This review concluded that end-to-end anastomosis was associated with increased anastomotic leak compared with other anastomotic configurations; the authors suggest that results from future randomised controlled trials are needed before any definitive conclusions can be drawn. This review has a number of methodological weaknesses, but the authors’ conclusions are in line with the evidence presented and seem appropriate.

To compare end-to-end anastomosis (EEA) with other anastomotic configurations (OAC) after intestinal resection for patients with Crohn’s disease.

Studies in any language were identified through a computerised bibliographic search of MEDLINE, EMBASE and the Cochrane Library from 1985 to February 2006; the search terms were reported. Additional articles were identified through use of the ‘Related Articles’ function of PubMed.

Study designs of evaluations included in the review

All comparative study designs were eligible for inclusion.

Specific interventions included in the review

Studies comparing EEA with OAC were eligible for inclusion. Of the 712 anastomoses (performed on 661 patients), 54% were sutured end-to-end and 46% were OAC, the majority being side-to-side anastomoses.

Participants included in the review

Studies in patients undergoing intestinal resection and anastomosis for Crohn’s disease were eligible for inclusion. The mean age of the patients ranged from 26 to 40 years.

Outcomes assessed in the review

Studies that reported short-term complications, length of post-operative hospital stay or long-term complications were eligible for inclusion. Short-term complications included mortality, anastomotic leak and general post-operative complications. Long-term complications included perianastomotic recurrence and need for reoperation because of perianastomotic recurrence.

How were decisions on the relevance of primary studies made?

The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.

Two reviewers assessed study quality using a modified version of the Newcastle-Ottawa Scale. Factors related to patient selection, comparability of the study groups and outcome assessment were evaluated. Studies that scored 7 or more points out of a possible 11 were considered high quality.

Two reviewers independently extracted the data. Data on numbers of events in each group, means and standard deviations were used to calculate odds ratios (ORs) with 95% confidence intervals (CIs) for dichotomous variables (short-term complications), weighted mean differences (WMDs) with 95% CIs for continuous variables (post-operative hospital stay), and hazards ratios (HRs) with 95% CIs for long-term complications.

How were the studies combined?

The effect measures and corresponding 95% CIs were calculated using the DerSimonian and Laird random-effects model. Publication bias was assessed visually using funnel plots.

How were differences between studies investigated?

A chi-squared test was used to test for statistical heterogeneity in the meta-analyses. I-squared values were also calculated. Sensitivity analyses were conducted for the following subgroups; studies of higher quality, studies published during or after 2000, studies comparing side-to-side anastomosis and EEA, and studies that included only ileocolonic resections.

Eight studies (n=661) were included in the review: 2 randomised controlled trials (RCTs; n=161), 1 non-randomised prospective study (n=42) and 5 non-randomised retrospective studies (n=458). Three studies reported and analysed anastomotic events, while it appears that the remaining studies reported and analysed events by patient. The meta-analyses were based on all anastomotic events (n=712).

The methodological quality of the included studies was reported. The scores ranged from 5 to 9, with 5 studies scoring 7 points or more.

Short-term complications: anastomotic leak rates were significantly higher for EEA compared with OAC (6 studies; OR 4.37, 95% CI: 1.3, 14.72); there was no significant heterogeneity between the studies. There were no significant differences for any other short-term outcomes.

Post-operative hospital stay: there were no significant differences between the groups (4 studies; WMD 0.88, 95% CI: -1.43, 3.2), although there was significant heterogeneity for this outcome.

Long-term complications: there were no significant differences between the groups for perianastomotic recurrence (3 studies: HR 1.23, 95% CI: 0.52, 2.92) or the need for reoperation (4 studies: HR 1.76, 95% CI: 0.81, 3.83).

Sensitivity analyses showed that in studies reporting on side-to-side anastomosis compared with EEA, a significant difference favouring side-to-side was observed in anastomotic leak (OR 4.37, 95% CI: 1.3, 14.72), overall post-operative complications (OR 2.64, 95% CI: 1.49, 4.67) and post-operative hospital stay (WMD 2.81, 95% CI: 0.79, 4.84). Studies reporting on anastomosis after ileocolonic resection showed a significantly reduced rate of anastomotic leaks (OR 3.8, 95% CI: 1.01, 14.29) for OAC compared with EEA, while no differences were found for other outcomes. Analyses of high-quality studies and of more recent studies (since 2000) did not change the findings.

Evidence of publication bias was identified graphically.

EEA was associated with increased anastomotic leak compared with OAC. However because the majority of included studies were retrospective, results from future RCTs are needed before any definitive conclusions can be drawn.

The review addressed a clear question in terms of the intervention and outcomes of interest. However, the authors failed to state which study designs were eligible for inclusion and information about the eligibility of the patients was limited. There was no indication that the authors attempted to identify unpublished studies. Validity was assessed, although the results were reported as composite scores only; this makes it difficult for the reader to make their own assessment of study quality. Steps were taken to minimise bias and errors in the quality assessment and data extraction processes, but the methods used to make decisions about the relevance of primary studies were not reported.

Limited details about the included studies were reported: in particular, there was little information on patient characteristics and this makes it difficult to assess the generalisability of the review. The studies were combined by meta-analysis and sensitivity analyses were conducted to investigate the influence of individual study results. Statistical heterogeneity was investigated and was not significant for most outcomes. It was unclear whether all studies were based on anastomotic events or on patients. Combining the event rates of 3 studies with the patient rates in the remaining studies impacts on the significance of the findings; a subgroup analysis might have been helpful.

The authors’ conclusions are appropriately cautious and appear to be based on the data. Given the small number of trials, small sample sizes and the inclusion of retrospective data, the studies included in the review are not sufficient to draw conclusions about anastomosis configurations.

Practice: The authors did not state any implications for practice.

Research: The authors stated that further RCTs are needed to investigate anastomotic leak, post-operative complications, length of stay and perianastomotic recurrence.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.