Syndesmotic Ankle Injuries: A Comparison Of Etiology, Mechanism Of Injury, And Return-to-Play Time In NFL And NHL Athletes

OBJECTIVES: The purpose of our study is to examine the differences in mechanism of injury, severity, and diagnosis of syndesmotic injuries and determine if the variation among these factors is associated with return-to-play time among US professional football and hockey players. Notably prior literature indicates an unexplained disparity in time lost to injury for syndesmotic pathologies: Wright et al. (2004) reported a mean return-to-play time of 45 days for 14 examined NHL players, whereas Osbahr et al. (2013) reported a mean return-to-play time of 15 days for 36 examined NFL players. Specifically we posit that patterns of injury, verified by expert MRI evaluation, account for the greater time losses reported among NHL athletes, when compared to NFL athletes. METHODS: We reviewed the injury databases of the Pittsburgh Penguins and Pittsburgh Steelers in order to identify athletes who incurred syndesmotic injuries (without co-existing pathology) within the last 10 years. To be eligible for the study, subjects must have received diagnostic MRI imaging within 7 days of injury and no prior history of ipsilateral syndesmotic injury. We then extracted the following data: return-to-practice time (days), age, height, weight, position within sport, date of injury, side of injury, proportion of season remaining, mechanism of injury, and MRI studies. The MRI studies evaluated injuries to the AITFL, PITFL, interosseous membrane, and interosseous ligament, as well as edema height and syndesmotic widening. We generated descriptive statistics of the outcome variables and predictors, including frequencies and proportions for categorical variables and measures of central tendency and dispersion for continuous variables. We assessed univariable measures of association of the predictors with each sport, using the appropriate non-parametric tests for small sample sizes (Mood’s median test and Fisher’s exact test for continuous and categorical predictors, respectively). We then tested differences in return-to-play by sport using survival analysis methods using Kaplan-Meier Estimators and later Cox Proportional Hazards (CPH) Regression Models, allowing us to model time-to-return as a function of multiple continuous and discrete predictors. RESULTS: The initial investigation revealed nine professional hockey and thirty-three professional football players that suffered injuries over the course of the ten-year study horizon. The median return-to-play times were 34 days (mean=35.2, range=1-79) and 11 days (mean=14.8, range=1-68) for hockey and football athletes, respectively. Non-parametric univariable analysis indicated preferential involvement of the PITFL (p<.05) and minimal syndesmotic width widening (p<.05) among NHL athletes. Sport-specific survival curves indicated statistically significant differences in return-to-play time between the second and fourth weeks of recovery, further supported by the statistical significance of sport (p<.05) as a predictor of return-to-play time in the regression analyses. CONCLUSION: Our study, unique in its direct access to injury databases of two professional teams from different sports, confirms a known disparity in recovery times among professional athletes. Specifically the results suggest that the preferential involvement of the PITFL combined with a lesser degree of syndesmotic widening in NHL athletes may account for the greater time losses, when compared to NFL athletes.