ORIGINAL ARTICLE. Quan-Xing Liu a,, Yuan Qiu b,, Xu-Feng Deng a, Jia-Xin Min a and Ji-Gang Dai a, * Abstract INTRODUCTION

Transcription

1 European Journal of Cardio-Thoracic Surgery 47 (2015) e118 e123 doi: /ejcts/ezu457 Advance Access publication 4 December 2014 ORIGINAL ARTICLE Cite this article as: Liu Q-X, Qiu Y, Deng X-F, Min J-X, Dai J-G. Comparison of outcomes following end-to-end hand-sewn and mechanical oesophagogastric anastomosis after oesophagectomy for carcinoma: a prospective randomized controlled trial. Eur J Cardiothorac Surg 2015;47:e a Comparison of outcomes following end-to-end hand-sewn and mechanical oesophagogastric anastomosis after oesophagectomy for carcinoma: a prospective randomized controlled trial Quan-Xing Liu a,, Yuan Qiu b,, Xu-Feng Deng a, Jia-Xin Min a and Ji-Gang Dai a, * Department of Thoracic Surgery, Xinqiao Hospital, Third Military Medical University, Chongqing, China b Department of General Surgery, Xinqiao Hospital, Third Military Medical University, Chongqing, China * Corresponding author. Department of Thoracic Surgery, Xinqiao Hospital, The Third Military Medical University, Chongqing , PR China. Tel: ; fax: ; @qq.com ( J. Dai). Received 1 August 2014; received in revised form 29 October 2014; accepted 4 November 2014 Abstract OBJECTIVES: The effects of the use of the stapler or hand-sewn method in oesophagogastric anastomosis on postoperative morbidity, mortality and quality of life after oesophagectomy remain controversial. The purpose of his study was to compare clinical outcomes of hand-sewn and stapler techniques in oesophagogastric anastomosis after oesophagectomy for oesophageal carcinoma. METHODS: We performed a prospective randomized controlled trial on 478 patients treated for oesophageal tumour between February 2009 and December Patients were randomly assigned to two treatment groups with 237 patients in the hand-sewn group and 241 patients in the circular stapler group ( ChiCTR-TRC ). RESULTS: The mean follow-up time was 18 months. The mean operating time of the stapled group and the hand-sewn group were 193 and 226 min, respectively (P < 0.001). Seventeen clinical and radiological leakages occurred in the hand-sewn group compared with 7 in the stapler group (P = 0.033). In the stapler group hospital mortality occurred in 10 patients (4.3%) and in the hand-sewn group in 9 patients (3.9%) (P = 0.837). Anastomotic strictures were noted in 31 patients from the stapler group (14.2%) and in 16 patients from the hand-sewn group (7.5%) (P = 0.027). CONCLUSIONS: Using the circular stapler method in oesophagogastric anastomoses had a lower anastomotic leakage rate and shorter operating time compared with the hand-sewn method. However, the circular stapler method was associated with a significantly increased risk of anastomotic strictures. Keywords: Anastomotic leakage Hand-sewn stapler Oesophagectomy INTRODUCTION Resection for carcinoma of the oesophagus with gastric transposition is currently considered as the standard surgical treatment for cure or palliation. Oesophagogastric anastomoses can be performed in the neck or the chest, by a hand-sewn method or by using a mechanical stapling device. It is reported that the survival rate of patients undergoing surgery for oesophageal cancer has improved in recent years, partly as a result of improvement in surgical techniques and perioperative management [1 4]. The patient s outcome is clearly correlated with the success of oesophagogastric anastomosis for early complications, such as anastomotic leakage, which can cause significant morbidity or mortality. Furthermore, anastomotic stricture as one of the main The first two authors contributed equally to this study. complications of surgery can contribute to the recurrence of dysphagia, with a need to restore normal swallowing function. The circular stapler is regarded as safer, less traumatic and less time-consuming, which has been one of the successful means adopted in an attempt to eliminate or at least reduce anastomotic complications. Many reports have shown that the stapler would not increase the rate of leakage after oesophagogastrostomy when