The authors reported that the review included 30 randomised controlled trials (RCTs)(n not reported).
Liver resection (3 RCTs): there was a statistically non significant decrease in infected intra-abdominal collections with no drainage compared with drainage (OR 2.83, 95% CI: 0.82, 9.71). There was no statistically significant difference between treatments for bile collections (OR 1.15, 95% CI: 0.36, 3.68) or pulmonary complications (OR 1.40, 95% CI: 0.73, 2.68). There was no statistically significant heterogeneity between the trials observed in any of the three meta-analyses.
Cholecystectomy (number of studies not reported): all trials failed to demonstrate a reduction in post-operative complications by routine drainage.
Pancreatic resection (1 RCT and 1 cohort study): both studies failed to demonstrate a reduction in complications with drainage compared with a no-drainage treatment strategy.
Oesophageal, gastric and duodenal surgery: no studies were identified on prophylactic drainage versus no drainage after oesophageal or gastric surgery. One prospective cohort study that assessed prophylactic drains after surgery for perforated duodenal ulcer found that drainage had no effect on the incidence of intra-abdominal fluid collections, including abscesses, or the duration of hospital stay.
Colorectal surgery (8 RCTs): there was no statistically significant difference between drainage and no drainage for clinical leakage (OR 1.38, 95% CI: 0.77, 2.49), wound infections (OR 1.41, 95% CI: 0.87, 2.29) and pulmonary complications (OR 0.83, 95% CI: 0.52, 1.32). No significant statistical heterogeneity was observed between the trials.
Acute Appendectomy (2 RCTs): one RCT found a significantly higher wound infection rate in drained patients with acute or simple appendicitis, whereas the second RCT found similar wound and intra-abdominal infection rates between the drained and non-drained groups.
Appendectomy for gangrenous and perforated appendicitis: there was no significant difference in the rates of intra-abdominal infection between the drained and non-drained patients (3 RCTs; OR 1.43, 95% CI: 0.39, 5.29). There was significant statistical heterogeneity between the 3 trials. There were no significant differences between the rates of wound infections between drained and non-drained patients (4 RCTs; OR 1.75, 95% CI: 0.96, 3.19). There was a significant reduction in faecal fistulas with no drainage compared with drainage (3 RCTs; OR 12.4, 95% CI: 1.14, 135). No significant statistical heterogeneity was observed between the trials in either of these meta-analyses.