Clinical evaluation and quantitative analysis of knee varus thrust: A reliability and validity study

Abstract

Purpose: Identified by clinical observation, a varus thrust is defined as a dynamic bowing-out of the knee in the lateral direction during the weight bearing phase of gait and accompanied by return to a less varus alignment during the swing phase of gait. A varus thrust may further increase medial knee load of the knee joint and therefore be of detriment to people with medial knee osteoarthritis (OA). Varus thrust has been identified in 17-31% of individuals with knee OA and has been associated with a 4-fold increased likelihood of radiographic disease progression. As varus thrust has recently gained attention, little is known about inter-rater reliability of this clinically-observed test and the ability to quantify this risk factor using clinically available video technology. The aim of this study was to assess the intra-rater and inter-rater reliability of the clinical test to identify the presence of knee varus or valgus thrust. We also aimed to evaluate if varus or valgus thrust can be quantified using two-dimensional video analysis in older healthy adults. Methods: Fifty participants with no clinical signs of knee OA, aged 60-79, were recruited from the community. Participants walked barefoot at their comfortable walking speed along a 10 meter walkway, at the end of which a video camera was positioned. The limb was exposed for video recording and the dominant limb was the index knee with markers placed anteriorly on the anterior superior iliac spine, patella and midway between medial and lateral malleoli. Each participant completed five gait trails with the three middle trails used for quantifying knee motion. To describe the knee function of the sample, participants completed the Knee Injury and Osteoarthritis Outcome Score (KOOS). Clinically observed varus or valgus thrust was scored by two examiners who reviewed videos of over two sessions and the results were analysed for intra-rater and inter-rater reliability using the Kappa coefficient. Consensus on thrust presence classifications for all participants was reached to validate the quantitative measure of thrust using two-dimensional video analysis (P&O Clinical Movement Data, Siliconcoach Ltd, Dunedin, NZ). Video analysis was conducted by manually locating the markers and were used to define the hip-knee-ankle (HKA) angle. The change in HKA angle between initial contact and mid-stance was used to evaluate validity of this measure. We conducted a receiver operator curve (ROC) to determine the accuracy of the quantitative measure in distinguishing thrust presence. Results: Intra-rater reliability in the clinical evaluation of thrust demonstrated substantial agreement (p<0.001, k=0.74-0.84). Inter-rater reliability revealed moderate agreement between observers (k=0.77, p<0.001). After reaching consensus on classification of thrust, 70% of the participants demonstrated no thrust, 26% had a varus thrust and 4% had a valgus thrust. Quantitative analysis demonstrates that participants with no thrust had a mean HKA angle of 1.13° (SD = 1.1), those with varus thrust had 3.21° (SD = 0.8) and two participants with valgus thrust had -0.8° (SD = 1.7). We identified a statically significant difference (p<0.001) in the HKA angle change between participants who had no thrust and those with varus thrust. There were no statistically significant differences in participant characteristics between thrust classifications. The change in HKA angle was found to be 90.2% accurate in distinguishing participants with varus thrust and those with no thrust (p<0.001). Conclusions: Clinical evaluation of varus thrust demonstrates moderate to substantial reliability. Both clinical evaluation and quantitative analysis using simple readily available tools can be used to classify and quantify a varus thrust. Both methods of assessment have the potential to be easily transferred into the clinical setting and be used as a predictive tool or to objectively monitor severity of varus thrust.