compared with the hand-sewn method which calls for the development of the circular stapler [3, 4]; however, controversy still exists on which is the better approach. Obviously, in the recent randomized controlled trials, the stapling method appears to be superior to the hand-sewn method in preventing anastomotic leakage (1/41 vs 0/42 [5] and 2/58 vs 4/59 [3] and 1/14 vs 3/ 18 [4]), but there are no statistical differences because of limited samples. The present study was designed to clarify the controversy of using the hand-sewn method and circular stapler method in oesophagogastric anastomosis after oesophagectomy and gastric The Author Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 Q.-X. Liu et al. / European Journal of Cardio-Thoracic Surgery e119 tube reconstruction. The main end points of the study were anastomotic leakage, anastomotic stricture and perioperative morbidity and mortality. MATERIALS AND METHODS During the period of February 2009 and December 2011 we conducted a randomized trial in Department of Thoracic Surgery of Xinqiao Hospital, which was designed to compare the short- and medium-term results of the hand-sewn and stapler anastomosis after oesophagectomy with gastric tube reconstruction. This study was based on randomized control trials and the consolidated standards of reporting trials (CONSORT) guidelines of 2010 [6]. This study was registered at the Chinese Clinical Trial Registry (ChiCTR-TRC ). Our institute s ethics committee approved the study before randomization and all participants signed their written informed consent. The total number of patients to be included in this study was calculated based on the assumption that stapled oesophagogastric anastomosis would decrease the leak rate to <3% from an existing leak rate of 10% prevalent with the hand-sewn oesophagogastric anastomosis, thereby showing an absolute difference of 7% with regard to the occurrence of postoperative anastomotic leakage. Keeping in mind the aforementioned assumption, the study was designed with a power of 90% and alpha error of 0.05, and a sample size of 197 was required. Assuming that 5% of the patients were lost to the follow-up, we planned 210 cases eventually in each group. Randomization and masking All of the patients with oesophageal carcinoma were diagnosed and grouped according to the tumor, node, metastasis classification of oesophagus and oesophagogastric junctions in the seventh edition of the American Joint Committee on Cancer (AJCC) cancer staging manual, which was published by the AJCC and International Union Against Cancer [7]. The pathological examination of the endoscopic biopsy samples was performed for a definite diagnosis for all patients, but not all patients received a position emission tomography scan. All 478 patients who had oesophageal carcinoma were randomized by the closed-envelope method to have the end-to-end oesophagogastric anastomosis constructed by hand or by stapler after a Lewis Tanner oesophagectomy. The patients were randomly assigned to permuted blocks of 4 or 6 patients. Thus, 237 patients in the hand-sewn group and 241 patients in the stapler group were included in this study. Two oesophageal surgical teams in our department performed all the operations, respectively, and were not involved in the randomized classification process. Preoperative concurrent chemoradiotherapy (CCRT) with a standardized regimen (nedaplatin, paclitaxel and a total dose of 3600 cgy of radiation) was allowed. Inclusion and exclusion criteria Inclusion criteria for this study were as follows: (i) age above 18 years; (ii) biopsy-proven stage I, stage II or stage III untreated oesophageal or oesophagogastric junction cancer (including neoplasms of the upper, middle or lower third of the oesophagus). Exclusion criteria were as follows: (i) prior gastric surgery; (ii) patients who had advanced tumour stage, with distant metastasis (T4 disease and M1 disease); (iii) an increased cardiac risk (cardiac insufficiency grade IV The New York Heart Association or status post myocardial infarction) or poor pulmonary reserve (forced expiratory volume, <50% of normal); (iv) benign