Comparative effectiveness of different wound dressings for patients with partial-thickness burns: study protocol of a systematic review and a Bayesian framework network meta-analysis.

Abstract

Selecting a suitable wound dressing for patients with partial-thickness burns (PTBs) is important in wound care. However, the comparative effectiveness of different dressings has not been studied. We report the protocol of a network meta-analysis designed to combine direct and indirect evidence of wound dressings in the management of PTB. We will search for randomised controlled trials (RCTs) evaluating the wound-healing effect of a wound dressing in the management of PTB. Searches will be conducted in MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials, the Cochrane Wounds Group Specialised Register and CINAHL. A comprehensive search strategy is developed to retrieve articles reporting potentially eligible RCTs. Besides, we will contact the experts in the field and review the conference proceedings to locate non-published studies. The reference lists of articles will be reviewed for any candidate studies. Two independent reviewers will screen titles and abstracts of the candidate articles. All eligible RCTs will be obtained in full text to perform a review. Disagreement on eligibility of an RCT will be solved by group discussion. The information of participants, interventions, comparisons and outcomes from included RCTs will be recorded and summarised. The primary outcome is time to complete wound healing. Secondary outcomes include the proportion of burns completely healed at the end of treatment, change in wound surface area at the end of treatment, incidence of adverse events, etc. The result of this review will provide evidence for the comparative effectiveness of different wound dressings in the management of PTB. It will also facilitate decision-making in choosing a suitable wound dressing. We will disseminate the review through a peer-review journal and conference abstracts or posters. PROSPERO CRD42016041574; Pre-results.