Correlates of hand abnormalities and measures of hand pain and function in older adults

Purpose: Ultrasound is increasingly used in clinical practice, yet associations between individual ultrasound features of hand osteoarthritis and clinical findings remain unclear. Therefore, we aimed to describe cross-sectional associations between abnormal features seen on ultrasound and clinical joint examination and measures of hand pain, physical function limitation, stiffness, and grip strength in a cohort of community-dwelling older adults (n=519). Methods: Joints in both hands were clinically examined for presence of tenderness, soft tissue swelling, hard tissue enlargement (nodules) and deformity using American College of Rheumatology (ACR) criteria for hand OA. Joints in the dominant hand was assessed by ultrasound for osteophytes, grey-scale synovitis, and Power Doppler Imaging (PDI) synovitis using established protocols. Hand pain, function, and stiffness were assessed using visual analogue scale (VAS) score and Australian/Canadian hand osteoarthritis index (AUSCAN). Grip strength was assessed using dynamometer. Data was analysed using hurdle and linear models, adjusted for age, sex, BMI, and other clinical or ultrasound features. Results: Participants with pain had higher proportions of women and ACR-defined HOA, had weaker grip strength, and had more clinical assessment and ultrasound features of HOA compared to participants without pain. Ultrasound-visualised osteophytes and grey-scale synovitis were ubiquitous (100% had osteophytes and 99% had grey-scale synovitis on at least one joint), therefore the presence of osteophytes and grey-scale synovitis were dichotomised at ≥2, while PDI synovitis was common (33%) in this population. Number of tender joints was associated with greater pain (target hand pain on VAS, β=1.57 (95% confidence interval; 1.11, 2.03); AUSCAN pain, β=10.57 (4.00, 17.13)) and self-reported physical function limitation (β=4.07 (1.28, 6.86)) and poorer grip strength (β=-0.07psi (-0.14, -0.005)); associations with function were partially mediated by pain. Number of joints with PDI synovitis was associated with greater pain intensity (VAS, β=2.61 (1.03, 4.19); AUSCAN pain, β=13.07 (3.82, 22.32)) but not function and stiffness whereas grey-scale synovitis was independently associated with poorer grip strength (β=-0.22 (-0.41, -0.04)) but not pain. Number of deformed joints was associated with poorer function (AUSCAN function, β=4.51 (1.75, 7.26); grip strength, β=-0.13 (-0.23, -0.04)) but not pain and stiffness, Nodules (β=0.27 (0.08, 0.46)) and osteophytes (β=0.51 (0.03, 0.99)) were independently associated with stiffness. Further adjustment of ultrasound features for clinical features (and vice versa) did not significantly change the associations, except for a significant reduction in the effect between nodules and stiffness when further adjusted for osteophytes. Conclusions: Tender, deformed, nodulous joints on hand examination and ultrasound-assessed PDI synovitis and grey-scale synovitis were independently associated with hand pain, physical function or grip strength cross sectionally. Associations with poor physical function were predominantly mediated through pain; however, tenderness and deformity affected function even after taking pain into account. Therefore, interventions to treat synovitis may improve hand pain, but preventing deformity and pain may be required to improve poor hand function.