Baseline bone marrow lesion volume and bone marrow lesion regression are associated with knee osteoarthritis progression: data from the osteoarthritis initiative

Abstract

Listen

Translate article

gait between asymptomatic individuals and symptomatic individuals diagnosed with moderate knee OA, all with Kellgren-Lawrence (KL) osteoarthritis (OA) radiographic grade 2 or 3. Methods: Individuals withmoderate medial compartment knee OA and asymptomatic adults underwent gait testing. Three-dimensional joint angles and net external knee joint moments of the affected (OA) or randomized (asymptomatic) knee were calculated from a synchronized motion capture and force platform system during self-selected speed gait. KL grades of standard anterior-posterior knee radiographs were scored by an orthopaedic surgeon (WDS). Principal component analysis (PCA) extracted major patterns of variability from the angle and moment waveforms individually (i.e. 6 PCA analyses). Participant waveforms were projected onto each principal component to calculate a set of PC scores. Asymptomatic and symptomatic PC scores were compared using an ANOVA. Results: Twenty-seven asymptomatic individuals and 37 symptomatic individuals diagnosed with moderate OA had a KL score of 2 or 3. There were no significant group differences in age, or in any features of the frontal and transverse plane knee angles. The asymptomatic group walked with a significantly faster self-selected walking speed, had smaller BMI, and lower WOMAC and higher SF-36 total and physical function scores (Table 1). The only significant kinematic difference between the groups was in the overall magnitude of the knee flexion angle (PC1, p 1⁄4 0.0006), with the asymptomatic group having higher flexion angles. The asymptomatic group had lower overall magnitude of the knee adduction moment (PC1, p <0.0001), and greater difference between the first peak and themid-late stance adductionmoment (PC2, p 1⁄4 0.05) than the OA group. The asymptomatic group had earlier peak extension moments prior to toe-off (PC3, p <0.0001), as well as more constant internal rotation moment (PC1, p <0.0001) and earlier peak rotation moments (PC3, p 1⁄4 0.01) during stance (Figure 2). Conclusions: It was interesting that twenty-seven of 28 asymptomatic individuals tested had radiographic evidence (KL 2,3) of OA in their knee joints. Compared to the gait patterns of those diagnosed with moderate OA with the same radiographic scores, the asymptomatic group walked faster with higher overall knee flexion angles, earlier peak extension moments, lower and less constant knee adduction moment loading and more constant knee rotation moment loading during gait. The lower overall adduction moment would suggest that high adduction moments may be a symptomatic feature of OA and not an early mechanism of OA development. The sagittal and frontal plane pattern differences have been previously associated with higher walking speeds and may be a reflection of a less symptomatic joint. The altered rotation moment pattern difference is not speed or function related according to previous study and may support hypotheses that altered rotational joint mechanics is a factor in early knee OA development, prior to the onset of symptoms.