Loss of excess weight (EWL) over time (mean and interquartile range).
Table 1 Post-operative complications Complication
%
Number
Oesophageal bougie injury
0.2%
1
Splenic injury
0.2%
1
Staple line abscess/leak
0.2%
1
Post-operative haemorrhage 0.6%
1 2
Portal vein thrombosis
0.2%
1
DVT or PE† Late port-site hernia
0 0.6%
0 3
Stricture Wound infection Other minor Total
0.2% 0 0.8% 3%
1 0 4 15
Management Laparoscopic repair (during case) Laparoscopic splenectomy (during case) Drainage and antibiotics Conservative Transfusion/ laparoscopic washout Anticoagulation for 3 months Laparoscopic mesh repair Endoscopic dilatation
†Deep venous thrombosis; pulmonary embolus.
surgical complications occurred in four (1.2%) patients. There was only one proximal staple line leak, managed by laparoscopic washout and drainage allowing control of a chronic low volume fistula. Other complications occurred early in three (0.6%) patients, one of which had developed portal venous thrombosis as diagnosed on computed tomography. Late complications occurred in four (0.8%) patients, including three 15 mm port-site incisional hernias in the period prior to suture closure. Two hundred twenty-five (45%) patients reported GORD symptoms preoperatively, with 130 (26%) already on acid lowering medi-
With the increased use of LSG as a definitive bariatric operation, it has become increasingly important to ascertain its safety and efficacy. This series has demonstrated that the sleeve gastrectomy can lead to significant medium-term excess weight loss (EWL) with a low complication rate and an excellent functional result with minimum reintervention required. Our results of EWL showed that they peaked around 12 months post-operatively and were maintained to 36 months, favourable in comparison with other studies.13 Like Himpens,4 we believe complete fundal excision is required to achieve such results and also to prevent GORD. Critically, complete posterior fundal mobilization is required and achieved by division of both anterior and posterior short gastric vessels with exposure of the entire left crus and lateral edge of the phreno-oesophageal ligament. We also believe in the importance of a complete sleeve gastrectomy (as distinct from the ‘vertical’ sleeve gastrectomy, popular in North America and elsewhere) with stapling initiated very close to the pylorus (within 1–2 cm). This variant results in distinct horizontal (antral) and vertical (body and fundus) staple lines, and may have several advantages over the more common antral-preserving vertical sleeve gastrectomy. There is a greater reduction in gastric volume, and possibly an increased rate of gastric emptying, particularly to liquids, an effect which may not occur to the same extent with antral sparing.14 This latter point may explain the dumping-like effect frequently reported by patients following the consumption of liquids high in sugar or fat such as milk shakes, helping to reduce the desire for such substances. There may also be beneficial diabetic effects from earlier gastric emptying with nutrients reaching the distal small bowel sooner.15 Another benefit of distinct horizontal and vertical components is that it naturally allows for a greater margin at the angula incisura. The importance of a wider margin at the angula must be emphasized; a natural change in plane occurs between the gastric body and antrum of approximately 30 degrees. The sleeve may twist and narrow at this point if this is not allowed for, increasing the risk of complications including GORD, food intolerance and even proximal leak.16 Our series has demonstrated a low requirement for reintervention due to food intolerance (one patient) and we believe that this is due to the respect we have for the angula incisura. We did not perform gastropexy sutures to control the final position of the sleeve staple line, and feel they are not required provided the angula is respected, and the stomach is not pulled or distorted during the stapling phase. Overall complication rates in our series also compare favourably13 and importantly no complication had led to any mortality. The single leak (0.2%) that we experienced presented as a non-communicating abscess adjacent to the proximal staple line. The majority of reported © 2013 Royal Australasian College of Surgeons
Review 500 laparoscopic sleeve gastrectomies
leaks from the staple line occur proximally. It is unclear whether this is a pressure, technical or ischaemic phenomenon. Significant controversy exists as to the best method of leak prevention. Musella et al.16 have shown that over sewing the staple line itself does not reduce staple line leaks but increases stenosis of the sleeve, whereas Ser et al. have demonstrated a significantly reduced leak rate (10% versus 0%) when the staple line was reinforced.16,17 Although patching the proximal staple line appears effective in protecting the proximal staple line, buttressing with Seamguard saves time and avoids the risk of hematoma albeit at slightly increased cost.18 We have used two techniques in this series to reinforce the proximal staple line: a simple fat pad/omental patch over the proximal corner, and commercial buttressed staple line reinforcement (Seamguard). While both techniques appear to be equally effective, we have evolved our technique to use Seamguard and omental patching, avoiding use of the fat pad after the only leak occurred. Additionally, in prevention, we believe respecting a 1 cm margin between the staple line and the lateral side of the oesophagus is important. The potential for chronicity of leaks after sleeve gastrectomy remains a key challenge for the procedure. We have learnt other lessons during our experience. Following our experience with a bougie perforation, our technique quite deliberately was to align the lesser curve to the oesophagus thereby facilitating passage. The splenic bleed we experienced was significantly because of the dense adhesion from a previous band. Given that, we also had three patients with post-operative haemorrhage, we have become quite deliberate in removing any tension from the short gastrics while the sealing device was activated. Post-operative anticoagulation may contribute but must be balanced against the risk of thromboembolism. While we did not experience any deep venous thromboses, one patient developed a portal vein thrombosis. Finally, reassuringly, port-site hernia was abolished after placing closure suture to the 15 mm lateral port. Unlike in some series, symptoms of GORD improved in our cohort.4,19 We feel this is due to our aggressive approach to the hiatus with routine assessment and repair of the hiatus for laxity or hernia (required in more than half of patients). This approach is facilitated by commencing the procedure at the hiatus thus clarifying whether hiatal repair is needed, identifying the precise position of the oesophagus and opening the angle of His to aid fundal mobilization. Limitations of our study primarily relate to its retrospective nature. Exemplifying this is that we have not controlled the study during changes in technique thereby preventing direct comparison. However, the large numbers and consistency of a single surgeon practice do make the results significant. In conclusion, the sleeve gastrectomy is a safe bariatric procedure with a low risk of complications. Tenants to successful sleeve gastrectomy include repair of hiatal laxity or hernia, generous width at the angula incisura, avoidance of twists and complete resection of the posterior fundus.
