Stapled versus handsewn methods for colorectal anastomosis surgery

Background Previous systematic reviews comparing stapled and handsewn colorectal anastomosis that are available in the medical literature have not shown either technique to be superior. An update of this systematic review was performed to find out if there are any data that properly answer this question.

To compare the safety and effectiveness of stapled and handsewn colorectal anastomosis surgery. The following primary hypothesis was tested: the stapled technique is more effective because it decreases the level of complications. Search methods A computerized search was performed in the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE according to the strategies of the Colorectal Cancer Group of The Cochrane Collaboration. There were no limits upon language, date or other criteria. A revised search strategy was performed for this updated version of the review May 2011. Selection criteria All randomised controlled trials (RCTs) in which stapled and handsewn colorectal anastomosis techniques were compared. Participants were adult patients undergoing elective colorectal anastomosis surgery. The interventions were endoluminal circular stapler and handsewn colorectal anastomosis surgery. Outcomes considered were a) mortality; b) overall anastomotic dehiscence; c) clinical anastomotic dehiscence; d) radiological anastomotic dehiscence; e) stricture; f) anastomotic haemorrhage; g) reoperation; h) wound infection; i) anastomosis duration; and j) hospital stay. Data collection and analysis Data were independently analysed by the two review authors (CBN, SASL) and cross‐checked. The methodological quality of each trial was assessed by the same two authors. After searching the literature for this update, no study was added to those in the previous version of this review. Details of randomizations (generation and concealment), blinding, whether an intention‐to‐treat analysis was done or not, and the number of patients lost to follow‐up were recorded. The analysis of the risk of bias was updated according to the software Review Manager 5.1. The results of each RCT were summarized on an intention‐to‐treat basis in 2 x 2 tables for each outcome. External validity was defined by the characteristics of the participants, interventions and the outcomes. The RCTs were stratified according to the level of colorectal anastomosis. The risk difference (RD) method (random‐effects model) and number needed to treat (NNT) for dichotomous outcome measures and weighted mean differences (WMD) for continuous outcomes measures, with the corresponding 95% confidence intervals (CI), were presented in this review. Statistical heterogeneity was evaluated using a funnel plot and the Chi2 test. Main results Of the 1233 patients enrolled in nine identified trials, 622 were treated with staples and 611 with manual suture. The following main results were obtained. a) Mortality, result based on 901 patients: RD ‐0.6%, 95% CI ‐2.8% to +1.6%. b) Overall dehiscence, result based on 1233 patients: RD 0.2%, 95% CI ‐5.0% to +5.3%. c) Clinical anastomotic dehiscence, result based on 1233 patients: RD ‐1.4%, 95% CI ‐5.2 to +2.3%. d) Radiological anastomotic dehiscence, result based on 825 patients: RD 1.2%, 95% CI ‐4.8% to +7.3%. e) Stricture, result based on 1042 patients: RD 4.6%, 95% CI 1.2% to 8.1%; NNT 17, 95% CI 12 to 31. f) Anastomotic haemorrhage, result based on 662 patients: RD 2.7%, 95% CI ‐0.1% to +5.5%. g) Reoperation, result based on 544 patients: RD 3.9%, 95% CI 0.3% to 7.4%. h) Wound infection, result based on 567 patients: RD 1.0%, 95% CI ‐2.2% to +4.3%. i) Anastomosis duration, result based on one study (159 patients): WMD ‐7.6 minutes, 95% CI ‐12.9 to ‐2.2 minutes. j) Hospital stay, result based on one study (159 patients): WMD 2.0 days, 95% CI ‐3.27 to +7.2 days. Authors' conclusions The evidence found was insufficient to demonstrate any superiority of stapled over handsewn techniques in colorectal anastomosis surgery, regardless of the level of anastomosis. There were no randomised clinical trials comparing these two types of anastomosis in elective conditions in the last decade. The relevance of this research question has possibly lost its strength where elective surgery is concerned. However, in risk situations, such as emergency surgery, trauma and inflammatory bowel disease, new clinical trials are needed.

Adult; Humans; Suture Techniques; Anastomosis, Surgical; Anastomosis, Surgical/methods; Colon; Colon/surgery; Randomized Controlled Trials as Topic; Rectum; Rectum/surgery; Surgical Stapling Plain language summary The evidence found did not indicate superiority of stapled over handsewn technique in colorectal anastomosis, regardless of the anastomotic level. The review with nine randomised controlled trials (1233 patients, 622 with stapling and 611 with the handsewing technique) compared the safety and effectiveness of stapled versus handsewn colorectal anastomosis surgery. Meta‐analysis was performed using the risk difference and weighted mean difference, with corresponding 95% confidence intervals. Outcome measures were mortality, anastomotic dehiscence, narrowing (stricture), haemorrhage, need for reoperation, wound infection, anastomosis duration (time taken to perform the anastomosis) and hospital stay. No significant statistical differences were found except that stricture was more frequent with stapling (P < 0.05) and the time taken to perform the anastomosis was longer with handsewn techniques.

Since stapling instruments have been introduced into the current practice of surgery of the digestive tract, by Ravitch and Steichen (Ravitch 1979), the safety, usability and cost effectiveness of stapling has been questioned. In colorectal surgery, the advantages of the stapled technique are said to be a lower percentage of complications, such as leaks, better blood supply, reduced tissue manipulation, less edema, uniformity of sutures and shorter hospital stay and operation time (Korolija 2008). A smaller number of leaks might be a factor in the lower incidence of both tumour recurrence and cancer‐specific mortality (Docherty 1995). The medical literature concerning the comparison of stapling and handsewn techniques is conflicting. The use of staplers for intra‐peritoneal anastomosis has been questioned (Beart 1991). Systematic reviews have shown that both techniques are effective and the choice of technique may be based on personal preference (MacRae 1998) and local availability. Prospective and randomised trials have shown different results, shown by no significant inter‐group difference found in regard to time for anastomosis construction or occurrence of complications in colorectal anastomosis (Cajozzo 1990); the routine use of stapling instruments for intra‐peritoneal colorectal anastomosis could not be recommended because of a higher incidence of mishaps and strictures, even though the operation took less time to perform and anastomotic leakage occurred less often (Fingerhut 1994). Taking into consideration the conflicting data available so far, an update of this systematic review comparing handsewn and stapled colorectal anastomosis surgery is needed.

To compare the relative safety and effectiveness of stapled anastomosis with that of handsewn colorectal anastomosis surgery. The following primary hypothesis was tested: the stapled technique is more effective because it decreases the level of complications.

