Importance: Surgical site infection (SSI) complicates 2-5% of surgeries in the United States. Severity of SSI ranges from superficial skin infection to life-threatening conditions such as severe sepsis, and SSIs are responsible for increased morbidity, mortality, and economic burden associated with surgery. Staphylococcus aureus (S. aureus) is a commonly-isolated organism for SSI, and methicillin-resistant S. aureus SSI incidence is increasing globally.

Objective: The objective of this systematic review was to characterize risk factors for SSI within observational studies describing incidence of SSI in a real-world setting.

Evidence review: An initial search identified 328 titles published in 2002-2012; 57 were identified as relevant for data extraction. Extracted information included study design and methodology, reported cumulative incidence and post-surgical time until onset of SSI, and odds ratios and associated variability for all factors considered in univariate and/or multivariable analyses.

Findings: Median SSI incidence was 3.7%, ranging from 0.1% to 50.4%. Incidence of overall SSI and S. aureus SSI were both highest in tumor-related and transplant surgeries. Median time until SSI onset was 17.0 days, with longer time-to-onset for orthopedic and transplant surgeries. Risk factors consistently identified as associated with SSI included co-morbidities, advanced age, risk indices, patient frailty, and surgery complexity. Thirteen studies considered diabetes as a risk factor in multivariable analysis; 85% found a significant association with SSI, with odds ratios ranging from 1.5-24.3. Longer surgeries were associated with increased SSI risk, with a median odds ratio of 2.3 across 11 studies reporting significant results.

Conclusions and relevance: In a broad review of published literature, risk factors for SSI were characterized as describing reduced fitness, patient frailty, surgery duration, and complexity. Recognition of risk factors frequently associated with SSI allows for identification of such patients with the greatest need for optimal preventive measures to be identified and pre-treatment prior to surgery.

Competing Interests:The authors have the following interests: this study was sponsored by MedImmune, the employer of Frangiscos Sifakis, Hasan S. Jafri, Mathew Lo, Moe H. Kyaw at the time of the study. Ellen Korol, Karissa Johnston and Nathalie Waser are employed by Oxford Outcomes, and served as consultants for MedImmune in performing this research. There are no patents, products in development or marketed products to declare. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials, as detailed online in the guide for authors.