Closed-Incision Prophylactic Negative Pressure Wound Therapy in Emergency Laparotomy: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.

Surgical site infection (SSI) is common after emergency laparotomy, and closed-incision prophylactic negative pressure wound therapy (ciNPWT) applied over the closed incision has been proposed to reduce it, though trials disagree. We searched MEDLINE, Embase, CENTRAL, Scopus, and two trial registries to 21 June 2026, with no date or language restriction, for randomized controlled trials (RCTs) comparing ciNPWT with standard dressings in adults undergoing emergency laparotomy. The primary outcome was any incisional SSI within 30 days. We used random-effects meta-analysis with restricted maximum likelihood, Hartung-Knapp-Sidik-Jonkman confidence intervals, and prediction intervals, together with RoB 2 and GRADE; because heterogeneity tracked study quality, we present the primary outcome stratified by risk of bias. Seven RCTs were included (1,351 patients; largest approximately 840), of which five contributed to the primary analysis and two to sensitivity analyses only - one because it reported outcomes only for a mixed elective and emergency whole cohort rather than an extractable emergency subgroup, and one because it was at high risk of bias with an implausibly high control-arm infection rate. The single trial at low risk of bias showed no clinically important effect on incisional SSI [risk ratio (RR) 1.04, 95% confidence interval (CI) 0.83-1.30], whereas four small trials with some concerns showed a homogeneous reduction (RR 0.41, 0.28-0.59; I²=0%). The overall estimate was heterogeneous and non-significant (RR 0.54, 0.27-1.06; I²=67%; 95% prediction interval 0.09-3.02) and, being dominated by the single large trial, is not interpreted as a single effect; in leave-one-out analysis it became significant only when that trial was removed. Length of hospital stay (ratio of means 0.96, 0.86-1.07) and 30-day mortality (RR 0.95, 0.46-1.96) showed no difference. Among secondary outcomes, seroma was lower with ciNPWT (RR 0.59) and fascial dehiscence was uninformative (RR 1.15); the apparent SSI reduction was confined to the more subjectively assessed superficial component (RR 0.49), with no effect on deep SSI (RR 1.01). Certainty was low for the low-risk estimate (no benefit) and very low for the some-concern estimate (apparent benefit). Current evidence does not provide even low-certainty support that ciNPWT reduces SSI after emergency laparotomy; adequately powered trials with masked outcome assessment are needed.