Infection and Nonunion Rates in Open Fractures: Description of 6,042 Fractures from the FLOW and PREP-IT Trials

Infection and nonunion are common sequelae of open fractures. Studies have shown infection and nonunion rates ranging from 2% to 52% and 12% to 17%, respectively. The present article describes the rates of surgical site infection (SSI) and delayed union/nonunion following open fractures in a large contemporary series of patients from prospective clinical trials with adjudicated outcomes.

A descriptive analysis was performed with use of patient data from the FLOW, Aqueous-PREP, and PREPARE-Open studies. These studies, published within the past 10 years, included multiple international trauma centers and shared definitions for SSI and delayed union/nonunion. SSI and delayed union/nonunion rates were stratified by the OTA/AO fracture and Gustilo-Anderson open fracture classification systems. Kaplan-Meier estimators were utilized to obtain point estimates, and the log-log transformation approach was utilized to calculate 95% confidence intervals (CIs) for outcome rates.

A total of 6,042 open fractures were included. The cumulative SSI rates at 12 months for Gustilo-Anderson Types 1, 2, 3A, 3B, and 3C were 5.1%, 9.7%, 13.8%, 28.9%, and 26.2%, respectively. The cumulative rates of delayed union/nonunion at 12 months for Gustilo-Anderson Types 1, 2, 3A, 3B, and 3C were 3.0%, 5.2%, 8.0%, 14.0%, and 17.0%, respectively. Utilizing the OTA/AO fracture classification to increase the point estimate granularity, the estimated 12-month SSI and delayed union/nonunion rates in 156 Gustilo-Anderson type 3B open tibial shaft fractures (OTA/AO 42) were 34.7% (95% CI, 26.7% to 41.9%) and 18.4% (95% CI, 12.0% to 24.4%), respectively. A companion website with SSI and delayed union/nonunion rates was developed to supplement this article.

Open fractures are a substantial problem with complications that include infection and nonunion. The present data are useful for prognosis, research study design, and informing public awareness and policy. These results show that, despite current treatment approaches, the rates of SSI and delayed union/nonunion following treatment of open fractures remain high at 1 year and are not substantially improved from historical rates spanning several decades. Although open fracture sequelae remain a burden for patients, orthopaedic surgeons, and health-care systems, there may be opportunities for improvement in outcomes.

Prognostic Level I. See Instructions for Authors for a complete description of levels of evidence.