Abstract
Overuse anterior knee pain (AKP) is precipitated by activity and affects up to 30% of young and active populations. There is little empirical evidence for the multitude of cited risk factors for the condition and a lack of prospective studies. The main aim of this PhD was to examine the role of gait kinematics as a risk factor for AKP. The first study examined variables other than gait that may need to be controlled or statistically adjusted for in future studies to avoid masking true risk factors or effects. A prospective study of military recruits was undertaken into the effect of prior activity levels, aerobic fitness and social and medical history on the development of AKP. The incidence of AKP was high (8.6%; 95% CI: 6.8-10.4) despite the short 12-week exposure to training. Heavy smokers (odds ratio (OR): 6.37) and individuals with a previous ankle injury (OR; 2.48) had an increased risk of AKP that was independent of lifestyle factors. The association between 3D gait kinematics and patellofemoral pain syndrome (PFPS) was then explored prospectively. Principle components analysis was applied to reduce the gait data into its main factors and multivariate logistic regression was used to explore the association between these factors and PFPS. Three factors from treadmill running explained 47% of the variance between individuals who developed PFPS and those that remained injury-free. These factors contained increased hip and tibial internal rotation, increased hip adduction and decreased knee internal rotation during stance. These results contradicted findings from case-control studies. The association between variability in gait movement patterns and PFPS was assessed using the continuous relative phase method. The main risk factor was reduced inter-stride variability in the joint coordination relationships that contained tibial rotation. The main limitation of the gait study was the small sample size of the PFPS group (n=7). A study was thus undertaken to cross-validate the findings in a new sample with PFPS. This new sample was captured using a 3-year follow up study of the original gait cohort. The results were not replicated in the new PFPS group, and there were no other gait characteristics correlated with PFPS. The lack of validation was attributed to differences in the symptom-complex between the case groups of the two studies. To date, all published evidence for an association between gait and AKP originates from case-control studies. The key issue with this design is inferring the correct temporal sequence of a finding. Thus, to assess the effect of PFPS on gait and inform the interpretation of these studies, a repeated measures study of 6 subjects before and after the onset of PFPS was undertaken. Despite the mild symptoms of the group and the absence of pain during testing, the subjects showed some subtle gait inhibition post onset of PFPS. This questions the use of the case-control study to validly quantify risk factors in gait. Future research should cross-validate the significant risk factors found' in these studies, explore other potentially salient variables such as patellofemoral alignment and examine the causes of these risk factors. It is hoped that such work will benefit the prevention and treatment of AKP.