Prediction of outcome after ankle fracture

Study Design: Prospective inception cohort study of 62 consecutive patients seen in 2 orthopaedic clinics following ankle fracture. Objectives: To investigate 4 putative predictors (age, fracture classification, acute management [surgical or nonsurgical], and ankle dorsiflexion range of motion measured at the time of cast removal) of outcome after ankle fracture and to develop simple predictive models of outcome after ankle fracture. Background: Ankle fracture is a common condition. However, few studies have investigated factors that predict outcome after ankle fracture. Methods and Measures: Sixty-two consecutive subjects aged 17 or older with ankle fractures were recruited from 2 hospital orthopaedic clinics. Outcome measures recorded at 6 weeks and 6 months after cast removal included 2 functional questionnaires, patients' ratings of global improvement, and a measure of ankle dorsiflexion. The predictive value of 4 variables selected a priori was analyzed using bivariate and stepwise multiple linear regression. Results: Ankle dorsiflexion and fracture classification predicted outcome 6 weeks and 6 months after cast removal for all outcome measures used (P<.05, r² = 0.09-0.47). Fracture management (surgical or nonsurgical) inconsistently predicted outcome at both 6 weeks and 6 months, and age did not predict outcome at either 6 weeks or 6 months after cast removal. The predictive models explain between 19% and 58% of the variance in outcomes 6 weeks after cast removal and 19% to 52% of the variance in outcomes 6 months after cast removal. Conclusion: Ankle dorsiflexion measured at the time of cast removal and fracture classification are clinically significant predictors of outcome after ankle fracture; however, much unexplained variation in outcomes still exists.