Pain detect questionnaire and pain catastrophizing scale affect gait pattern in patients with knee osteoarthritis

Purpose Although pain phenotype affects clinical score in patients with knee osteoarthritis (OA), little information has been available on the relationship between pain phenotype and gait analysis. The purpose was to investigate the relationship between pain phenotype and gait parameters. Methods A total of 34 patients (24 females and 10 males) with end‐stage medial compartmental knee OA participated. All the patients were evaluated based on pain detect questionnaire (PD‐Q) and pain catastrophizing scale (PCS). They were divided into two categories: Group Low (PD‐Q score ≤ 12) and Group High (PD‐Q score > 12), PCS + (PCS ≥ 23) and PCS‐ (PCS < 23). Gait analysis was performed using three‐dimensional motion analysis system. Statistical analysis was done to compare gait parameters between groups for each allocation of PD‐Q or PCS, separately. Results Peak vertical ground reaction forces in Group Low and High were 0.99 ± 0.054 and 0.82 ± 0.17, respectively ( P  = 0.015). Peak knee adduction moments in Group Low and High were 0.70 ± 0.19 and 0.39 ± 0.14, respectively ( P  = 0.0022). For PCS allocation, knee extension limitation during mid‐stance during gait were significantly larger in PCS‐ ( P  = 0.038). Conclusions Patients with high PD‐Q score had atypical gait pattern with smaller peak vertical ground reaction force and knee adduction moment, compared to patients with low PD‐Q score. Moreover, patient with low PCS had different gait pattern in extension limitation, compared to those with high PCS. PD‐Q and PCS would affect gait pattern in patients with knee OA. Level of evidence: III.