Criteria for considering studies for this review Types of studies Randomized controlled trials. Types of participants Adults submitted electively to stapled and handsewn colorectal anastomosis. There was no stratification according to underlying disease. Types of interventions Endoluminal circular stapler using the models end‐to‐end anastomosis (EEA), disposable end‐to‐end anastomosis (DEEA) and intraluminal anastomosis (ILS); no restriction on the material and technique used for handsewn anastomosis surgery. Types of outcome measures a) Specific mortality: number of in‐hospital deaths due to anastomotic complications. b) Overall anastomotic dehiscence: total number of anastomotic dehiscences. c) Clinical anastomotic dehiscence: clinical dehiscence defined as a discharge of faeces from the anastomosis site. d) Radiological anastomotic dehiscence: presence of a leak with the control postoperative enema in a patient who had no evidence of clinical anastomotic leak. e) Stricture: narrowing in the bowel lumen due to anastomotic healing. f) Anastomotic haemorrhage: postoperative bleeding per anus from the anastomotic site. g) Reoperation: surgical re‐intervention for anastomotic complication. h) Wound infection: defined as the presence of infection in the abdominal wound. i) Anastomosis duration: time to perform the anastomosis. j) Hospital stay: time from operation until discharge from the hospital. Search methods for identification of studies All publications describing (or which might describe) stapled versus handsewn colorectal anastomosis were sought through computerised searches of the Cochrane Central Register of Controlled Trials (CENTRAL) (May 2011), MEDLINE (1980 to 2011), EMBASE (1980 to 2011). For the search terms used please see Appendix 1 (MEDLINE), Appendix 2 (EMBASE) and Appendix 3 (CENTRAL). Data collection and analysis Locating and selecting studies Two review authors (CBN and SASL) independently selected the trials which were included in the review. Previous systematic reviews have been reviewed. Critical appraisal of studies The methodological quality of each trial was assessed by the same two review authors. Details of the randomizations method, blinding, whether an intention‐to‐treat analysis was performed, and the number of patients lost to follow‐up were recorded. The risk of bias was stratified by CBN and EMKS according to the Cochrane Collaboration criteria (Higgins 2011). Collecting data Data were independently extracted by the review authors and cross‐checked. The results of each trial were summarized on an intention‐to‐treat basis in 2 x 2 tables for each outcome. Only trials classified as randomised score A and B have been used in the review. In the analysis, trials of each type of suture (stapled or handsewn) and anatomical anastomotic site (extra or intraperitoneal) were stratified to assess whether there may be important differences between them. The external validity of the studies has been assessed by analysis of the following. Participants: category of disease (cancer, inflammatory disease, non‐inflammatory disease), male and female, location and duration of study. The calculation of the sample size and the sample representativeness was also analysed. Interventions: endoluminal circular stapler, models end‐to‐end anastomosis (EEA), disposable end‐to‐end anastomosis (DEEA) and intra‐luminal anastomosis (ILS); no restriction was made on materials and techniques used for handsewn anastomosis surgery. Outcomes: a) mortality; b) overall anastomotic dehiscence; c) clinical anastomotic dehiscence; d) radiological anastomotic dehiscence; e) stricture; f) anastomotic haemorrhage; g) reoperation; h) wound infection; i) anastomosis duration; j) hospital stay. Analysing and presenting results The results of each RCT were summarized on an intention‐to‐treat basis in 2 x 2 tables for each outcome. External validity was defined by the characteristics of the participants, the interventions and the outcomes. The RCTs have been stratified according to the level of colorectal anastomosis. The risk difference (RD) method (random‐effects model) and number needed to treat (NNT) for dichotomous outcome measures and weighted mean difference (WMD) for continuous outcomes measures, with the corresponding 95% confidence intervals (CI), have been presented in this review. Meta‐analysis was performed using RevMan 2011. Statistical heterogeneity in the results of the trials were assessed both by inspection of a graphical presentation (funnel plot: plotting the study weight or sample size [on the 'y' axis] against the odds ratio [on the 'x' axis]) and by calculating a test of heterogeneity (standard Chi2 test on N degrees of freedom, where N equals the number of trials contributing data minus one). Three possible reasons for heterogeneity were pre‐specified: (i) that responses differed according to a difference in the quality of the trials; (ii) that responses differed according to sample size; (iii) that responses differed according to clinical heterogeneity. These have been assessed by looking at separate subgroups of trials. Clinical heterogeneity was assessed by the authors of this review (CBN, SASL and DM).

Description of studies In summary, nine studies were selected for this systematic review. They were published in English (six trials), French (two trials) and German (one trial). The aim of these studies was to compare stapled anastomosis performed with EEA, DEEA and ILS instruments (622 patients) versus handsewn anastomosis (611 patients); altogether (1231patients) were submitted to elective operations and two patients were submitted to emergency operations. The outcome measures analysed in the studies were: a) specific mortality (7 studies); b) overall anastomotic dehiscence (6 studies); c) clinical anastomotic dehiscence (9 studies); d) radiological anastomotic dehiscence (6 studies); e) stricture (7 studies); f) anastomotic haemorrhage (4 studies); g) reoperation (3 studies); h) wound infection (6 studies); i) anastomosis duration (1 study); j) hospital stay (1 study). Risk of bias in included studies The process of generation of random numbers was reported in six studies and concealment of allocation was kept until the time of the anastomotic procedure (the surgeon opened the envelope when he found out that both techniques were feasible) in seven studies; in two studies the allocation concealment was not reported. Blinding was not described in any study. The follow up of the patients was confined to the period within hospital in three studies, for six to 10 months in three studies, for more than a year in two studies, and was not described in one study. Intention‐to‐treat analysis was applied in three studies for all outcome measures; in four studies the stricture anastomotic variable was not described using intention‐to‐treat analysis; in one study this principle was not used at all; in one study local recurrence was not evaluated according to the intention‐to‐treat principle. A sample size calculation was described in three studies. The representativeness of the patients was reported in four studies. The learning curve for surgeons to perform stapled colorectal anastomosis was reported in four studies. Dropouts and withdrawals were reported for all outcome measures in one study, in three studies for stricture only, and were not described in four studies. From the data available, the risk of bias of each study was stratified according to the risk of bias classification of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Effects of interventions Stapled versus handsewn See comparison 01.00.00 Of the 1233 patients enrolled (09 trials), 622 were treated with stapled and 611 with handsewn anastomoses. The results of the comparison between patients assigned to stapled anastomosis with the patients assigned to the handsewn procedure are as follows. a) Specific mortality (as a dichotomous outcome): this result was based on seven studies (901 patients); 2.4% (11 of 453 patients) versus 3.6% (16 of 448 patients); risk difference (random‐effects model) ‐0.6% (95% CI ‐2.8% to 1.6%). b) Overall dehiscence (as a dichotomous outcome): this result was based on nine studies (1233 patients); 13.0% (81 of 622 patients) versus 13.4% (82 of 611) patients; risk difference (random‐effects model) 0.2% (95% CI ‐5.0% to 5.3%). c) Clinical anastomotic dehiscence (as a dichotomous outcome): this result was based on nine studies (1233 patients); 6.3% (39 of 616 patients) versus 7.1% (44 of 617 patients); risk difference (random‐effects model) ‐1.4% (95% CI ‐5.2% to 2.3%). d) Radiological anastomotic dehiscence (as a dichotomous outcome): this result was based on six studies (825 patients); 7.8% (33 of 421patients) versus 7.2% (30 of 414 patients); risk difference (random‐effects model) 1.2% (95% CI ‐4.8% to 7.3%). e) Stricture (as a dichotomous outcome): this result was based on seven studies (1042 patients); 8% (40 of 500 patients) versus 2% (10 of 496 patients); risk difference (random‐effects model) 4.6%, (95% CI 1.2% to 8.1%); NNT 17 (95% CI 12 to 31). f) Anastomotic haemorrhage (as a dichotomous outcome): this result was based on four studies (662 patients); 5.4% (18 of 336 patients) versus 3.1% (10 of 326 patients); risk difference (random‐effects model) 2.7% (95% CI ‐0.1% to 5.5%). g) Reoperation (as a dichotomous outcome ): this result was based on three studies (544 patients); 7.6% (21 of 278 patients) versus 4.1% (11 of 266 patients); risk difference (random‐effects model) 3.9 % (95% CI 0.3% to 7.4%). h) Wound infection (as a dichotomous outcome): this result was based on six studies (567 patients), 5.9% (17 of 286) versus 4.3% (12 of 282); risk difference (random‐effects model) 1.0% (95% CI ‐2.2% to 4.3%). i) Anastomosis duration, time to complete the anastomosis (as a continuous outcome): this result was based on one study (159 patients); average value of ‐7.6 minutes (95% CI ‐12.9 to ‐2.2 minutes). j) Hospital stay (as a continuous outcome): this result was based on one study (159 patients), average value of 2.0 days (95% CI ‐3.2 to 7.2 days). The results were stratified according to level of anastomosis (intra versus extra‐peritoneal) and are shown in the graphs.