disease. The CONSORT flow diagram is showninfig.1. Surgical procedure Oesophagectomy with mediastinal lymphadenectomy was performed via open right thoracotomy. For patients with forced expiratory volume between 50 and 60% of normal (impaired pulmonary function), transhiatal blunt dissection was performed. The lesser curvature is resected through a midline laparotomy to form a gastric tube with a diameter of 3 cm. The tubular stomach was created by means of a 75-mm linear cutter stapler (Proximate Linear CutterTLC75, Johnson & Johnson Medical (China) Co. Ltd, Shanghai) [8]. Then careful preservation of the right gastroepiploic artery is ensured while pulling the gastric tube through the anterior mediastinum for end-to-end oesophagogastric anastomoses in all patients for ease of construction and standardization of the procedure. When the anastomosis was to be performed, before any further management, a randomization of the two groups was performed by patients chart number (randomly assigned on admission). In the hand-sewn group the oesophagogastric anastomosis was performed with a single layer of interrupted absorbable monofilament sutures. Thus, the muscle fibres and mucosa of the oesophagus and stomach were approximated with interrupted sutures. ILS circular staplers (Proximate ILS Stapler, Johnson & Johnson Medical (China) Co. Ltd, Shanghai) were used in the stapled anastomosis group. The operation was started with resection of the lesser curvature and the cardiac portion of the stomach and was completed with the anastomosis in the right side of the chest. The stapling device was introduced through a gastrotomy created in the region where the lesser curvature was resected. The gastrectomy was closed using the aforementioned liner stapler technique. The same technique of the stapler was used for both thoracic and transhiatal resections. The size of the oesophagus was stratified to be large ( 30 mm) or small (<30 mm) by measuring the outer diameter of the divided oesophagus after a Satinsky clamp was applied. According to our evaluation of the size of the oesophagus of our patients, different staplers were used in the patients in the stapler group of this study and the doughnuts were verified routinely. Clinical outcomes The main end points of this study were anastomotic leakage, benign stricture rates and perioperative morbidity and mortality. In addition, we also recorded the operative procedures and the operating time during surgery. After operation the assessment of anastomotic leakage for all patients was performed by the barium swallow examination or the endoscopic examination, or spill of saliva through the drains during 7 10 days. We added 5 ml of methylene blue via a nasogastric tube so that we could see blue saliva through the chest drain if anastomotic leakage had occurred. Other complications were also studied, THORACIC

3 e120 Q.-X. Liu et al. / European Journal of Cardio-Thoracic Surgery Figure 1: Consolidated standards of reporting trials flow chart. including cardiopulmonary morbidity, pulmonary complications, the time of stay in the intensive care unit (USA) and the hospital mortality. All patients were observed after discharge every 2 weeks for 2 months, monthly thereafter for 1 year and then at 3-month intervals until the present review period or the date of death. If symptoms of dysphagia returned, the barium swallow and endoscopic examinations were performed and the benign anastomotic stricture was diagnosed when the flexible endoscope with a 10-mm diameter could not be passed. Histological evidence was used to diagnose a malignant stricture. In the evaluation of benign stricture development, the following patients were excluded: (i) patients who died in hospital; (ii) patients who developed malignant recurrence; (iii) patients with a tumour infiltrated resection margin that was diagnosed on histopathological examination after surgery; (iv) patients who received postoperative radiation therapy. Statistical analysis The Fisher s exact test, χ 2 test and the Mann Whitney U-test were used for comparisons of different groups using the SPSS13.0 statistical analysis software (SPSS, Chicago, IL, USA). Fisher s exact test and the χ 2 test was used to compare categorical and dichotomous data, whereas the Mann Whitney U-test was typically used to compare ordinal data. Continuous data, such as hospital stay or age, were expressed as the mean and standard deviation and analysed by Student s t-test. P < 0.05 was considered statistically significant. RESULTS Between February 2009 and December 2011, this study included a total of 478 patients, with 237 in the hand-sewn group and 241

