Comparison of Prevalence of Osteoarthritis in the Hand: A Multicenter Retrospective Cohort Study

Introduction Several studies have previously been undertaken to investigate the prevalence of radiologic hand osteoarthritis (OA) in Caucasian, Japanese, Chinese, and Arabic populations. To date, there has been no direct international comparison study on the prevalence of radiologic hand OA between a Western and a Southeast Asian population. We hypothesize that there is difference in the pattern of joint involvement among individuals of both populations. Materials and Methods Consecutive hand radiographs from individuals aged 60 years and above were reviewed. Radiographic evidence of osteoarthritis in the various joints in the hands was graded using Kellgren-Lawrence (K-L) score. Chi-square test, Fisher's exact test, or Student's t-test was used as appropriate. Multiple logistic regression analysis was performed to evaluate the associations and relationships of radiographic hand OA between joints. Results A total of 194 patients exhibited radiographic OA. Singaporean patients were more likely to have radiographic OA of the thumb interphalangeal joint (IPJ) (65.1%) compared with the thumb carpometacarpal joint (CMCJ; 40.2%); British patients were more likely to have CMCJ (43.3%) OA than thumb IPJ (21.0%). The difference was statistically significant (p = 0.00026). In the Singapore population, it was more likely that a patient had concurrent OA of both thumb and finger joints, whereas in the British population, this was apparent in only the above-80-year-old group. The most commonly affected joint was the left thumb IPJ (joint specific prevalence of 35%) in the Singapore population, and the right index finger distal interphalangeal joint (DIPJ; joint specific prevalence of 41.5%) in the British population. Conclusion Our study, the first study to compare between Singaporean and British populations, showed statistically significant difference in the prevalence of OA in the hands. These findings suggest ethnic and cultural factors play a part in such a distribution trend.