When we performed a search on the available literature comparing manual and stapled suture techniques in colorectal elective surgery in order to update this systematic review we asked ourselves if this was a relevant question and do we have consolidated scientific evidence to sustain our current surgical routine? The basic premise of this systematic review was that by grouping studies without statistical power but with adequate methodological quality, a sample large enough to detect any possible significant differences could be obtained. This was not observed, given that the majority (7/9) of the variables analysed were not significantly different. Perhaps this fact can be explained by insufficient overall sample sizes for demonstrating the magnitude of possible differences. This reinforces the necessity for periodic updating of this review. It is possible that the results may in some way have been influenced by aspects of a learning curve related to differences in experience between surgeons participating in the included studies. In addition to this learning curve, another factor related to the results from colorectal anastomoses is the adequate functioning of the instrument used. In this systematic review, some authors (Beart 1991; Fingerhut 1994; McGinn 1985) analysed this aspect together with the experience of the surgeon. It is accepted that these two parameters, failure of instruments and experience of surgeons, should be analysed independently (Matos 1996). Colorectal anastomoses have their healing progressively compromised the closer they are to the anal margin. In a prospective study of handsewn colorectal anastomosis in 370 patients, an incidence of fistula in supra‐peritoneal anastomoses of 0.6% was seen, and of 7.0% in infra‐peritoneal anastomoses (Mann 1996). In a retrospective study of 219 patients submitted to radical anterior resection because of adenocarcinoma of the rectum, there was an incidence of 11%, all in anastomoses situated within 6 cm of the anal margin. The risk of complications is greater for infra‐peritoneal, more distal and colorectal anastomoses (Karanjia 1994). The classification criteria for colorectal anastomoses in relation to position were not uniform in the literature surveyed in this review. Thus, some authors refer to endoscopic measurements for classifying anastomoses (Beart 1991; Fingerhut 1994; Fingerhut 1995; McGinn 1985) while others classify anastomoses as high or low without referring to the criteria adopted (Kracht 1991). In fact, the heterogeneity of measures may lead to misinterpretation of some results in this review. Various procedures may alter the security of a colorectal anastomosis. Protection colostomy (Mealy 1992), epiploplasty, complementary suturing and the performance or otherwise of an integrity test on the anastomosis (Beart 1991) are procedures frequently used by surgeons, but not in a uniform manner. In this review, the great majority of authors (8/9) used such procedures and this may have influenced the conclusions. Future trials and analyses should stratify for these variables. The data relating to specific mortality show that clinical dehiscence was responsible for four deaths in the group sutured by stapling, and for two deaths in the handsewn group. The rate of anastomotic dehiscence, evaluated clinically and radiologically, was not significantly different between the two techniques of stapling and handsewing. In various non‐randomised studies, a greater incidence of stenosis has been attributed to the technique of stapling (Docherty 1995; Fain 1975; Heald 1981; Ravitch 1979; Shahinian 1980). In the present systematic review it was observed that the length of follow‐up for assessing stenosis varied a lot between the different studies, which made comparison excessively difficult and may have made the overall result imprecise. Actuarial calculations, instead of crude ones, may be more appropriate in this context. This systematic review has shown that stenosis occurs to a significant extent in patients undergoing stapled colorectal anastomosis, especially in infra‐peritoneal anastomoses. This may be considered scientifically relevant in this review. However, the majority of the studies (7/9) considered this complication to be irrelevant from a clinical point of view, given the favourable outcome with conservative treatment in the great majority of cases. No patient in any series had to be re‐operated on for anastomotic stenosis. Besides, the data favouring handsewn anastomosis regarding this outcome (stricture) in this review are the result of a single study for infra‐peritoneal anastomosis, and so conclusions may be tendentious. The time taken to perform the anastomosis was significantly shorter in colorectal anastomoses performed with the stapler than for handsewn anastomoses. A limitation to the analysis of this variable was that only one study (Fingerhut 1995) provided data (average and standard deviation) that could be used for statistical analysis, just as we have seen in the stricture analysis. The time taken to perform the anastomosis may, when analysed in isolation, have some importance. It may influence the total length of the operative procedure or hospitalisation of the patient. The other variables analysed did not demonstrate any advantage of one technique over the other. The question of cost, which was not analysed in this review, is related to the length of the operative procedure, length of hospitalisation, price of sutures and the value of devices used, among other factors. This represents a variable of great importance, deserving special attention in studies with that specific objective. We also consider that a more detailed study of costs in this review would become necessary in the event of evidence that the stapling technique was more advantageous. When only the cost of the material used in the anastomosis is taken into consideration, the stapler is more expensive. The cost of an operative procedure, however, must be analysed within a wider context involving not only the monetary value of the materials but also the value resulting from the ease of execution, total time consumed, cost of complications related to the method employed among other factors. In the literature review we found some studies comparing the stapling technique and manual suture in less ordinary conditions, such as trauma and other emergency situations. One of them was a randomised clinical trial (Catena 2004) and another was a prospective analysis (Demetriades 2002) and both advocate that the method of anastomosis does not affect the incidence of abdominal complications. More randomised clinical data are needed to support these results. In the same way, a few clinical trials and non‐randomised studies speculate that stapled anastomosis may have better outcomes in Crohn disease than manual suture (Resegotti 2005), and there are controversial results concerning the oncological outcome of the stapled anastomosis (Docherty 1995; Lovegrove 2006). These points remain unclear in this updated review of the available data. The practical conclusion, from this systematic review, is that the evidence found was insufficient to demonstrate superiority of the stapler to handsewn anastomosis, independent of the level of anastomosis. The fact that few studies were found comparing those two techniques for elective colorectal anastomosis means, from our point of view, that the decision about which technique to use depends on clinical circumstances and available resources and is not currently decided for the coloproctogical surgeon. However, we think this is a still a very relevant question when we concern risk situations, such as trauma, inflammatory bowel disease, obstruction, etc. and a number of randomised trials exist to help answer this question. Authors' conclusions Implications for practice. The evidence found was insufficient to demonstrate any superiority of stapled over handsewn techniques in colorectal anastomosis surgery, regardless of the level of anastomosis. The choice of which technique is applied depends on the surgeon's personal preferences, clinical circumstances and available resources. There are no randomised clinical trials comparing these two types of anastomosis in elective surgery in the last decade. The relevance of this research question has possibly lost its strength where elective surgery is concerned. However, in risk situations, such as emergency surgery, trauma and inflammatory bowel disease, new clinical trials are needed. Implications for research. From the results of this systematic review, we propose that the design of future trials should