Abstract
CONTEXT: Adolescence is a critical period for bone development. Due to an elevated risk for low body mass index (BMI) and/or low energy availability, certain adolescent athletes are at increased risk of bone stress injury (BSI) and eating disorders (EDs). Despite this risk, the incidence of EDs in patients with BSI is unclear. OBJECTIVE: To compare the incidence of a premorbid drop in BMI percentile in a sample of adolescent athletes diagnosed with a BSI between 2005 and 2016. We also evaluated the relative risk of developing an ED for this sample, compared with a sex-matched, age-matched (±6 months), and time-matched (±2 years) sample of active adolescents with knee pain. DESIGN: Retrospective cohort study. SETTING: Data collected through the Rochester Epidemiology Project. PATIENTS OR OTHER PARTICIPANTS: One hundred eighty-seven adolescent (aged 13-18) athletes with BSI and 187 active adolescents with knee pain. MAIN OUTCOME MEASURES: Conditional logistic regression was used to assess whether premorbid weight loss was a risk factor for BSI. A stratified log-rank test was used to evaluate the association between ED diagnosis and case-control status up to 5 years postinjury. RESULTS: Patients with a BSI had 1.53 times the odds of premorbid BMI percentile drop compared with controls (95% CI = 0.83, 2.12; P = .17). A total of 14 patients with a BSI had ED diagnoses after injury compared with 8 controls (P = .13). Patients from weight-based or aesthetic sports had 1.93 times the odds of a BSI compared with controls (95% CI = 1.23, 3.02; P = .0040). The Kaplan-Meier estimates indicate that the case group had a greater risk of ED diagnosis. CONCLUSIONS: Results suggest a trend toward a greater drop in BMI percentile and increased frequency of ED diagnoses in the BSI group compared with controls. The documented co-occurrence of weight loss, stress fractures, and EDs in adolescent athletes should inform guidelines for screening and treatment of adolescent athletes with BSIs.