4 Q.-X. Liu et al. / European Journal of Cardio-Thoracic Surgery e121 Table 1: Demographic information of the two study groups Hand-sewn (n = 237) (%) Stapler (n = 241) (%) P-value Statistical method Age(years, mean ± SD) 61 ± 9 62 ± Student s t-test Sex χ 2 test Female 61 (25.7) 58 (24.1) Male 176 (74.3) 183 (75.9) Chronic illness a 76 (32.1) 91 (37.8) χ 2 test Site of tumour χ 2 test Upper thoracic 42 (17.7) 40 (16.6) Middle thoracic 138 (58.2) 145 (61.2) Lower thoracic 57 (24.1) 56 (23.2) Stage of tumour Mann Whitney U-test I 26 (11.0) 28 (11.6) IIa 65 (27.4) 70 (29.0) IIb 80 (33.8) 79 (32.8) III 66 (27.8) 64 (26.6) Concurrent chemoradiotherapy 29 (12.2) 35 (14.8) χ 2 test a Pulmonary, heart, renal diseases or diabetes. in the stapler group. The demographic data for the two groups are given in Table 1. There were no remarkable differences in age, sex, localization of the tumour, type of cancer, tumour stage and preoperative CCRT. In this study a total of 64 patients accepted the preoperative CCRT. There were 5 patients in the hand-sewn group and 6 patients in the stapler group who did not undergo surgery for refusal of subsequent surgery, or complications of CCRT. In the remaining 467 patients, radical resection of the carcinoma was performed through transhiatal blunt dissection (113 patients) with the oesophagogastric anastomosis performed on the cervical, or right thoracotomy (354 patients) with an oesophagogastric anastomosis in the chest. After oesophagectomy the oesophagogastric anastomoses of all patients were randomly performed by the hand-sewn technique (232 patients) or by the circular stapler method (235 patients) and no crossover operation was made between the two groups. A small-sized oesophagus was found in 144 patients of the handsewn group and 147 patients of the stapler group (P = 0.915). The types of stapler used in the stapler group stratified for size are listed in Table 2. The duration of operation, the operative procedures, proximal resection margins, operative complications, ICU and hospital stay, perioperative morbidity and leakage rates are summarized in Table 3. When compared with the stapler group, the patients in the hand-sewn group had longer operation time (P < 0.001). In 17 patients of the hand-sewn group and 7 patients of the stapler group, radiological and clinical anastomotic leakage developed. Transhiatal oesophagogastrectomy (THO) with cervical oesophagogastric anastomoses was performed in 54 patients of the handsewn group and 59 patients of the stapled group. However, the anastomotic leakage rates between the two subgroups have no significant difference (4 patients vs 2 patients, P = 0.423). After operation, all these patients were managed with intravenous nutrition and chest tube drainage. Twenty-one patients healed on conservative management. There were 19 patients who died within 30 days after operation, 9 in the hand-sewn group and 10 in the stapler group. Six of these patients in the hand-sewn group died of pneumonia, and the other 3 patients died of anastomotic Table 2: The types of stapler used in the stapler group stratified by size Stapler size (n = 235) Oesophageal size (n = 235) <30 mm 30 mm Large (ILS33) a 0 6 Medium (ILS29) Small (ILS25) Total a ILS (Ethicon, Inc., Somerville, NJ, USA). leakage. In the stapler group, 7 patients died of pneumonia, and the other 3 patients died of heart failure. There were 35 patients were excluded when evaluating the benign stricture for anastomotic recurrence development, hospital mortality and being performed postoperative radiotherapy to infiltrated proximal resection margin. There were, thus, 213 patients