Delays to Surgery and Coronal Malalignment Are Associated with Reoperation after Open Tibia Fractures in Tanzania

摘要

Background Treatment of diaphyseal open tibia fractures often results in reoperation and impaired quality of life. Few studies, particularly in resource-limited settings, have described factors associated with outcomes after these fractures. Questions/purposes (1) Which patient demographic, perioperative, and treatment characteristics are associated with an increased risk of reoperation after treatment of open tibia fractures with intramedullary nailing or an external fixation device in Tanzania? (2) Which patient demographic, perioperative, and treatment characteristics are associated with worse 1-year quality of life after treatment of open tibia fractures with intramedullary nailing or an external fixation device in Tanzania? Methods A prospective study was completed in parallel to a similarly conducted RCT at a tertiary referral center in Tanzania that enrolled adult patients with diaphyseal open tibia fractures from December 2015 to March 2017. Patients were treated with either a statically locked intramedullary nail or external fixator and examined at 2 weeks, 6 weeks, 3 months, 6 months, and 1 year postoperatively. The primary outcome, reoperation, was any deep infection or nonunion treated with a secondary intervention. The secondary outcome was the 1-year EuroQol-5D (EQ-5D) index score. There were 394 patients screened and ultimately, 267 patients enrolled in the study (240 from the primary RCT and 27 followed for the purposes of this study). Of these, 90% (240 of 267) completed 1-year follow-up and were included in the final analysis. This group comprised 110 patients who underwent IMN and 130 who had external fixation; follow-up was similar between study groups. Patients were an average of 33 years old and were primarily males who sustained road traffic injuries resulting in AO/Orthopaedic Trauma Association (OTA) classification type A or B fractures. There were 51 reoperations. For the purposes of analysis, all patients were pooled to identify all other factors, in addition to treatment type, associated with increased risk of reoperation and 1-year quality of life. An exploratory bivariable analysis identifying various factors associated with reoperation risk and EQ-5D was subsequently included in a multivariate modeling procedure to control for confounding of effect on our primary outcome. Multivariable modeling was performed using standard hierarchical modeling simplification procedures with log-likelihood ratios. Alpha levels were set to 0.05. Results After controlling for potentially confounding variables such as gender, smoking status, mechanism of injury, and treatment type, the following factors were independently associated with reoperation: Time from hospital presentation to surgery more than 24 hours (odds ratio 7.7 [95% confidence interval 2.1 to 27.8; p = 0.002), AO/OTA fracture classification Type 42C fracture (OR 4.2 [95% CI 1.2 to 14.0]; p = 0.02), OTA-Open Fracture Classification muscle loss (OR 7.5 [95% CI 1.3 to 42.2]; p = 0.02), and varus coronal angle on an immediate postoperative AP radiograph (OR 4.8 [95% CI 1.2 to 14.0]; p = 0.002). After again controlling for confounding variables such as gender, smoking status, mechanism of injury, and treatment type factors independently associated with worse 1-year EQ-5D scores included: Wound length >= 10 cm (ss = [change in EQ-5D score] -0.081 [95% CI -0.139 to -0.023]; p = 0.006), OTA-Open Fracture Classification muscle loss (ss = -0.133 [95% CI -0.215 to -0.051]; p = 0.002), and OTA-Open Fracture Classification bone loss (ss = -0.111 [95% CI -0.208 to -0.