include the following. a) Objective: to compare the efficiency of handsewn and stapled supra and infra‐peritoneal colorectal anastomosis separately. b) Design: Multicentre randomised controlled trial with adequate follow‐up and analysis of anastomotic strictures. The intention‐to‐treat principle should be used with a full description of dropouts and withdrawals. d) Participants: consecutive elective patients. e) Intervention: anastomotic devices should be standardized. Surgical interventions performed during the learning curve period should not be included. f) Control group: use standardized techniques of handsewn colorectal anastomosis. g) Factors influencing safety of anastomosis: take into account, and stratify for, ancillary procedures such as diverting colostomy, air‐tightness test, oversewn and omental wrap as major factors influencing the safety of anastomosis. h) Outcome measures: mortality, overall anastomotic dehiscence, clinical anastomotic dehiscence, radiological anastomotic dehiscence, stricture, anastomotic haemorrhage, reoperation, anastomosis duration and hospital stay. i) Data analysis: with a sample size of 578 patients in each group, there will be 90% statistical power to detect a 50% reduction (0,1 to 0,05) in the clinical anastomotic dehiscence, with 95% confidence interval. j) Compare the efficiency of handsewn and stapled colorectal anastomosis surgery in risk situations (emergency procedures for trauma, active inflammatory bowel disease, obstruction and others). What's new | Date | Event | Description | |---|---|---| | 26 October 2011 | New search has been performed | New search 5 May 2011. | | 26 October 2011 | New citation required but conclusions have not changed | No new studies included, but 14 eligible studies, identified and commented on, were excluded in this update. | History Protocol first published: Issue 4, 1998 Review first published: Issue 3, 2001 | Date | Event | Description | |---|---|---| | 25 May 2001 | New citation required and conclusions have changed | Substantive amendment |

We would like to acknowledge Marija Barbateskovic for performing the search strategy. Appendices Appendix 1. MEDLINE search strategy MEDLINE (Ovid) 30.05.11 – 152 hits Search history 1. exp Colorectal Surgery/ 2. exp Intestines/ 3. (colon* or rect* or colorect* or intestin* or bowel).mp. 4. 1 or 2 or 3 5. exp Anastomosis, Surgical/ 6. anastomos*.mp. 7. 5 or 6 8. exp Surgical Stapling/ 9. exp Surgical Staplers/ 10. stapl*.mp. 11. 8 or 9 or 10 12. exp Sutures/ 13. exp Suture Techniques/ 14. (sutur* or sew* or handsew* or stitch).mp. 15. 12 or 13 or 14 16. 4 and 7 and 11 and 15 17. randomized controlled trial.pt. 18. controlled clinical trial.pt. 19. randomized.ab. 20. placebo.ab. 21. clinical trial.sh. 22. randomly.ab. 23. trial.ti. 24. 17 or 18 or 19 or 20 or 21 or 22 or 23 25. humans.sh. 26. 24 and 25 27. 16 and 26 Appendix 2. EMBASE search strategy EMBASE (Ovid) 30.05.11 – 354 hits Search history 1. exp colorectal surgery/ 2. exp intestine/ 3. (colon* or rect* or colorect* or intestin* or bowel).mp. 4. 1 or 2 or 3 5. exp anastomosis/ 6. anastomos*.mp. 7. 5 or 6 8. exp surgical stapling/ 9. exp surgical stapler/ 10. stapl*.mp. 11. 8 or 9 or 10 12. exp suture/ 13. exp suturing method/ 14. (sutur* or sew* or handsew* or stitch).mp. 15. 12 or 13 or 14 16. 4 and 7 and 11 and 15 17. randomized controlled trial/ 18. randomization/ 19. controlled study/ 20. multicenter study/ 21. phase 3 clinical trial/ 22. phase 4 clinical trial/ 23. double blind procedure/ 24. single blind procedure/ 25. ((singl* or doubl* or trebl* or tripl*) adj (blind* or mask*)).ti,ab. 26. (random* or cross* over* or factorial* or placebo* or volunteer*).ti,ab. 27. 22 or 19 or 23 or 25 or 18 or 24 or 20 or 17 or 26 or 21 28. "human*".ti,ab. 29. (animal* or nonhuman*).ti,ab. 30. 29 and 28 31. 29 not 30 32. 27 not 31 33. 16 and 32 Appendix 3. CENTRAL search strategy The Cochrane Library30.05.11 – 98 hits (87 hits in CENTRAL) #1 MeSH descriptor Colorectal Surgery explode all trees #2 MeSH descriptor Intestines explode all trees #3 (colon* or rect* or colorect* or intestin* or bowel):ti,ab,kw #4 (#1 OR #2 OR #3) #5 MeSH descriptor Anastomosis, Surgical explode all trees #6 (anastomos*):ti,ab,kw #7 (#5 OR #6) #8 MeSH descriptor Surgical Stapling explode all trees #9 MeSH descriptor Surgical Staplers explode all trees #10 (stapl*):ti,ab,kw #11 (#8 OR #9 OR #10) #12 MeSH descriptor Sutures explode all trees #13 MeSH descriptor Suture Techniques explode all trees #14 (sutur* or sew* or handsew* or stitch):ti,ab,kw #15 (#12 OR #13 OR #14) #16 (#4 AND #7 AND #11 AND #15) Data and analyses Comparison 1. All studies. | Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size | |---|---|---|---|---| | 1 mortality | 7 | 901 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.69 [0.32, 1.49] | | 1.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.64 [0.16, 2.67] | | 1.2 SUPRAPERITONEAL ANASTOMOSIS | 2 | 229 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.78 [0.18, 17.29] | | 1.3 COLORECTAL ANASTOMOSIS | 3 | 441 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.60 [0.22, 1.61] | | 2 overall dehiscence | 9 | 1233 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.99 [0.71, 1.40] | | 2.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.88 [0.97, 3.63] | | 2.2 SUPRAPERITONEAL ANASTOMOSIS | 3 | 289 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.03 [0.42, 2.53] | | 2.3 COLORECTAL ANASTOMOSIS | 4 | 713 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.74 [0.47, 1.15] | | 3 clinical dehiscence | 9 | 1233 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.80 [0.51, 1.24] | | 3.1 INFRAPERITONEAL ANASTOMOSIS | 3 | 344 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.74 [0.36, 1.55] | | 3.2 SUPRAPERITONEAL ANASTOMOSIS | 4 | 448 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.97 [0.34, 2.74] | | 3.3 COLORECTAL ANASTOMOSIS | 3 | 441 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.79 [0.41, 1.51] | | 4 radiological dehiscence | 6 | 835 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.10 [0.66, 1.85] | | 4.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 2.04 [0.92, 4.50] | | 4.2 SUPRAPERITONEAL ANASTOMOSIS | 2 | 219 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.34 [0.50, 3.61] | | 4.3 COLORECTAL ANASTOMOSIS | 2 | 385 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.38 [0.15, 0.98] | | 5 stricture | 7 | 996 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 3.59 [2.02, 6.35] | | 5.1 INFRAPERITONEAL ANASTOMOSIS | 1 | 102 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 3.90 [1.06, 14.30] | | 5.2 SUPRAPERITONEAL ANASTOMOSIS | 2 | 214 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.75 [0.34, 8.92] | | 5.3 COLORECTAL ANASTOMOSIS | 4 | 680 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 3.99 [2.00, 7.96] | | 6 hemorrhage | 4 | 662 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.78 [0.84, 3.81] | | 6.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.86 [0.22, 3.31] | | 6.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 6.82 [1.15, 40.41] | | 6.3 COLORECTAL ANASTOMOSIS | 1 | 272 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.74 [0.59, 5.09] | | 7 reoperation | 3 | 544 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.94 [0.95, 3.98] | | 7.1 INFRAPERITONEAL ANASTOMOSIS | 1 | 113 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.60 [0.48, 5.27] | | 7.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 6.73 [0.93, 48.93] | | 7.3 COLORECTAL ANASTOMOSIS | 1 | 272 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.62 [0.59, 4.44] | | 8 wound infection | 6 | 568 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.43 [0.67, 3.04] | | 8.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 2.43 [0.68, 8.68] | | 8.2 SUPRAPERITONEAL ANASTOMOSIS | 3 | 289 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.08 [0.40, 2.92] | | 8.3 COLORECTAL ANASTOMOSIS | 1 | 48 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.0 [0.06, 16.47] | | 9 anastomosis duration | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | ‐7.60 [‐12.92, ‐2.28] | | 9.1 INFRAPERITONEAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | | 9.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | ‐7.60 [‐12.92, ‐2.28] | | 9.3 COLORECTAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | | 10 hospital stay | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | 2.0 [‐3.27, 7.27] | | 10.1 INFRAPERITONEAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | | 10.