Predictors of reoperation for adult femoral shaft fractures managed operatively in a Sub-Saharan country

Background: The optimal treatment for femoral shaft fractures in low-resource settings has yet to be established, in part, because of a lack of data supporting operative treatment modalities. We aimed to determine the reoperation rate among femoral fractures managed operatively and to identify risk factors for reoperation at a hospital in a Sub-Saharan country. Methods: We conducted a prospective clinical study at a single tertiary care center in Tanzania, enrolling all skeletally mature patients with diaphyseal femoral fractures managed operatively from July 2012 to July 2013. Patients were followed at regular intervals for 1 year postoperatively. The primary outcome was a complication requiring reoperation for any reason. Secondary outcomes were scores on the EuroQol (EQ)-5D, radiographic union score for tibial fractures (RUST), and squat-and-smile test. Results: There were a total of 331 femoral fractures (329 patients) enrolled in the study, with a follow-up rate at 1 year of 82.2% (272 of 331). Among the patients with complete follow-up, 4 injuries were managed with plate fixation and 268 were managed with use of an intramedullary nail. The reoperation rate for plate fixation was 25% (1 of 4) compared with 5.2% (14 of 268) for intramedullary nailing (p = 0.204). As found in a multivariate logistic regression, a small nail diameter, a Winquist type-3 fracture pattern, and varus malalignment of proximal fractures were associated with reoperation. The mean EQ-5D score at 1 year was 0.95 for patients who did not require reoperation compared with 0.83 for patients who required reoperation (p = 0.0002). Conclusions: Intramedullary nailing for femoral shaft fractures was associated with low risk of reoperation and a nearly full return to baseline health-related quality of life at 1 year of follow-up. There are potentially modifiable risk factors for reoperation that can be identified and addressed through education and dissemination of these findings. Level of Evidence: Prognostic Level I. See Instructions for Authors for a complete description of levels of evidence.