References 1. Omana JJ, Nguyen SQ, Herron D, Kini S. Comparison of comorbidity resolution and improvement between laparoscopic sleeve gastrectomy
© 2013 Royal Australasian College of Surgeons
677
2.
3.
4. 5.
6. 7.
8.
9.
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
and laparoscopic adjustable gastric banding. Surg. Endosc. 2010; 24: 2513–7. Bellanger DE, Greenway FL. Laparoscopic sleeve gastrectomy, 529 cases without a leak: short-term results and technical considerations. Obes. Surg. 2011; 21: 146–50. Gadiot RP, Biter LU, Zengerink HJ et al. Laparoscopic sleeve gastrectomy with an extensive posterior mobilization: technique and preliminary results. Obes. Surg. 2012; 22: 320–9. Himpens J, Dobbeleir J, Peeters G. Long-term results of laparoscopic sleeve gastrectomy for obesity. Ann. Surg. 2010; 252: 319–24. Brethauer SA, Hammel JP, Schauer PR. Systematic review of sleeve gastrectomy as staging and primary bariatric procedure. Surg. Obes. Relat. Dis. 2009; 5: 469–75. Aggarwal S, Kini SU, Herron DM. Laparoscopic sleeve gastrectomy for morbid obesity: a review. Surg. Obes. Relat. Dis. 2007; 3: 189–94. Lalor PF, Tucker ON, Szomstein S, Rosenthal RJ. Complications after laparoscopic sleeve gastrectomy. Surg. Obes. Relat. Dis. 2008; 4: 33– 8. Kelogrigoris M, Sotiropoulou E, Stathopoulos K, Georgiadou V, Philippousis P, Thanos L. CT-guided percutaneous drainage of infected collections due to gastric leak after sleeve gastrectomy for morbid obesity: initial experience. Cardiovasc. Intervent. Radiol. 2011; 34: 585–9. Stroh C, Birk D, Flade-Kuthe R et al. Results of sleeve gastrectomy-data from a nationwide survey on bariatric surgery in Germany. Obes. Surg. 2009; 19: 632–40. 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. Keidar A, Appelbaum L, Schweiger C, Elazary R, Baltasar A. Dilated upper sleeve can be associated with severe postoperative gastroesophageal dysmotility and reflux. Obes. Surg. 2010; 20: 140–7. Rosenthal RJ, International Sleeve Gastrectomy Expert P, Diaz AA et al. International sleeve gastrectomy expert panel consensus statement: best practice guidelines based on experience of >12 000 cases. Surg. Obes. Relat. Dis. 2012; 8: 8–19. Shi X, Karmali S, Sharma AM, Birch DW. A review of laparoscopic sleeve gastrectomy for morbid obesity. Obes. Surg 2010; 20: 1171–7. Bernstine H, Tzioni-Yehoshua R, Groshar D et al. Gastric emptying is not affected by sleeve gastrectomy–scintigraphic evaluation of gastric emptying after sleeve gastrectomy without removal of the gastric antrum. Obes. Surg. 2009; 19: 293–8. Shah S, Shah P, Todkar J, Gagner M, Sonar S, Solav S. Prospective controlled study of effect of laparoscopic sleeve gastrectomy on small bowel transit time and gastric emptying half-time in morbidly obese patients with type 2 diabetes mellitus. Surg. Obes. Relat. Dis. 2010; 6: 152–7. Musella M, Milone M, Bellini M, Leongito M, Guarino R, Milone F. Laparoscopic sleeve gastrectomy. Do we need to oversew the staple line? Ann. Ital. Chir. 2011; 82: 273–7. Ser KH, Lee WJ, Lee YC, Chen JC, Su YH, Chen SC. Experience in laparoscopic sleeve gastrectomy for morbidly obese Taiwanese: stapleline reinforcement is important for preventing leakage. Surg. Endosc. 2010; 24: 2253–9. Dapri G, Cadiere GB, Himpens J. Reinforcing the staple line during laparoscopic sleeve gastrectomy: prospective randomized clinical study comparing three different techniques. Obes. Surg. 2010; 20: 462–7. Chiu S, Birch DW, Shi X, Sharma AM, Karmali S. Effect of sleeve gastrectomy on gastroesophageal reflux disease: a systematic review. Surg. Obes. Relat. Dis. 2011; 7: 510–5.
