UCSF

Background

Infection remains a costly, devastating complication following the treatment of open fractures. The appropriate timing of debridement is controversial, and available evidence has been conflicting.

Methods

This study is a retrospective analysis of the SIGN (Surgical Implant Generation Network) Surgical Database (SSDB), a prospective registry of fracture cases in predominantly low-resource settings. Skeletally mature patients (≥16 years of age) who returned for follow-up at any time point after intramedullary nailing of an open femoral or tibial fracture were included. Patients were excluded if they had delays in debridement exceeding 7 days after the injury. Utilizing a model adjusting for potential confounders, including patient demographic characteristics, injury characteristics, country income level, and hospital type and resources, local logistic regression analysis was performed to evaluate the probability of infection with increasing time to debridement in 6-hour increments.

Results

In this study, 27.3% of patients met the eligibility criteria and returned for follow-up, with a total of 10,651 fractures from 61 countries included. Overall, the probability of infection increased by 0.17% for every 6-hour delay in debridement. On subgroup analysis, the probability of infection increased by 0.23% every 6 hours for Gustilo-Anderson type-III injuries compared with 0.13% for Gustilo-Anderson type-I or II injuries. The infection risk increased every 6 hours by 0.18% for tibial fractures compared with 0.13% for femoral fractures.

Conclusions

There was a linear and cumulative increased risk of infection with delays in debridement for open femoral and tibial fractures. Such injuries should be debrided promptly and expeditiously. The size and international nature of this cohort make these findings uniquely generalizable to nearly all environments where such injuries are treated.

Level of evidence

Therapeutic Level III . See Instructions for Authors for a complete description of levels of evidence.