Risk factors for shoulder osteoarthritis with rotator cuff tear in the elderly general population

The relationship between shoulder osteoarthritis (OA) and rotator cuff tear (RCT) is unclear. We hypothesized that there is a difference between the pathogenesis of OA complicating RCT and that of RCT complicating OA. In this study, our primary objective was to determine the prevalence of shoulder OA without RCT, RCT without OA, and OA with RCT in the general older population. Our secondary objective was to identify risk factors for the association with OA+RCT in shoulder OA alone or RCT alone, respectively.

We enrolled patients from the public health checkup conducted in Gunma prefecture (Japan) in 2014. Subjects' shoulder pain at rest, during motion, and at night was evaluated using a questionnaire. Moreover, active and passive range of motions (ROMs) in flexion, abduction, and external rotation were measured. For RCT parameters, we evaluated as no tear, partial-thickness supraspinatus (SSP) tear, full-thickness SSP tear, and SSP-infraspinatus tears. For further analysis, the shoulders were divided into three groups according to the presence of RCT and/or OA: OA, RCT, and OA + RCT groups. Risk factors for OA + RCT were identified in a logistic regression analysis.

Overall, 944 of 1148 shoulders were eligible for inclusion. The prevalence rates of shoulder OA, RCT, and OA + RCT were 5.8%, 21.1%, and 4.2%, respectively. Furthermore, 650 shoulders were excluded, and 55, 199, and 40 shoulders had OA, RCT, and OA + RCT, respectively. There were significant differences for age, ROM of active external rotation, strength of abduction, external rotation, and morphology of the rotator tears. However, there were no significant differences for pain visual analog scale score, passive ROM, Simple Shoulder Test, and grades of OA. Older age decreased active ROM in external rotation, and the presence of both subscapularis and SSP-infraspinatus tears was a risk factor for the association of OA with an RCT shoulder. Older age, weaker power in external rotation, and affected dominant side were risk factors for the association of RCT with an OA shoulder.

This study is the first to report risk factors by considering both shoulder OA and RCT in the general population. Our findings will be useful for the treatment and management of OA and RCT as well as for the prevention of these conditions in the older adults.