to be evaluated in the hand-sewn group and 219 patients in the stapler group. The mean follow-up time (standard deviation of the mean) of the patients was 17.8 (3.2) months for the handsewn group and 18.3 (3.4) months for the stapled group (P = 0.116). Data for anastomotic stricture rates are listed in Table 4. In the hand-sewn group, anastomotic stricture developed in 16 patients (7.5%) compared with 31 patients (14.2%) in the stapled group (P = 0.027). There were 5 patients (8.2%) in the hand-sewn group with cervical oesophagogastric anastomoses and 8 (14.8%) patients in the stapled group with cervical oesophagogastric anastomoses resulting in anastomotic stricture (P = 0.563). From operation to the development of benign stricture, the median duration was 3.6 months and 3.7 months in the hand-sewn group and the stapler group, respectively. All those strictures occurred within 1 year after operation. There also was no difference in the time to the development of stricture when the two groups were compared. THORACIC

5 e122 Q.-X. Liu et al. / European Journal of Cardio-Thoracic Surgery Table 3: Operative and preoperative data Hand-sewn (n = 232) (%) Stapler (n = 235)(%) P-value Statistical method Operative procedure χ 2 test TTO, n (%) 178 (76.7) 176 (74.9) THO, n (%) 54 (23.3) 59 (25.1) Operation time (min ± SD) 226 ± ± 16 <0.001 Student s t-test Proximal resection margins (cm ± SD) 8.5 ± ± Student s t-test Pulmonary complications a, n (%) 38 (16.4) 36 (15.3) χ 2 test Cardiac complications b, n (%) 12 (5.2) 15 (6.4) χ 2 test Abdominal or thoracic infections 9 (3.9) 8 (3.4) χ 2 test Anastomotic leaks, n (%) 17 (7.3) 7 (3.0) χ 2 test ICU stays (days, mean ± SD) 9.3 ± ± Student s t-test Hospital stay (days, mean ± SD) 18.9 ± ± Student s t-test a Aspiration pneumonia, pulmonary atelectasis and wound sepsis. b Arrhythmia, cardiac failure and myocardial infarction. TTO: transthoracic oesophagogastrectomy; THO: transhiatal oesophagogastrectomy. Table 4: Incidence of benign anastomotic strictures with regard to oesophageal size and stricture development time Hand-sewn (n = 213) (%) Stapler (n = 219) (%) P-value Statistical method Anastomotic strictures 16 (7.5) 31 (14.2) χ 2 test Time to development of stricture (months) 3.6 ± ± Student s t-test Oesophageal size χ 2 test <30 mm 144 (67.6) 147 (67.1) 30 mm 69 (32.4) 72 (32.9) The differences in stricture rates when the two techniques were compared were evident in both small- and large-size oesophagi. The effect of oesophageal size in each method did not cause significant difference in stricture rates. Stapler sizes were classified into three groups: large (ILS33), medium (ILS29) and small (ILS25). Notably, the hand-sewn method had significantly less chance of having stricture development compared with that of the small-size group (P = 0.001). Anastomotic recurrence occurred in 5 and 4 patients, respectively, in the hand-sewn and stapler groups. This represented 1.9% of patients who were discharged. DISCUSSION Many studies have described that oesophagogastric anastomoses can be performed by hand-sewn or circular stapling device. Currently, the use of mechanical staplers in operations on the digestive tract has been widely adopted in recent years, but in oesophageal surgery, the use of staplers for constructing an anastomosis is still a hotly debated method, and the indications are clearly far from being fully accepted. Recently, many prospective, randomized trials that compared hand-sewn and circular stapler methods have been published, but the results were not conclusive due to suturing methods, stapling instruments, heterogeneity in population, the location of anastomosis and the limited sample size [3, 9, 10]. In this prospective randomized trial we have shown that both hand-sewn and stapled techniques had low leakage rates in the construction of an intrathoracic and cervical oesophagogastric anastomosis, but the stapled technique is superior in preventing anastomosis leakage compared with the hand-sewn method. Although perioperative morbidity and mortality, middle-term anastomotic recurrence and perioperative complications were the same for