Poster 124: Intrinsic Anthropometric Factors are Associated with Bone Stress Injuries in Collegiate Distance Runners: New Risk Metrics & Screening Tools?

Abstract

Objectives:

Lower limb bone stress injuries (BSI) of the pelvis, femur, and tibia are prevalent in collegiate track & field distance runners. However, it is not known the population’s bone, body composition (BComp), and anthropometric parameters prior to first respective collegiate injury compared to non-injured counterpart. This respective injury has been reported to require an extended timeframe for recovery as well as increases the risk, prospectively, for sustaining another similar injury. Extrinsic factors such as greater running frequency, mileage, pace, and change in terrain is known to influence risk which, during a competitive season, commonly occurs in collegiate runners. Intrinsic risk factors related to bone health and body composition of the runner have also been identified as important factors that further influence risk indicating proper screening of these respective intrinsic factors are necessary and recommended in order to better identify risk and prompt further evaluation. In light of previous observations and a clinical need in this unique population, the purpose of this study was to characterize bone mineral density, body composition, and skeletal dimensions assessed vai dual energy x-ray absorpitometry (DXA) in collegiate male and female distance runners prior to lower limb BSI status (i.e. injured or non-injured) during a competitive season as well as provide indices of risk specific DXA assessment. We hypothesized that 1) collegiate distance runners with a lower limb BSI will exhibit lower BMD relative to runners without injury; 2) total and regional body composition (i.e. fat and lean) as well as skeletal dimension differences will be observed between respective BSI status groups thus, allowing for identification of potential threshold cutoffs for this population; 3) measures of body composition and skeletal dimensions could be used to develop prediction models for regional and total BMD that may be utilized in the abscence of DXA for those with out readily available access.

Methods:

This was a retrospective case-control study design that was conducted at a University athletic training facility during the fall (August-November) 2013-2019 cross country seasons. Study measurements included a total body (TB) DXA scan, weight (kg), height (cm), and body mass index (BMI; kg/m²). The inclusion scan for the study participants with a lower limb BSI was the scan before first respective injury of collegiate career while study participants with no injury during collegiate career had their last scan of eligible fall season included. This study included male and female distance running athletes who were members of a NCAA D1A collegiate track and field team at a single University. Inclusion criteria included participation in sport-related activity prior to and after DXA scan for three months as well as only scans prior to first collegiate lower limb BSI. Exclusion criteria included any bone-related injury or recovery from such injury at time of DXA scan that prohibited prospective sport-related activities as well as scans prior to reoccurring lower limb BSI. Enrolled runners completed a DXA scan (August-November) and participated in sport activities before scan. Three months following scan, electronic medical records were reviewed to determine whether or not a BSI occurred. Independent samples t-test was used to compare BMD, BComp, and anthropometric measures between BSI vs non-injured (included subgroup analysis by sex). Multiple linear regression with stepwise removal was used to determine measures most predictive of BMD. Significance was set at p<0.05.

Results:

Demographics are shown in Table 1. Figure 1 shows athlete BMD characteristics among athletes (i.e. all, males, and females) with or without lower limb BSI. Overall, those in the NoBSI group had significantly higher BMD values for total, spine, pelvis, and legs (p<0.001). In males, the NoBSI group had significantly higher total, pelvis, and leg BMD (p<0.05) while no difference was observed in spine BMD compared to BSI group. Within females, the NoBSI group had significantly higher BMD in all respective measures compared to BSI group (p<0.001). Regression analysis revealed BComp and anthropometric measures were predictive of total bone mass [BM = (0.046 x age, yrs) + (0.024 x weight, kg) + (0.014 x %Fat) + (-0.017 x ARML) + (0.017 x SHWIDTH) + (-0.009 x TRNKL) + (0.037 x LL)]; P<0.05, R² = 0.61 g/cm², percent error = 11.01). In addition to predicted BM, as shown in Figure 2, other BComp and anthropometric measures [age, height, weight, BMI, FM, %Fat, BM, LL, SHWIDTH, TRNKL, and ARML) were predictive of total and regional BMD (P<0.05, R² = 0.64-0.80 g/cm², %error = 3.8-4.8). A physiologic comparison of those with and without long bone BSIs is shown in Table 2. No differences were found between BSI and NoBSI group demographics (i.e. height, weight, age, and BMI) in all athletes including males and females. BComp values in males found FM to be significantly lower in BSI compared to NoBSI while women leg LM in BSI group was significantly lower than NoBSI. Among all athletes, the BSI group had significantly shorter arm and leg limb lengths as well as a lower leg/trunk ratio compared to NoBSI group.

Conclusions:

Lower limb bone stress injuries are common in collegiate distance runners wherby total and regional BMD measures tend to be lower compared to runners who were not injured. Shorter limb lengths, lower fat mass (male), and lower leg lean mass (female) may also be indicative of BSI risk in this population. Interestingly, BComp and anthropometric measures are highly predictive of BMD in this sport population. These metrics may serve as sport-specific benchmarks for lower limb BSI risk and the equations developed here may be suitable for BMD screening in collegiate distance runners without readily availalbe access to DXA.

Figure 1.

BMD Comparison. Data are presented as means +/- 95% CI for athlete bone mineral density (BMD) with long bone stress injury (BSI) or no long hone stress injury (NoBSI). Type I error set at α = 0.05. For each significant comparison, effect size is presented as Cohen’s d statistic, whereby 0.1 indicates negligible effect size; 0.1 to 0.3, small effect size; 0,3 to 0.5, moderate effect size; 0.5 to 0.7, large (L.) effect size; and 0.0.7, very large (VL) effeel size. NS, not significant.

Figure 2

Prediction of Total and Regional BMD, Data are presented as plots with r², standard error (SEE), and percent error (%error) values for results of multiple linear regression alysis predict total and regional BMD (n - 79). All predictions presented are significant at p <0.05. Coefficient abbreviations: age (in year), height (HT; in centimeters), weight (in kilograms; KG), body mass index (BMI; weighv/ht²), percent fat (%Fat), fat mass (FM; in kilograms), lean mass (LM; in kilograms), total bone mass (BM; in kilograms), leg length (LL; in centimeters), shoulder width (SHWIDTH; in centimeters), trunk length (TRNK1; in centimeters), and arm length (ARML; in centimeters).

Table 1.

Athlete Demographics

Table 2.

Comparison Between Those With and Without Bone Stress Injury (BSI)