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | 2.0 [‐3.27, 7.27] | | 10.3 COLORECTAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | Comparison 2. Studies with adequate allocation concealment. | Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size | |---|---|---|---|---| | 1 mortality | 6 | 788 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.50 [0.21, 1.20] | | 1.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.64 [0.16, 2.67] | | 1.2 SUPRAPERITONEAL ANASTOMOSIS | 2 | 229 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.78 [0.18, 17.29] | | 1.3 COLORECTAL ANASTOMOSIS | 2 | 328 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.28 [0.08, 0.99] | | 2 overall dehiscence | 7 | 1060 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.99 [0.68, 1.43] | | 2.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.88 [0.97, 3.63] | | 2.2 SUPRAPERITONEAL ANASTOMOSIS | 2 | 229 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.26 [0.39, 4.03] | | 2.3 COLORECTAL ANASTOMOSIS | 3 | 600 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.66 [0.41, 1.08] | | 3 clinical dehiscence | 6 | 1051 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.71 [0.44, 1.14] | | 3.1 INFRAPERITONEAL ANASTOMOSIS | 3 | 503 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.56 [0.32, 0.97] | | 3.2 SUPRAPERITONEAL ANASTOMOSIS | 3 | 501 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.47 [0.59, 3.62] | | 3.3 COLORECTAL ANASTOMOSIS | 1 | 47 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.13 [0.00, 6.54] | | 4 radiological dehiscence | 4 | 662 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.03 [0.59, 1.81] | | 4.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 2.04 [0.92, 4.50] | | 4.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.32 [0.37, 4.75] | | 4.3 COLORECTAL ANASTOMOSIS | 1 | 272 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.27 [0.10, 0.76] | | 5 stricture | 5 | 826 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 3.86 [2.03, 7.36] | | 5.1 INFRAPERITONEAL ANASTOMOSIS | 1 | 102 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 3.90 [1.06, 14.30] | | 5.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 154 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.75 [0.34, 8.92] | | 5.3 COLORECTAL ANASTOMOSIS | 3 | 570 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 4.74 [2.06, 10.90] | | 6 hemorrhage | 4 | 662 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.78 [0.84, 3.81] | | 6.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.86 [0.22, 3.31] | | 6.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 6.82 [1.15, 40.41] | | 6.3 COLORECTAL ANASTOMOSIS | 1 | 272 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.74 [0.59, 5.09] | | 7 reoperation | 3 | 544 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.94 [0.95, 3.98] | | 7.1 INFRAPERITONEAL ANASTOMOSIS | 1 | 113 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.60 [0.48, 5.27] | | 7.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 6.73 [0.93, 48.93] | | 7.3 COLORECTAL ANASTOMOSIS | 1 | 272 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.62 [0.59, 4.44] | | 8 wound infection | 5 | 508 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.21 [0.49, 2.95] | | 8.1 INFRAPERITONEAL ANASTOMOSIS | 2 | 231 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 2.43 [0.68, 8.68] | | 8.2 SUPRAPERITONEAL ANASTOMOSIS | 2 | 229 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.54 [0.13, 2.19] | | 8.3 COLORECTAL ANASTOMOSIS | 1 | 48 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.0 [0.06, 16.47] | | 9 anastomosis duration | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | ‐7.60 [‐12.92, ‐2.28] | | 9.1 INFRAPERITONEAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | | 9.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | ‐7.60 [‐12.92, ‐2.28] | | 9.3 COLORECTAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | | 10 hospital stay | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | 2.0 [‐3.27, 7.27] | | 10.1 INFRAPERITONEAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | | 10.2 SUPRAPERITONEAL ANASTOMOSIS | 1 | 159 | Mean Difference (IV, Fixed, 95% CI) | 2.0 [‐3.27, 7.27] | | 10.3 COLORECTAL ANASTOMOSIS | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | Characteristics of studies Characteristics of included studies [ordered by study ID] Beart 1991. | Methods | RCT | | | Participants | 1) Inclusion criteria: any patient requiring an anterior anastomosis 2) Type of disease: malignant=53, prolapse of the rectum=06, diverticulosis=11 3) Exclusion criteria: patients with cancer the rectum within 5 cm of the dentate line were excluded, but those with benign disease in this region were included 4) Type of anastomosis: end‐to‐end 5) Number of participants: 70, S=35, HS=35 6) Age: 64 (25‐88) years, S=62, HS=65 7) Sex: not described 8) Location of study: Mayo Clinic and Mayo Foundation, Rochester, Minnesota, USA 9) Time of study: June 1978 to January 1980 | | | Interventions | 1) Group S (n = 35): stapler, EEA 2) Group HS (n = 35): hand sewn with two layer | | | Outcomes | 1) wound infection 2) pelvic infection 3) anastomotic dehiscence 4) mortality 5) urinary infection, urinary retention, pneumonitis, phlebitis 6) time of anastomosis 7) duration of hospital stay 8) mishaps | | | Notes | 1) Sample size: not described 2) Representativity: not described 3) Intention to treat: done 4) Learning curve: after becoming familiar with the EEA stapler in the laboratory and though clinical use in performing 20 rectal anastomosis, they initiated a randomised prospective evaluation of the end‐to‐end stapler 5) Distance anastomosis and anal verge (cm): S= 9.4, HS=9.8 6) Dukes: not described 7) Factors influencing safety of anastomosis: loop colostomy 8) Others: not described 9) Follow up not described | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Low risk | Computer randomizations sequence | | Allocation concealment (selection bias) | Low risk | At operation, if either anastomosis could be performed | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Unclear risk | Not described | | Selective reporting (reporting bias) | Low risk | All outcomes described | Elhadad 1990. | Methods | RCT | | | Participants | 1) Inclusion criteria: elective colorectal resection 2) Type of disease: neoplastic and diverticular disease 3) Exclusion criteria: emergency operator and inflammatory bowel disease 4) Type of anastomosis: end‐to‐end, end‐to‐side, side‐to end 5) Number of participants: 272 S=139, HS=133 6) Age: not described 7) Sex: 135 men, 136 women 8) Location of study: 26 centres, Aulnay‐sous‐Bois, France 9) Time of study: not described | | | Interventions | 1) Group S (n=139) stapler: EEA, DEEA and ILS 2) Group HS (n=133) interrupted or not interrupted suture | | | Outcomes | 1) clinical leakage 2) radiological leakage 3) duration of anastomosis 4) duration of operation 5) haemorrhage 6) stricture 7) cost | | | Notes | 11) Sample size: not described 2) Representativity: not described 3) Intention to treat: not described to stricture 4) Learning curve: not described 5) Distance anastomosis and anal verge(cm): not described 6) Dukes: not described 7) Factors influencing safety of anastomosis: diverting colostomy and omental wrap 8) Others: not described 9) Follow up not described | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Unclear risk | Not described | | Allocation concealment (selection bias) | Low risk | At operation, if either anastomosis could be performed | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Unclear risk | Not described | | Selective reporting (reporting bias) | Low risk | All outcomes described | Fingerhut 1994. | Methods | RCT | | | Participants | 1) Inclusion criteria: only patients who underwent immediate infra‐peritoneal anastomosis after elective left colorectal resection. 