Abstract
Background: Diabetic foot ulcers (DFUs) carry high risks of infection, amputation, and mortality. We systematically reviewed randomized controlled trials (RCTs) of negative-pressure wound therapy (NPWT), including single-use systems, for clinically uninfected DFUs (with sensitivity analyses for mixed/infected cohorts). Methods: We searched PubMed and Scopus (1 January 2004–30 June 2024). Dual reviewers performed screening and extraction; risk of bias was assessed with Cochrane Risk of Bias 2 (RoB 2) and certainty of evidence with GRADE. When ≥2 trials reported comparable outcomes, we used random-effects meta-analysis. The DiaFu cohort reported in two publications was counted once across analyses. Results: Eleven RCT publications (n = 1699; 10 unique cohorts) met criteria; eight trials (n = 1456) informed the primary endpoint. Trials largely excluded severe ischemia; findings therefore apply mainly to neuropathic or mixed-etiology DFUs with adequate perfusion. NPWT increased complete healing at 12–16 weeks (risk ratio [RR] 1.46, 95% CI 1.21–1.76; I2 = 48%) and shortened time to healing (mean difference –18 days, 95% CI −28 to −8). Effects were similar for conventional and single-use NPWT. Outcomes did not vary systematically within commonly used pressure ranges (approximately −80 to −125 mmHg). Only two RCTs reported direct cost data (exploratory). Moderate heterogeneity (Higgins’ I2 48–68%) reflected variation in ulcer severity, device type/settings, dressing-change frequency, and off-loading protocols. Conclusions: NPWT probably improves short-term healing of clinically uninfected DFUs compared with standard care and may reduce minor amputations, without increasing adverse events. Certainty is moderate for healing and low for most secondary outcomes. Benefits appear consistent across device classes and may support earlier discharge and community-based care. Evidence gaps include ischemia-dominated ulcers, long-term outcomes (recurrence and limb preservation), adherence mechanisms, and contemporary cost-effectiveness.

This publication has 31 references indexed in Scilit: