Background Previous studies on RA features between different ethnic groups have mainly focused on disability and disease activity scores, structural damage, genetic factors and health inequalities. However, data on RA distribution of joint involvement according to ethnicity are scarce.
Objectives To investigate the joint involvement distribution and other clinical features in a cohort of ethnically diverse patients with RA.
Methods We included patients with active RA who previously failed to respond to 2 non-biologic DMARDs, and were being considered for TNFi therapy between 2001 and 2012. Data collected included tender joint counts (TJC), swollen joint counts (SJC), inflammatory markers, visual analogue scale (VAS) and DAS28 score. We examined the distribution of joint involvement and other clinical features by race, classified as Caucasian, Asian, Afro-Caribbean (AC), and other/mixed race, with the Caucasian group serving as the referent.
Results The study sample included 401 patients with active RA. Of these, 266 (66%) were Caucasian, 88 (22%) Asian and 28 (7%) Afro-Caribbean, and 19 (5%) were other/mixed race. Compared with Asians, Caucasians were older (62 vs. 53 years, respectively; p<0.001) and heavier (76 kg vs. 67 kg, respectively; p<0.001). Compared with Caucasians, Asians had a higher ESR (42 and 36, respectively; p=0.04), which was confirmed after controlling for age, weight, SJC, TJC and smoking (β 11.74; p=0.003); and a lower CRP (26.8 vs. 30; p=0.6). The overall DAS score was also slightly higher in Asians compared with Caucasian (6.63 vs. 6.39; p=0.09). There were no significant differences with regards to other DAS28 components (i.e. VAS, TJC, SJC). Compared with Caucasians, AC had a higher ESR (47 vs. 36, respectively; p=0.04), which was confirmed after adjustment for age, weight, SJC, TJC and smoking (β 10.9; p=0.05), and slightly higher CRP (32.4 vs. 30; p=0.5). There were no significant differences with regards to age, VAS, TJC, SJC or DAS28. PIPJ involvement (presence of swelling or both tenderness and swelling of any PIP joint) was more common in Caucasians than in Asians (84% vs. 71%, respectively; p=0.006), and AC patients (84% vs. 64%, respectively; p=0.01), with right PIP involvement more commonly seen among Caucasians (70% vs. 53%; p=0.005 and 70% vs. 54%; p=0.07), particularly of the right PIP 1-3 joints. There were no differences with regards to other joint involvement distribution.
Conclusions Our results show that Caucasian patients are more likely to have PIPJ involvement than Asian and AC patients, but with a similar distribution other joint involvement. In contrast, Asian and AC patients are more likely to have a higher ESR than Caucasians, in line with previous studies. Our data provides further evidence for ethnic variation in ESR, independent of joint involvement. In contrast to previous studies no differences in tender and swollen joint counts and VAS scores were observed.
Disclosure of Interest None Declared