2) Type of disease: carcinoma 102, S=46, HS=56; diverticulosis 9, S=7, HS=2; benign tumour 2, S=1, HS=1 3) Exclusion criteria: emergency procedure, colitis, reestablishment of continuity or supra peritoneal anastomosis 4) Anastomosis: end‐to‐end 57, S=37, HS=20; side‐to‐end 50, S=11, HS=39; end‐to‐side 6, S=6, HS=0 5) Number of participants: 113, S=54, HS=59 6) Age: 66.5 (30‐87) years 7) Sex: 48 men, 65 women 8) Location of study: 24 different institutions, Bois‐Colombe, France 9) Time of study: May 1986‐October 1988 10) Number of participants: 113 (48 male; 65 female) 11) Age: 67±12 years (range: 30 to 87 years) 12) Location of study: 24 different institutions (Bois‐Colombe, France) 13) Time of study: May 1986 ‐ October 1988 | | | Interventions | Group HS (n=59): hand‐sewn, one extra mucosal layer, with continuous or interrupted, with 3/0 polyglycolic acid or polyglactin in general | | | Outcomes | 1) duration of anastomosis 2) duration of operation 3) duration of hospital stay 4) mortality 5) clinical leakage 6) radiological leakage 7) wound abscess 8) localized and generalized peritonitis 9) haemorrhage per anum 10) reoperation 11) mishaps 12) extra‐abdominal complications 13) strictures | | | Notes | 1) Sample size: the number of patients required was based on the hope of improving the rate of anastomotic failure from 15% to 5% with an a priori gamma risk at 5% according to the pragmatic method of Schwartz et al 2) Representativity: 113 consecutive patients 3) Intention to treat: done, except in stricture 4) Learning curve: all participating surgeons had at least 3 years experience with stapling devices and had accomplished at least 30 anastomoses 5) Distance anastomosis and anal verge cm: 6 (2‐13) 6) Dukes carcinoma: 102, S=46, HS=46 A e B: 48, S=17, HS=31 C: 54, S=29, HS=25 Paliativo: 17, S=06, HS=11 7) Factors influencing safety of anastomosis: tested for air‐tightness: 52, S=29, HS=23 leakage detected: 02, S=02, HS=0 extra sutures added: 01, S=01, HS=0 doughnuts verified: 49, S=49, HS=0 defect detected: 06 sutures added: 02 reanastomosis: 02 diverting colostomy: 28, S=13, HS=15 omental wrap: 25, S=08, HS=17 8) Others: all patients had mechanical colonic preparation and antibiotic prophylaxis 9) Follow up: 6‐10 months | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Low risk | Adequate: computerized random number tables. | | Allocation concealment (selection bias) | Low risk | At operation, if either anastomosis could be performed | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Low risk | Withdraws and dropouts:11 patients (10.8%) | | Selective reporting (reporting bias) | Low risk | All outcomes described | Fingerhut 1995. | Methods | RCT | | | Participants | 1) Inclusion criteria: only patients who underwent immediate supra‐peritoneal anastomosis after elective colorectal resection. 2) Type of disease: carcinoma: 92, S=53, HS=39; diverticulosis: 56, S=28, HS=28; benign tumour: 11, S=04, HS=07 3) Exclusion criteria: patients who were operated on for an emergency procedure, those with colitis or who had reestablishment of digestive continuity after Hartmann's procedure or patients undergoing ileocolonic, colocolonic, ileorectal or intra‐peritoneal anastomosis 4) Anastomosis: end‐to‐end: 82, S=40, HS=42; side‐to‐end: 72, S=41, HS=31; end‐to‐side: 05. S=04, HS=01 5) Number of participants: 159, S=85, HS=74 6) Age: 65.8 (33‐99) 7) Sex:: 88 men, 71 women 8) Location of study: 24 centres, Bois‐Colombes, France 9) Time of study: May 1986‐October 1988 | | | Interventions | Group S (n=85): stapler: EEA, DDEA and ILS Group HS (n=74): handsewn(one extra‐mucosal layer with continuous or interrupted sutures, preferably using 3/0 polyglycolic acid or polyglactin sutures) | | | Outcomes | 1) clinical leakage 2) radiological leakage 3) wound abscess 4) deep abscess 5) generalized peritonitis 6) haemorrhage per anus 7) reoperation 8) extra‐abdominal complications 9) mortality 10) mishaps 11) duration of anastomosis 12) duration of hospital stay 13) duration of operation | | | Notes | 1) Sample size: the number of patients required was based on the hope of improving the rate of anastomotic leakage from 12% (found in preceding studies) to 5% with a priori gamma risk set at 5% according to Schwartz's pragmatic method. However a posteriori gamma risk up to 15%. Seventh patients required in each group. 2) Representativity: 159 consecutive patients 3) Intention to treat: done, except in stricture 4) Learning curve: all participating surgeons had at least 3 years of experience with the stapling device and had accomplished at least 30 anastomosis. 5) Distance anastomosis and anal verge (cm): 11(6‐20) 6) Dukes carcinoma: 92, S=53 HS=39 A e B: 25, S=13 HS=12 C: 53, S=31 HS=22 palliative: 14, S=09 HS=05 7) Factors influencing safety of anastomosis tested for air tightness: 63, S=47 HS=16 leakage detected: 05, S=04 HS=01 extra sutures added: 05, S=04 HS=01 doghnuts verified: 83, S=83 HS=0 defect detected: 04, S=04 HS=0 sutures added: 03, S=03 HS=0 reanastomosis: 03, S=03 HS=0 doughnuts not verified: 02, S=02 HS=0 diverting colostomy: 09, S=04 HS=05 omental wrap: 28, S=14 HS=16 8) Others: all patients had mechanical colonic preparation and antibiotic prophylaxis. *a: one leakage (R) occurred even though extra sutures had been added. *b: in 2 sutures were added fist and than judged to be insufficient. *c: all these anastomoses were redone mechanically with one ensuing post‐operative leak. 9) Follow up: 8 months | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Low risk | Adequate: computerized random number tables. | | Allocation concealment (selection bias) | Low risk | At operation, if either anastomosis could be performed | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Low risk | Withdraws and dropouts: 5 patients (3.2%) | | Selective reporting (reporting bias) | Low risk | All outcomes described | Gonzalez 1987. | Methods | RCT | | | Participants | 1) Inclusion criteria: only the elective operations performed in patients with tumours located 5 to 15 cm above the anal verge 2) Type of disease: rectal cancer 3) Exclusion criteria: not described 4) Type of anastomosis: end‐to‐end 5) Number of participants: 113, S=55 HS=58 6) Age: 59, S=61 HS=58 7) Sex: not described 8) Location of study: Madri , Spain 9) Time of study: from 1979 to 1985 | | | Interventions | 1) Group S (n=55): stapler EEA 2) Group HS (n=58): handsewn | | | Outcomes | 1) general complications 2) infectious complications 3) mortality 4) clinical leakage 5) radiological leakage 6) local recurrence | | | Notes | 1) Sample size: not described 2) Representativity: consecutive patients 3) Intention to treat: done 4) Learning curve: not described 5) Distance anastomosis and anal verge(cm): S=4.2 HS=3.2 6) Dukes: A=12, S=08 HS=04 B=52, S=23 HS=29 C=41, S=20 HS= 21 D=08, S=04 HS=04 7) Factors influencing safety of anastomosis: not described 8) Others: not described 9) Follow up: 3 years at least | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Unclear risk | Not described | | Allocation concealment (selection bias) | Unclear risk | Not described | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Low risk | Withdraws and dropouts: none | | Selective reporting (reporting bias) | Low risk | All outcomes described | Kracht 1991. | Methods | RCT | | | Participants | 1) Inclusion criteria: all patients submitted to elective colorectal resection 2) Type of disease: neoplastic and diverticular disease 3) Exclusion criteria: emergency operation and inflammatory bowel disease 4) Type of anastomosis: end‐to‐end, end‐to‐side, side‐to‐end 5) Number of participants: 268, S=137 HS=131 6) Age: not described 7) Sex: not described 8) Location of study: Bois‐Colombes, France 9) Time of study: not described | | | Interventions | 1) Group S(n=137): stapler EEA, DEEA and ILS 2) Group HS (n=131): interrupted or not interrupted suture | | | Outcomes | 1) clinical leakage 2) mortality 3) stricture | | | Notes | 1) Sample size: anastomotic dehiscence: improving the results from 10% to 5%, gamma risk of 5% 2) Representativity: not described 3) Intention to treat: done, except in stricture 4) Learning curve: not described 5) Distance anastomosis and anal verge cm: not described 6) Dukes: not described 7) Factors influencing safety of anastomosis: diverting colostomy and omental wrap 8) Others: mechanical bowel preparation, fibreless diet, antibiotic prophylaxis | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Unclear risk | Generation sequence process not described | | Allocation concealment (selection bias) | Unclear risk | Not described | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Low risk | Dropout of 09 patients (3.5%) | | Selective reporting (reporting bias) | Low risk | All outcomes described | McGinn 1985. | Methods | RCT | | | Participants | 1) Inclusion criteria: low colorectal anastomosis in elective, except in 02 (two) cases, S=01 HS=01 2) Type of disease: carcinoma 107, S=51 HS=56, diverticular disease 10(07/03), endometriosis 01, S=0 HS=01 3) Exclusion criteria: higher anastomosis 4) Anastomosis: end‐to‐end 5) Number of participants: 118, S=58 HS=60 6) Age: 66.5 (37‐85) years 7) Sex: 60 men, 58 women 8) Location of study: South Hamptom General Hospital, United Kingdom 9) from September 1979 to 1984 | | | Interventions | 1) Group S (n=58): stapler EEA and ILS 2) Group HS (n=60): single layer, interrupted extra‐mucosal (3.0 braided polyamide) | | | Outcomes | 1) mishaps 2) clinical leakage 3) radiological leakage 4) wound infection 5) colostomy 