Antibiotic therapy for skin and soft tissue infections: a protocol for a systematic review and network meta-analysis

• Main Authors: Jessica J. Bartoszko, Dominik Mertz, Lehana Thabane, Mark Loeb
• Format: Article
• Language: English
• Published: BMC 2018-09-01
• Series: Systematic Reviews
• Subjects: Skin and soft tissue infection; Antibiotic therapy; Systematic review; Network meta-analysis; Protocol
• Online Access: http://link.springer.com/article/10.1186/s13643-018-0804-8

Abstract Background Skin and soft tissue infections (SSTIs) in hospital and community settings impose a substantial socio-economic burden. Therapeutic uncertainty due to the availability of a wide range of antibiotics and the need for empirical treatment decisions complicate SSTI clinical management. Completion of numerous pairwise meta-analyses to account for this variability in antibiotics is impractical, and many head-to-head comparisons of potential interest are likely not available. In comparing multiple antibiotics simultaneously, this network meta-analysis aims to identify the antibiotic(s) with the greatest value in the treatment of SSTIs, in terms of patient-important outcomes such as efficacy and safety. Methods We will conduct a systematic review to identify randomized controlled trials of persons with suspected or confirmed SSTI assigned to orally or parenterally administered antibiotic therapy that report results on at least one outcome of interest. We will search MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL), along with trial registries. Our primary outcome of interest is clinical success at the test-of-cure visit. Secondary outcomes may include (1) early clinical success (2–3 days after the therapy starts), (2) mortality, (3) adverse events, (4) treatment duration, and (5) length of hospital stay. Independent reviewers will complete screening of titles, abstracts, and full texts, data extraction, risk of bias assessment (using the Cochrane Risk of Bias tool), and evaluation of the certainty of evidence (using the GRADE approach) in duplicate. We will complete pairwise and network meta-analyses within the Bayesian framework when possible using a random effects model. We will stratify SSTIs by severity into uncomplicated and complicated SSTIs when possible. Subgroup analyses by age, infection type, comorbidity, and suspected or confirmed methicillin-resistant Staphylococcus aureus (MRSA)-associated infection are planned. Discussion This study has several strengths compared to previous reviews: inclusion of a wider range of infection types, antibiotics, and outcomes; a comprehensive search strategy; a priori subgroup analyses; application of GRADE; and improved interpretability of findings through visual presentation of results. We hope our findings will inform future research, health care professionals, and policy makers resulting in improved evidence-based clinical management of SSTIs. Systematic review registration PROSPERO CRD42018085607