OBJECTIVE: To examine the association of pain sensitization with knee joint loading during walking in people with knee osteoarthritis (OA). METHOD: For this cross-sectional study, we used baseline data from participants with symptomatic knee OA (n=104) enrolled in two clinical trials. We used pressure pain threshold (PPT) at the knee and wrist to assess sensitization. Using gait analyses during walking, we derived peaks and impulse of knee adduction moment (KAM) and knee flexion moment, and frontal and sagittal plane range of motion during stance. We used Analysis of Covariance (ANCOVA) adjusted for age, sex, body height, and body weight to examine associations of PPT (sex-specific tertiles) with gait outcomes. Sensitivity analyses included additional adjustment for gait speed. RESULTS: For knee PPT, lowest tertile had lower 1st peak KAM (∆=9.1 Nm [95% CI
0.3, 17.9]) compared to highest tertile. For wrist PPT, compared to highest tertile, lowest tertile had lower - 1st peak KAM (∆=9.1 Nm [1.3, 16.8]), 2nd peak KAM (∆=7.5 Nm [0.5, 14.6]), KAM impulse (∆=1.7 Nm*s [0.1, 3.2]), and middle tertile had lower - 2nd peak of KAM (∆=8.0 Nm [0.7, 15.4]) and KAM impulse (∆=2.0 Nm*s [0.4, 3.6]). The effect sizes for other gait measures were small and clinically not meaningful. The effect sizes were similar after adjusting for gait speed. CONCLUSION: Greater pain sensitivity as assessed with PPT was related to lower frontal plane joint loading during walking. These findings may reflect a motor adaptation to nociceptive alterations in this population.