Purpose: Knee osteoarthritis (OA) is the most common joint disease and is often characterized by pain, stiffness, swelling and limited tolerance to weight-bearing activities. It has been shown to be associated with biological inflammatory processes in the synovium, synovitis, and mechanical insults that lead to the deterioration of the joint structures. Synovitis is usually characterized by knee joint swelling and the mechanical knee joint environment during a loading stimulus can be studied in vivo using state-of-the-art gait analyses. The study objective is to compare knee mechanics and muscle activation patterns during gait in individuals with knee OA who have and do not have knee joint swelling. Methods: 45 patients diagnosed with moderate medial compartment knee OA were recruited. Knee swelling was measured at three locations (mid, medial, lateral) of the supra-patellar recess (SRD) using ultrasound imaging. A maximum depth, regardless of the location was determined. Effusion presence was defined as a SRD depth >4 mm. Knee joint motion was calculated from skin markers and net external moments calculated through inverse dynamics during walking on a dual-belt instrumented treadmill at self-selected speed. Electromyography (EMG) from vastus medialis (VM) and lateralis (VL), medial (MH) and lateral hamstrings (LH), medial (MG) and lateral gastrocnemius (LG) was recorded using standardized procedures. Two-sample unpaired t-tests were used to test for significant differences in sagittal plane knee joint angles, sagittal and frontal external moments, pain, Knee Osteoarthritis Outcome Scores, age, body mass index, stride characteristics, and knee extensor and flexor strength. Principal component analysis (PCA) was performed on the EMG waveforms to capture amplitude and temporal based features. Principal Pattern (PP)-scores were calculated from principal patterns that together explained >90% of the waveform variability. Two-way mixed-model Analysis of Variance models were performed on the PP-scores to identify muscle, group and interaction effects. Bonferroni corrected pairwise post-hoc comparisons were made. A P-value of 0.05 was used to determine statistical significance. Results: Of the 45 participants, 25 had effusion (SRD > 4mm). No significant differences were found between groups in subject demographic and anthropometrics, questionnaire outcomes, and knee strength (P > 0.05). Both groups had a similar radiographic grade distribution. No biomechanical differences existed between both groups, however the effusion group had greater overall amplitudes of VM and VL (PP1-scores) and greater early stance activation of the gastrocnemius muscles (PP2-scores). No other group differences or group by muscle interactions existed. Conclusions: These results, as shown in Figure 1, indicate that specific amplitude knee joint muscle activation patterns are altered in the presence of knee effusion during gait in individuals with knee OA. Greater stance phase amplitudes in the quadriceps and gastrocnemius muscles are consistent with responses aimed at providing neuromuscular knee joint support and increasing early stance active stiffness, respectively, in order to preserve knee function when effusion is present. These muscular adaptations have been previously reported with increased OA severity suggesting a possible link between gait mechanics and knee swelling as knee OA progresses. The 4 mm SRD threshold has previously been linked to a higher risk of knee OA progression, supporting findings of this study. Future studies can investigate different cut-off values for the detection of effusion sonographically to establish whether biomechanical subgroups may exist.View Large Image Figure ViewerDownload Hi-res image Download (PPT)
Read full abstract