Design: Meta-analysis of randomised trials with trial sequential analysis
Population: 10 RCTs, 2,617 patients having midline laparotomy
Primary outcome: Incisional hernia
Effect: RR 0.48 (0.33–0.70) beyond 12 months; RR 0.56 (0.36–0.89) beyond 24 months

This meta-analysis pooled ten randomised trials (2,617 patients) comparing small-bite closure — 5 mm bites at 5 mm spacing — with large-bite closure at 10 mm for midline laparotomy. The primary outcome was incisional hernia, with trial sequential analysis to test whether the evidence was conclusive.

Across trials with more than 12 months of follow-up, small-bite closure reduced incisional hernia (pooled RR 0.48, 95% CI 0.33–0.70). The benefit held in trials followed beyond 24 months (RR 0.56) and in the five trials using imaging as well as examination (RR 0.57, I² 44%). Surgical-site infection (RR 0.72) and total complications (RR 0.76) were also lower. Closure took about 4.8 minutes longer; pain, seroma and dehiscence did not differ. Trial sequential analysis judged the evidence for benefit firm.

Incisional hernia follows a large share of midline laparotomies and many need repair. Small-bite closure costs a few minutes and no new equipment. For units that still close with large bites, this is a reason to change the default technique and teach it to trainees.

Use small-bite closure — about 5 mm bites, 5 mm apart, fascia only — as the default for elective midline laparotomy
Teach the technique to trainees and audit adherence in operation notes
Allow roughly five extra minutes for closure in theatre planning
Consider it in emergency laparotomy too, while noting most trial data are elective

Why it matters

One of the few closure changes with durable randomised evidence and no added cost.

The statistics, in plain English

A risk ratio of 0.48 means about half as many hernias with small bites. The benefit was larger where hernias were judged clinically only (RR 0.31) than where scans were used (RR 0.57); imaging finds more small hernias, so 0.57 is probably the more realistic estimate. Moderate heterogeneity (I² 44%) means trials varied somewhat but agreed on direction.