both groups, the development of anastomotic stricture was much more prevalent with the stapled method. Anastomotic leakage after oesophageal resection for carcinoma is one of the most feared surgical complications after oesophagectomy because of its high morbidity and mortality [3, 11, 12]. Compared with other operations, oesophagogastrostomy anastomosis has a more problematic healing process, which was shown by the high complication rate and the poor healing quality [13]. Although local tissue and systemic factors are implicated, the main cause of an oesophageal anastomotic leakage is still considered multifactorial [14]. There are also the inherent properties of the oesophagus that are considered to increase the risk of leakage: (i) it has no serosa; (ii) the longitudinal muscles hold sutures poorly; (iii) the surgical exposure is often awkward. Surgical technique is thus likely to play a major role; using of the stapling device has lowered the incidence of leakage and has been advocated as the preferred method of anastomosis [15]. The result produced by a circular stapler is perhaps more uniform and less operatordependent; however, the result of the hand-sewn method is largely dependent on the experience of the surgeon to keep stabilization. In our prospective study comparing the two techniques, we confirmed beyond doubt that the leakage rates in the stapler

6 Q.-X. Liu et al. / European Journal of Cardio-Thoracic Surgery e123 group was lower than those of the hand-sewn group. However, the meta-analysis of randomized trials did not show a significant difference in anastomotic leakage for the two anastomotic methods [3, 16]. The reasons may be as follows: (i) In the handsewn group of our clinical trial, we performed the anastomosis using a single-layer interrupted suturing method rather than a layered or other modified hand-sewn technique. (ii) As the incidence of anastomotic leaks recently has been a low probability event (1%-10%) [17], the total sample size may have been inadequate to acquire a reliable result when performing a comparison of the two groups. Although clinical results of the present study showed that anastomotic leakage was reduced by using a stapled anastomosis method compared with the hand-sewn method, a two-fold increase in the incidence of anastomotic stricture was found with the stapled group. It is obvious that the anastomotic technique (whether hand-sewn or stapled) had the most significant bearing on stricture, and the reasons why stricture was more common with the stapler method may be as follows. First, the lack of accurate mucosa-to-mucosa apposition (with the edges separated by two thicknesses of bowel wall) may play an important role in causing the raw surface to heal by second intention with granulation tissue formation. Secondly, tissue necrosis beyond the staple line, inflammation and delayed epithelialization may also predispose to excessive fibrosis and stricture formation. Thirdly, the unabsorbed, circumferentially placed metal staples do not allow the lumen to dilate beyond the size obtained originally. In contrast, the hand-sewn method allows mucosa-to-mucosa apposition. When minimal tension is applied, an interrupted single-layer suture offers anastomosis with less risk of tissue strangulation than with the stapled method. All patients who had anastomotic stricture in our study were easily and safely treated by endoscopic dilatations. Between the two groups operative complications are also comparable. In this study, pulmonary and cardiac complications were found to be the two leading causes of in-hospital mortality. However, the technique of anastomosis would not be expected to have a great influence on pulmonary and cardiac complications after operation. The issues including chronic illness and route of reconstruction are considered as more important contributors to postoperative cardiac and pulmonary complications. In conclusion, although the circular stapler method increased the rates of benign stricture following oesophagogastric anastomosis after oesophagectomy for oesophageal cancer, the operating time in the stapler group was significantly