A Simplified Prediction Model for Lower Extremity Long Bone Stress Injuries in Male Endurance Running Athletes.

OBJECTIVE

Develop a prediction model for lower extremity long bone injuries (LBIs) in male endurance running athletes using dual-energy x-ray absorptiometry (DEXA).

DESIGN

Retrospective.

SETTING

Sports medicine department in a university athletic setting.

PARTICIPANTS

National Collegiate Athletic Association (NCAA) Division 1 white male endurance athletes (n = 27).

INDEPENDENT VARIABLES

Backward stepwise elimination was used to achieve a model that predicts LBI, by removing noncontributory variables (P > 0.10), using binary logistic regression. Independent prediction variables analyzed for model were as follows: (1) height (cm), body mass index (BMI) (kg/m), and total mass (kg); and (2) regional and total lean mass, fat mass, and bone density assessed using DEXA.

MAIN OUTCOME MEASURES

Dichotomous dependent variable was LBI.

RESULTS

Final constructed model predicted 96.3% of athletes with and without LBI. Prediction model were as follows: predict lower extremity long bone stress injury = 23.465 - 0.896 BMI + 1.043 (total upper-body mass) TUB - 34.536 leg bone mineral density (BMD). Predict lower extremity long bone stress injury is the LBI prediction, and TUB (kg) is total fat, muscle, and bone weight in trunk and arms.

CONCLUSIONS

These preliminary data suggest that Division 1 white male endurance running athletes are at risk of LBI with higher relative TUB and lower BMI in combination with a lower leg BMD.