6) rectal haemorrhage 7) duration of surgery 8) duration of hospital stay 9) cost 10) mortality | | | Notes | 1) Sample size: not described 2) Representativity: 118 consecutive patients 3) Intention to treat: it's done 4) Learning curve: senior registrars performed 24 of the 118 operations 5) Distance anastomosis and anal verge (cm): 3.5‐5.0 (43, S=22 HS=21) 5.5‐8.0 (54, S=26 HS=28); 8.5‐11 (21, S=10 HS=11) 6) Dukes: not described 7) Factors influencing safety of anastomosis: hydro pneumatic test, oversewn and colostomies 8) Others: mechanical colonic preparation and antibiotic prophylaxis 9) Follow up: intra‐hospital | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Unclear risk | Not described | | Allocation concealment (selection bias) | Low risk | At operation, if either anastomosis could be performed | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Unclear risk | Not described | | Selective reporting (reporting bias) | Low risk | All outcomes described | Sarker 1994. | Methods | RCT | | | Participants | 1) Inclusion criteria: cancer of the rectum judged to be resectable and located more than 6 cm from the anal verge 2) Type of disease: adenocarcinoma rectum 3) Exclusion criteria: not described 4) Type of anastomosis: end‐to‐end 5) Number of participants: 60, S=30 HS=30 6) Age: 60 (24‐76) years 7) Sex: 49 men, 11 women 8) Location of study: Army Reserch‐Referral Hospital New Delhi, India 9) Time of study: from 1984 to 1990 | | | Interventions | Group S (n=30): stapler EEA Group HS (n=30): two layers using 3/0 silk | | | Outcomes | 1) time of anastomosis 2) clinical leak 3) radiological leak 4) wound infection 5) local recurrence | | | Notes | 1) Sample size: not described 2) Representativiy: not described 3) Intention to treat: done, except in local recurrence 4) Learning curve: not described 5) Distance anastomosis and anal verge: 6) Dukes: A=04, S=02 HS=02 B=40, S=19 HS=21 C= 09, S=05 HS=04 D=07, S=04 HS=03 7) Factors influencing safety of anastomosis: hydro pneumatic test 8) Others: bowel preparation, low residue diet and antibiotic 9) Length of follow up: 1 to 7 years | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Unclear risk | Generation sequence process not described | | Allocation concealment (selection bias) | Unclear risk | Not described | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | Unclear risk | Not described | | Selective reporting (reporting bias) | Low risk | All outcomes described | Thiede 1984. | Methods | RCT | | | Participants | 1) Inclusion criteria: elective colorectal resection 2) Type of disease: neoplastic and diverticular disease 3) Exclusion criteria: not described 4) Type of anastomosis: end‐to‐end 5) Number of participants: 60, S=29 HS=31 6) Age: not described 7) Sex: 30 men, 30 women 8) Location of study: Cristian‐Albrechts‐Universitat Kiel, Kiel, Germany 9) Time of study: December 1979‐January 1982 | | | Interventions | 1) Group S (n=29): stapler EEA 2) Group HS (n=31): handsewn with two layer? | | | Outcomes | 1) clinical leakage 2) stricture 3) mortality 4) wound infection | | | Notes | 1) Sample size: not described 2) Representativity: not described 3) Intention to treat: not done 4) Learning curve: not described 5) Distance anastomosis and anal verge: not described 6) Dukes: not described 7) Factors influencing safety of anastomosis: not described 8) Others: not described | | | Risk of bias | || | Bias | Authors' judgement | Support for judgement | | Random sequence generation (selection bias) | Low risk | Adequate: random number table | | Allocation concealment (selection bias) | Low risk | At operation, if either anastomosis could be performed | | Blinding (performance bias and detection bias) All outcomes | Unclear risk | Not described | | Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Not described | | Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Not described | | Incomplete outcome data (attrition bias) All outcomes | High risk | Withdraws and dropouts: 13 (31.7%) | | Selective reporting (reporting bias) | Low risk | All outcomes described | Characteristics of excluded studies [ordered by study ID] | Study | Reason for exclusion | |---|---| | Bocassanta 2006 | Concerns anorectal anastomosis. | | Cajozzo 1990 | Inclusion criteria are very wide: right sided anastomoses were included. | | Catena 2004 | Only colorectal anastomoses in emergency patients were included. | | Choy 2007 | Concerns right sided anastomosis (ileocolic anastomosis). | | Demetriades 2002 | Retrospective study, no randomisation was done. | | Didolkar 1985 | Inclusion criteria are very wide: right sided anastomoses were included. | | Everett 1986 | The authors selected only colorectal anastomosis over 12 cm above the dentate line. | | Fukunaga 2007 | A novel technique is also included in the comparison. | | Ikeuchy 2000 | Inclusion criteria were wide: not only colorectal anastomoses included. | | Korolija 2008 | Inclusion criteria are very wide: all anastomosis of gastrointestinal tract were included. | | Laurent 2005 | Concerns ileoanal anastomosis. | | Lovegrove 2006 | Concerns ileoanal anastomosis. | | McLeod 2009 | Concerns ileocolic anastomosis. | | Nakagoe 2005 | Concerns coloanal anastomosis. | | Papp 2007 | Retrospective analysis, no randomisation was done. | | Reiling 1980 | Inclusion criteria are very wide: all anastomosis of gastrointestinal tract were included. | | Resegotti 2005 | Concerns ileoanal anastomosis. | | Shelygin 2010 | The comparison of the types of suture were about the ileostomy closure, not the colorectal anastomosis. | | West 1991 | Colorectal anastomosis in emergency patients were included. | | Wrighton 2008 | Our inclusion criteria accepts only adults. | Contributions of authors Conceiving the review: DM, SASL Designing the review: CBN, SASL, DM, IP, EMKS Coordinating the review: CBN Data collection for the review: CBN, SALS Developing search strategy: Marija Barbatescovic Undertaking searches: CBN, SASL Screening search results: CBN, SASL, DM Organising retrieval of papers: CBN, SASL,DM Screening retrieved papers against inclusion criteria: CBN, SASL Appraising quality of papers: CBN, SASL, DM, IP Abstracting data from papers: CBN, SASL, DM Writing to authors of papers for additional information: DM, SASL Providing additional data about papers: CBN, SASL, EMKS, IP Obtaining and screening data on unpublished studies: not applicable Data management for the review: CBN, SALS, DMS Entering data into RevMan: CBN, EMKS Analysis of data: CBN, DM, SALS Interpretation of data: CBN, SALS, DM, EMKS Providing a methodological perspective: CBN, EMKS Providing a clinical perspective: CBN, DM, SALS Providing a policy perspective: DM Providing a consumer perspective: none Writing the review: CBN, EKMS, DM Providing general advice on the review: DM, EKMS Securing funding for the review: DM, EKMS Performing previous work that was the foundation of current study: SALS, DM Sources of support Internal sources Clinical Trials and Meta‐analysis Unit, Federal University of São Paulo, Brazil. Gastroenterology Surgical Division, Federal University of São Paulo, Brazil. External sources Centro Universitario de Volta Redonda (UniFOA), Brazil. Fundação Coordenação de Aperfeiçoamento de Pessoal do Ensino Superior (CAPES), Brazil. Declarations of interest None. New search for studies and content updated (no change to conclusions)

References to studies included in this review Beart 1991 {published data only} - Beart Jr RW, Kelly KA. Randomized prospective evaluation of EEA stapler colorectal anastomoses. American Journal of Surgery 1981;141:143‐7. [DOI] [PubMed] [Google Scholar] Elhadad 1990 {published data only} - Elhaddad A. Anastomoses colo‐rectales: à la main ou à la machine. Der Chirurg; Zeitschrift fur alle Gebiete der operativen Medizen 1990;116:425‐8. [Google Scholar] Fingerhut 1994 {published data only} - Fingerhut A, Hay JM, Elhaddad A, Lacaine F, Flamant Y. Supraperitoneal colorectal anastomosis: handsewn versus circular staples: a controlled clinical trial. Surgery 1995;116:484‐90. [PubMed] [Google Scholar] Fingerhut 1995 {published data only} - Fingerhut A, Hay JM, Elhaddad A, Lacaine F, Flamant Y. Supraperitoneal colorectal anastomosis: handsewn vs. circular staples: a controlled clinical trial. Surgery 1995;118:479‐84. [DOI] [PubMed] [Google Scholar] Gonzalez 1987 {published data only} - Gonzalez EM, Selas PR, Molina DM, Sanz RG, Martinez RR, Gonzalez JS, et al. Results of surgery for cancer of the rectum with sphincter conservation. Acta Oncologica 1989;2:241‐4. [DOI] [PubMed] [Google Scholar] Kracht 1991 {published data only} - Kracht M. Le point sur les meilleurs anastomoses après reséction colique. Annales de chirurgie 1991;45:295‐8. [PubMed] [Google Scholar] McGinn 1985 {published data only} - McGinn FP, Gartell PC, Clifford PC, Brunton FJ. Staples or sutures to colorectal anastomoses: prospective randomized trial. The British Journal of Surgery 1985;72:603‐5. [DOI] [PubMed] [Google Scholar] Sarker 1994 {published data only} - Sarker SK, Chaudhry R, Sinhab VK. A comparison of stapled vs. handsewn anastomoses in anterior resection for carcinoma rectum. Indian Journal of Cancer 1994;31:133‐7. [PubMed] [Google Scholar] Thiede 1984 {published data only} - Thiede A, Schubert G, Poser HL, Jostardnt L. Technic of rectum anastomoses in rectum resection, a controlled study: instrumental suture versus hand suture [Zur technik der rectumanastomosen bei rectumresektionen]. Der Chirurg; Zeitschrift fur alle Gebiete der operativen Medizen 1984;55(5):326‐35. [PubMed] [Google Scholar]

