EClinicalMedicine (Aug 2023)

Incisional negative pressure wound therapy for the prevention of surgical site infection: an up-to-date meta-analysis and trial sequential analysisResearch in context

  • Hannah Groenen,
  • Hasti Jalalzadeh,
  • Dennis R. Buis,
  • Yasmine E.M. Dreissen,
  • Jon H.M. Goosen,
  • Mitchel Griekspoor,
  • Wouter J. Harmsen,
  • Frank F.A. IJpma,
  • Maarten J. van der Laan,
  • Roald R. Schaad,
  • Patrique Segers,
  • Wil C. van der Zwet,
  • Stijn W. de Jonge,
  • Ricardo G. Orsini,
  • Anne M. Eskes,
  • Niels Wolfhagen,
  • Marja A. Boermeester

Journal volume & issue
Vol. 62
p. 102105

Abstract

Read online

Summary: Background: The evidence on prophylactic use of negative pressure wound therapy on primary closed incisional wounds (iNPWT) for the prevention of surgical site infections (SSI) is confusing and ambiguous. Implementation in daily practice is impaired by inconsistent recommendations in current international guidelines and published meta-analyses. More recently, multiple new randomised controlled trials (RCTs) have been published. We aimed to provide an overview of all meta-analyses and their characteristics; to conduct a new and up-to-date systematic review and meta-analysis and Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment; and to explore the additive value of new RCTs with a trial sequential analysis (TSA). Methods: PubMed, Embase and Cochrane CENTRAL databases were searched from database inception to October 24, 2022. We identified existing meta-analyses covering all surgical specialties and RCTs studying the effect of iNPWT compared with standard dressings in all types of surgery on the incidence of SSI, wound dehiscence, reoperation, seroma, hematoma, mortality, readmission rate, skin blistering, skin necrosis, pain, and adverse effects of the intervention. We calculated relative risks (RR) with corresponding 95% confidence intervals (CI) using a Mantel-Haenszel random-effects model. We assessed publication bias with a comparison-adjusted funnel plot. TSA was used to assess the risk of random error. The certainty of evidence was evaluated using the Cochrane Risk of Bias-2 (RoB2) tool and GRADE approach. This study is registered with PROSPERO, CRD42022312995. Findings: We identified eight previously published general meta-analyses investigating iNPWT and compared their results to present meta-analysis. For the updated systematic review, 57 RCTs with 13,744 patients were included in the quantitative analysis for SSI, yielding a RR of 0.67 (95% CI: 0.59–0.76, I2 = 21%) for iNPWT compared with standard dressing. Certainty of evidence was high. Compared with previous meta-analyses, the RR stabilised, and the confidence interval narrowed. In the TSA, the cumulative Z-curve crossed the trial sequential monitoring boundary for benefit, confirming the robustness of the summary effect estimate from the meta-analysis. Interpretation: In this up-to-date meta-analysis, GRADE assessment shows high-certainty evidence that iNPWT is effective in reducing SSI, and uncertainty is less than in previous meta-analyses. TSA indicated that further trials are unlikely to change the effect estimate for the outcome SSI; therefore, if future research is to be conducted on iNPWT, it is crucial to consider what the findings will contribute to the existing robust evidence. Funding: Dutch Association for Quality Funds Medical Specialists.

Keywords