shorter, and the leakage rate was notably reduced. Other surgical outcomes between the two groups were comparable. Owing to the benefits of the staplers with their relative ease of application and lower leakage rates, and being less operator-dependent, we conclude that the stapler method could be used as a superior method for oesophagogastric anastomosis after oesophagectomy for oesophageal carcinoma. ACKNOWLEDGEMENTS The authors thank the study participants for their contribution to the research, as well as current and past researchers and staff. We specifically thank Xiaobing Huang, Yuan Qiu and Wenyue Xu for their research and administrative assistance. Funding The work was supported by the clinical funding from the Xinqiao Hospital of the Third Military Medical University. Conflict of interest: none declared. REFERENCES [1] Ando N, Ozawa S, Kitagawa Y, Shinozawa Y, Kitajima M. Improvement in the results of surgical treatment of advanced squamous esophageal carcinoma during 15 consecutive years. Ann Surg 2000;232: [2] Abo S, Kitamura M, Hashimoto M, Izumi K, Minamiya Y, Shikama T et al. Analysis of results of surgery performed over a 20-year period on 500 patients with cancer of the thoracic esophagus. Surg Today 1996;26: [3] Luechakiettisak P, Kasetsunthorn S. Comparison of hand-sewn and stapled in esophagogastric anastomosis after esophageal cancer resection: a prospective randomized study. J Med Assoc Thai 2008;91: [4] Okuyama M, Motoyama S, Suzuki H, Saito R, Maruyama K, Ogawa J. Hand-sewn cervical anastomosis versus stapled intrathoracic anastomosis after esophagectomy for middle or lower thoracic esophageal cancer: a prospective randomized controlled study. Surg Today 2007;37: [5] Walther B, Johansson J, Johnsson F, Von Holstein CS, Zilling T. Cervical or thoracic anastomosis after esophageal resection and gastric tube reconstruction: a prospective randomized trial comparing sutured neck anastomosis with stapled intrathoracic anastomosis. Ann Surg 2003;238: , [6] Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomized trials. Ann Intern Med 2010;152: [7] Rice TW, Blackstone EH, Rusch VW. 7th edition of the AJCC Cancer Staging Manual: esophagus and esophagogastric junction. Ann Surg Oncol 2010;17: [8] Wang WP, Gao Q, Wang KN, Shi H, Chen LQ. A prospective randomized controlled trial of semi-mechanical versus hand-sewn or circular stapled esophagogastrostomy for prevention of anastomotic stricture. World J Surg 2013;37: [9] Urschel JD. Esophagogastrostomy anastomotic leaks complicating esophagectomy: a review. Am J Surg 1995;169: [10] Hsu HH, Chen JS, Huang PM, Lee JM, Lee YC. Comparison of manual and mechanical cervical esophagogastric anastomosis after esophageal resection for squamous cell carcinoma: a prospective randomized controlled trial. Eur J Cardiothorac Surg 2004;25: [11] Turkyilmaz A, Eroglu A, Aydin Y, Tekinbas C, Muharrem EM, Karaoglanoglu N. The management of esophagogastric anastomotic leak after esophagectomy for esophageal carcinoma. Dis Esophagus 2009;22: [12] Law S, Fok M, Chu KM, Wong J. Comparison of hand-sewn and stapled esophagogastric anastomosis after esophageal resection for cancer: a prospective randomized controlled trial. Ann Surg 1997;226: [13] Gockel I, Sultanov FS, Domeyer M, Trinh TT, Gonner U, Junginger T. [Surgical therapy for esophageal carcinoma: a prospective 20-year analysis]. Zentralbl Chir 2008;133: [14] Zhang YS, Gao BR, Wang HJ, Su YF, Yang YZ, Zhang JH et al. Comparison of anastomotic leakage and stricture formation following layered and stapler oesophagogastric anastomosis for cancer: a prospective randomized controlled trial. J Int Med Res 2010;38: [15] Hopkins RA, Alexander JC, Postlethwait RW. Stapled esophagogastric anastomosis. Am J Surg 1984;147: [16] Markar SR, Arya S, Karthikesalingam A, Hanna GB. Technical factors that affect anastomotic integrity following esophagectomy: systematic review and meta-analysis. Ann Surg Oncol 2013;20: [17] Honda M, Kuriyama A, Noma H, Nunobe S, Furukawa TA. Hand-sewn versus mechanical esophagogastric anastomosis after esophagectomy: a systematic review and meta-analysis. Ann Surg 2013;257: THORACIC