Biomechanical predictors of retrospective tibial stress fractures in runners

https://doi.org/10.1016/j.jbiomech.2008.02.001Get rights and content

Abstract

Both kinematics and kinetics of the lower limb have been shown separately to be related with a history of tibial stress fractures (TSFs) in female runners. However, it is likely that these factors interact together to increase the risk of a TSF. This study was conducted to determine which combination of kinematic and kinetic factors are the best predictors of retrospective TSF in female distance runners. Total 30 female runners who had previously sustained a TSF were recruited, along with an age and mileage matched control group (n=30). Subjects ran overground at 3.7 m/s while kinematic and kinetic data were recorded. Five trials from each subject were used for data analysis and ensemble means were calculated for both groups. The kinematic variables of peak hip adduction (HADD), peak knee internal rotation (KIR) and knee adduction (KADD), peak rearfoot eversion (RFEV) were entered into a binary logistic regression along with the kinetic variables of vertical instantaneous load rate (VILR) and absolute free moment (FM). The variables HADD, FM and RFEV were able to correctly predict a history of TSF in 83% of cases. Increases in HADD, FM and RFEV (odds ratios of 1.29, 1.37 and 1.18) were associated with an elevated risk of having a history of TSF. The addition of VILR, KIR and KADD did not improve the ability to predict previous injury. Based on these results, HADD, FM and RFEV appear to be the most important of the variables of interest in terms of predicting retrospective TSF in female runners.

Introduction

Stress fractures are a common problem among runners and may account for between 6% and 14% of running related injuries (James et al., 1978; McBryde, 1985;Taunton et al., 2002). The tibia is the bone most likely to be affected, accounting for between 35% and 56% of all stress fracture injuries (Matheson et al., 1987; Romani et al., 2002). In addition, stress fractures are reported to be more prevalent in females (Arendt et al., 2003; Taunton et al., 2002). It has been postulated that this may be due to females having a lower percentage lean body mass in the lower limb, a history of menstrual disturbance, a low fat diet and lower bone density compared to males (Bennell et al., 1999).

The etiology of stress fractures is believed to be multifactorial in nature arising from issues related to physiology, training, structure and diet (Bennell et al., 1999; Romani et al., 2002). Nevertheless, there are a number of recent studies linking retrospective tibial stress fractures (TSFs) with running mechanics. In a cross-sectional study, Milner et al. (2006b) found that the occurrence of TSF in female runners was related to greater initial loading of the lower extremity. Instantaneous and average vertical ground reaction force load rates along with vertical peak tibial acceleration were found to be significantly greater in the stress fracture group compared to healthy controls. The tibia is also likely to experience torsional loading during running. Indeed, it has been shown that runners with a history of TSF display both an increased adduction and absolute free moment (FM) in comparison to controls (Milner et al., 2006a).

Kinematic variables may also play a role in the development of stress fractures since they may alter the normal alignment of the lower extremity. For example, compared to controls, TSF runners have been shown to demonstrate increased peak hip adduction (HADD) and knee internal rotation (KIR) along with a reduction in peak knee adduction (KADD) (Milner et al., 2005). In addition, rearfoot eversion (RFEV) has been shown to be greater in athletes with running related lower leg pain (Messier and Pittala, 1988; Willems et al., 2006). Abnormal kinematics during running may alter the normal loading pattern placed on the tibia.

In summary, a number of kinetic and kinematic parameters have been related with retrospective TSF in female runners. However, it is unclear which of these factors are most strongly related to the injury. It is likely that a combination of these variables will better distinguish individuals with a history of TSF from healthy controls. Hence the aim of this study was to determine which kinematic and kinetic parameters were the best predictors of retrospective TSF in female distance runners. It was hypothesised that a subset containing both kinetic and kinematic parameters would significantly predict group membership.

Section snippets

Subjects

All subjects gave written informed consent prior to the commencement of this study. Approval for all procedures was obtained from the Human Subjects Review Board at the University of Delaware before commencing the study. Sixty female runners who were between the ages of 18 and 45 years and rearfoot strikers participated in this project. Thirty subjects with a history of TSF were compared with 30 control subjects (CON) who had no previous lower extremity bony injuries. All subjects were

Results

High colinearity (r⩾0.65) was found between VILR and both VALR (r=0.936, P<0.001) and PPA (r=0.802, P<0.001). In addition, VALR was highly correlated to PPA (r=0.681, P<0.001). Therefore, it was decided to use VILR as the input variable for the regression calculations since it provided a good representation of the VALR and PPA. None of the other variables demonstrated colinearity.

The mean peak values of VILR, FM, HADD, KIR and RFEV were all greater in the TSF group compared to the controls (

Discussion

The aim of this investigation was to determine which kinematic and kinetic factors were the best predictors of a previous TSF in female runners. We found that increases in peak HADD, peak RFEV and absolute FM increased the likelihood that a subject had sustained a previous stress fracture.

Peak RFEV was found to be 2.7° greater in the TSF group compared to the controls. This value corresponds well with the differences of 2.6° and 1.9° reported between controls and runners with exercise related

Conflict of interest

All the authors confirm that there are no financial or personal relationships with other people or organisations that could inappropriately bias the content of this paper.

Acknowledgement

This study was supported by Department of Defense Grant DAMD17-00-1-0515. This support was financial in nature and no conflicts of interest were encountered.

References (30)

  • T. Cashmere et al.

    Medial longitudinal arch of the foot: stationary versus walking measures

    Foot and Ankle International

    (1999)
  • R.H. Daffner et al.

    Stress fractures—current concepts

    American Journal of Roentgenology

    (1992)
  • Gefen, A., 2003. Consequences of imbalanced joint-muscle loading of the femur and tibia: from bone cracking to bone...
  • E.S. Grood et al.

    A joint coordinate system for the clinical description of 3-dimensional motions—application to the knee

    Journal of Biomechanical Engineering—Transactions of the ASME

    (1983)
  • K.G. Hauret et al.

    The physical training and rehabilitation program: duration of rehabilitation and final outcome of injuries in basic combat training

    Military Medicine

    (2001)
  • Cited by (0)

    View full text