Operative versus non-operative management of adhesive small bowel obstruction: A systematic review and meta-analysis.

OBJECTIVES: To investigate outcomes of operative and non-operative management of adhesive small bowel obstruction (SBO).
METHODS: We performed a systematic review in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement standards. We conducted a search of electronic information sources to identify all randomised controlled trials (RCTs) and observational studies investigating outcomes of operative versus non-operative management of patients with adhesive SBO. We used the Cochrane risk of bias tool and the Newcastle-Ottawa scale to assess the risk of bias of RCTs and observational studies, respectively. Fixed-effect or random-effects models were applied to calculate pooled outcome data.
RESULTS: We found one RCT, two prospective and three retrospective observational studies, enrolling a total of 876 patients. The analyses showed that operative management of adhesive SBO was associated with a lower risk of future recurrence [odds ratio (OR) 0.54, 95% confidence interval (CI) 0.38-0.76, P = 0.0005] but a higher risk of mortality [risk difference (RD) 0.03, 95% CI 0.01-0.06, P = 0.01] and complications (OR 5.39, 95% CI 2.97-9.78, P < 0.00001). There was no difference in need for surgical re-intervention rate (OR 0.72, 95% CI 0.35-1.47, P = 0.36) and length of stay [mean difference (MD) 5.07, 95% CI -2.36-12.49, P = 1.0] between operative and non-operative managements. The baseline suspicion of strangulation was a major confounding factor. When the baseline suspicion of strangulation was higher in the operative group, the risk of mortality (RD 0.04, 95% CI 0.02-0.07, P = 0.0006) and complications (OR 8.14, 95% CI 4.16-15.94, P = 0.00001) were higher in the operative group but the risk of recurrence was lower (OR 0.62, 95% CI 0.43-0.90, P = 0.01). When the baseline suspicion of strangulation was low in both groups, there was no difference in any of the outcomes except recurrence (OR 0.09, 95% CI 0.02-0.37, P = 0.0009) which was lower in the operative group.
CONCLUSIONS: The difference in baseline suspicion of strangulation between operative and non-operative groups is a major confounding factor in current literature. The benefit of surgical treatment should be balanced with the risks associated with surgery, patient's co-morbidities, and presence or absence of strangulation. Based on the best available evidence it could be argued that surgical intervention could be preserved for cases with high suspicion or evidence of bowel strangulation. The controversy still remains for optimum length of conservative management and timing of surgery (early or late) for cases with low baseline suspicion of strangulation. Randomised controlled trials are required to compare outcomes of early operation (<24 h) versus late operation (>24 h) and early operation versus conservative management in patients with low suspicion of strangulation. Crown