A systematic review of intervention studies demonstrates the need to develop a minimum set of indicators to report the presence of burn wound infection

Introduction: Burn wound infections result in delayed healing and increased pain, scarring, sepsis risk and healthcare costs. Clinical decision making about burn wound infection should be supported by evidence syntheses. Validity of evidence from systematic reviews may be reduced if definitions of burn wound infectionvary between trials. This review aimed to determine whether burn wound infectionis defined, and whether there is variation in the indicators used to define burn wound infectionacross studies testing interventions for patients with burns. Method: Searches were carried out in four databases (Ovid Medline, Ovid Embase, Cinahl, CochraneRegister of Trials) to identify studies evaluating interventions for patients with burns and reporting a burn wound infection outcome. Pre-defined inclusion and exclusion criteria were systematically applied to select relevant studies. Data were systematically extracted and reported narratively. Results: 2056 studies were identified, of which 72 met the inclusion criteria, comprising 71 unique datasets. 52.1% of studies were randomised controlled trials. Twenty-eight (38.0%) studies reporting a burn wound infection outcome did not reporthow they had defined it. In the methods of included studies, 59 studies (83.1%) reported that they planned to measure burn wound infection as an outcome. Of these, 44 studies (74.6%) described how they had defined burn wound infection; 6 studies (13.6%) reported use of a previously developed consensus-informed definition of burn wound infection, and 41 studies (69.5%) described the specific indicators used to define it. Studies used between one (11 studies; 26.8%) and nine indicators (2 studies; 4.9%) to define burn wound infection (median = 3, inter-quartile range = 2). The most commonly used indicator was presence of bacteria in the wound (61.0% of studies). Only 13 studies (31.7%) defined burn wound infection using the same indicators as at least one other study. Discussion and conclusions: Within intervention studies reporting burn wound infection outcomes, a definition of this outcome is commonly not provided, or it varies between studies. This will prevent evidence synthesis to identify effective treatments for patients with burn injuries. Since there is no objective method for assessing burn wound infection, expert consensus is needed to agree a minimum set of indicators (Core Indicator Set) reported in all trials reporting burn wound infection as an outcome. (C) 2020 Elsevier Ltd and ISBI. All rights reserved.