to studies excluded from this review Bocassanta 2006 {published data only} - Boccasanta P, Venturi M, Barbieri S, Roviaro G. Impact of new technologies on the clinical and functional outcome of Altemeier's procedure: a randomized, controlled trial. Diseases of the Colon and Rectum 2006 May;49(5):652‐60. [DOI] [PubMed] [Google Scholar] Cajozzo 1990 {published data only} - Cajozzo M, Compagno G, DiTora P, Spallita SI, Bazan P. Advantages and disadvantages of mechanical vs. manual anastomoses in colorectal surgery. Acta Chirurgica Scandinavica 1990;156:167‐9. [PubMed] [Google Scholar] Catena 2004 {published data only} - Catena F, Donna M, Gagliardi S, Avanzolini A, Taffurelli M. Stapled versus hand‐sewn anastomoses in emergency intestinal surgery: results of a prospective randomized study. Surgery Today 2004;34(2):123‐6. [DOI] [PubMed] [Google Scholar] Choy 2007 {published data only} - Choy PYG, Bissett IP, Docherty JG, Parry BR, Merrie A, Fitzgerald A. Stapled versus handsewn methods for ileocolic anastomoses. Cochrane Database of Systematic Reviews 2011, Issue 9 Art. No.: CD004320.. [DOI: 10.1002/14651858.CD004320.pub3] [DOI] [PubMed] [Google Scholar] Demetriades 2002 {published data only} - Demetriades D, Murray JA, Chan LS, Ordonez C, Bowley D, Nagy KK, et al. Handsewn versus stapled anastomosis in penetrating colon injuries requiring resection: a multicenter study. The Journal of Trauma 2002 Jan;52(1):117‐21. [DOI] [PubMed] [Google Scholar] Didolkar 1985 {published data only} - Didolkar MS, Reed WP, Elias EG, Schnaper LA, Brown SD, Chaudhary SM. A prospective randomized study of sutured vs. stapled bowel anastomosis in patientes with cancer. Cancer 1986;57:456‐60. [DOI] [PubMed] [Google Scholar] Everett 1986 {published data only} - Everett WG. A comparison of layer and two layer techniques for colorectal anastomosis. The British Journal of Surgery 1975;62:135‐40. [DOI] [PubMed] [Google Scholar] Fukunaga 2007 {published data only} - Fukunaga Y, Higashino M, Tanimura S, Osugi H. Triangulating stapling technique for reconstruction after colectomy. Hepatogastroenterology 2007 Mar;54(74):414‐7. [PubMed] [Google Scholar] Ikeuchy 2000 {published data only} - Ikeuchi H, Kusunoki M, Yamamura T. Long‐term results of stapled and hand‐sewn anastomoses in patients with Crohn's disease. Digestive Surgery 2000;17(5):493‐6. [DOI] [PubMed] [Google Scholar] Korolija 2008 {published data only} - Korolija D. The current evidence on stapled versus hand‐sewn anastomoses in the digestive tract. Minimally Invasive Therapy & Allied Technologies 2008;17(3):151‐4. [DOI] [PubMed] [Google Scholar] Laurent 2005 {published data only} - Laurent A, Parc Y, McNamara D, Parc R, Tiret E. Colonic J‐pouch‐anal anastomosis for rectal cancer: a prospective, randomized study comparing handsewn vs. stapled anastomosis. Diseases of the Colon and Rectum 2005 Apr;48(4):729‐34. [DOI] [PubMed] [Google Scholar] Lovegrove 2006 {published data only} - Lovegrove RE, Constantinides VA, Heriot AG, Athanasiou T, Darzi A, Remzi FH, et al. A comparison of hand‐sewn versus stapled ileal pouch anal anastomosis (IPAA) following proctocolectomy: a meta‐analysis of 4183 patients. Annals of Surgery 2006 Jul;244(1):18‐26. [DOI] [PMC free article] [PubMed] [Google Scholar] McLeod 2009 {published data only} - McLeod RS, Wolff BG, Ross S, Parkes R, McKenzie M, Investigators of the CAST Trial. Recurrence of Crohn's disease after ileocolic resection is not affected by anastomotic type: results of a multicenter, randomized, controlled trial. Diseases of the Colon and Rectum 2009 May;52(5):919‐27. [DOI] [PubMed] [Google Scholar] Nakagoe 2005 {published data only} - Nakagoe T, Ishikawa H, Sawai T, Tsuji T, Takeshita H, Nanashima A, et al. Oncological outcome of ultra‐low anterior resection with total mesorectal excision for carcinoma of the lower third of the rectum: Comparison of intrapelvic double‐stapled anastomosis and transanal coloanal anastomosis. Hepatogastroenterology 2005 Nov‐Dec;52(66):1692‐7. [PubMed] [Google Scholar] Papp 2007 {published data only} - Papp Z, Szalai L, Molnar G, Czako T, Assefa A, Petri I. Evaluation of colonic anastomoses performed in the last twenty‐five years. Orvosi Hetilap 2007 Jan;148(3):117‐20. [DOI] [PubMed] [Google Scholar] Reiling 1980 {published data only} - Reiling RB, Reiling Jr WA, Bernie WA, Huffer AB, Perkins NC, Eliott DW. Prospective controlled study of gastrointestinal stapled anastomosis. American Journal of Surgery 1980;139:147‐52. [DOI] [PubMed] [Google Scholar] Resegotti 2005 {published data only} - Resegotti A, Astegiano M, Farina EC, Ciccone G, Avagnina G, Giustetto A, et al. Side‐to‐side stapled anastomosis strongly reduces anastomotic leak rates in Crohn’s disease surgery. Diseases of the Colon and Rectum 2005 March;48(3):464‐8. [DOI] [PubMed] [Google Scholar] Shelygin 2010 {published data only} - Shelygin YA, Chernyshov SV, Rybakov EG. Stapled ileostomy closure results in reduction of postoperative morbidity. Techniques in Coloproctology 2010 Mar;14(1):19‐23. [DOI] [PubMed] [Google Scholar] West 1991 {published data only} - West of Scotland and Highland Anastomosis Study Group. Suturing or stapling in gastrointestinal surgery: a prospective randomized study. The British Journal of Surgery 1991;78:337‐41. [DOI] [PubMed] [Google Scholar] Wrighton 2008 {published data only} - Wrighton L, Curtis JL, Gollin G. Stapled anastomosis in infants. Journal of Pediatric Surgery 2008 Dec;43(12):2231‐4. [DOI] [PubMed] [Google Scholar] Additional references Docherty 1995 - Docherty JG, McGregor JR, Akyol AM, Murray GD, Galloway DJ. Comparison of manually constructed and stapled anastomoses in colorectal surgery. West of Scotland and Highland Anastomosis Study Group. Annals of Surgery 1995 Feb;22(2):176‐84. [DOI] [PMC free article] [PubMed] [Google Scholar] Fain 1975 - Fain SN, Patin CS, Morgenstern L. Use of a mechanical suturing apparatus in low colorectal anastomosis. Archives of Surgery 1975;110:1079‐82. [DOI] [PubMed] [Google Scholar] Heald 1981 - Heald RJ, Leciester RJ. The low stapled anastomosis. The British Journal of Surgery 1981;68:333‐7. [DOI] [PubMed] [Google Scholar] Higgins 2011 - Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration. Available from www.cochrane‐handbook.org, 2011. [Google Scholar] Karanjia 1994 - Karanjia ND, Corder AP, Bern P, Heald RJ. Leakage from stapled low anastomosis after total mesorectal excision for carcinoma of the rectum. The British Journal of Surgery 1994;81:1224‐6. [DOI] [PubMed] [Google Scholar] MacRae 1998 - MacRae HM, McLeod RS. Handsewn vs. stapled anastomoses in colon and rectal surgery: a meta‐analysis. Diseases of the Colon and Rectum 1998;41(2):180‐9. [DOI] [PubMed] [Google Scholar] Mann 1996 - Mann B, Kleinschmidt S, Stremmel W. Prospective study of hand‐sutured anastomosis after colorectal resection. The British Journal of Surgery 1996;83:29‐31. [DOI] [PubMed] [Google Scholar] Matos 1996 - Matos D. Compression intestinal anastomosis with AKA Russian instrument: analysis of clinical, radiological and endoscopic results. Personal communication 1996. Mealy 1992 - Mealy K, Burke P, Hyland J. Anterior resection without a defunctioning colostomy: questions of safety. The British Journal of Surgery 1992 April;79(4):305‐7. [DOI] [PubMed] [Google Scholar] Ravitch 1979 - Ravitch MM, Steichen FM. A stapling instrument for end‐to‐end inverting anastomosis in the gastrointestinal tract. Annals of Surgery 1979;189(6):791‐7. [DOI] [PMC free article] [PubMed] [Google Scholar] Shahinian 1980 - Shahinian TK, Bowen JR, Dorman BA, Sderberger CH, Thompson WR. Experience with the EEA stapling device. American Journal of Surgery 1980;139:549‐53. [DOI] [PubMed] [Google Scholar]