Nodular rheumatoid arthritis (RA): A distinct disease subtype, initiated by cadmium inhalation inducing pulmonary nodule formation and subsequent RA-associated autoantibody generation.

Nodular rheumatoid arthritis (RA) patients have raised rheumatoid factor (RF) and anti-citrullinated protein antibody (ACPA) levels, and are more likely to smoke than RA patients without nodules. Subcutaneous and pulmonary rheumatoid nodules (granulomas) frequently co-exist. Pulmonary rheumatoid nodules develop prior to RA development and have the immunological machinery to generate RF and ACPAs. Pulmonary granulomas have been observed in animal models exposed to cadmium (Cd) inhalation. Cigarette smoke increases pulmonary Cd exposure. It has been suggested that dust and cigarette smoke co-exposure increases localised pulmonary Cd adsorption. We hypothesise that subcutaneous nodular RA represents a distinct disease subtype induced by pulmonary rheumatoid nodule formation and the generation of high levels of RA associated autoantibodies initiated by Cd inhalation via cigarette smoke. Cohorts of RA patients attending rheumatology clinics in Cornwall, UK (total n = 504) were studied to determine the prevalence of nodular RA, with matched analysis (age, gender and social class) to compare urinary Cd, RF and ACPA levels stratifying for nodular disease and smoking. In cohort 1 45/303 (14.9%) of the RA patients under regular follow up had nodular disease. Of the RA smokers, 30/155 (19%) were nodular and of the RA non-smokers 15/148 (10%) were nodular. Smoking was significantly associated with nodular RA, odds ratio (OR) = 2.48 95% confidence interval (CI) 1.26-4.88, p = 0.008. Raised urinary Cd levels were significantly associated with nodular RA in non-dust exposed individuals, OR 2.26 (95% CI 1.08-4.73), p = 0.03 compared to dust exposed individuals, OR 0.78 (95% CI 0.35-1.76), p = 0.557, despite fewer pack years (py) at diagnosis (16 vs 20 py). Nodular RA smokers had significantly raised RF levels compared to RA smokers without nodular disease (median RF 171.5 (interquartile range (IQR) 48-394) vs median RF 31.7 (IQR 10.3-170.3), p < 0.00001). RF positivity was significantly more prevalent in nodular RA smokers compared to RA smokers without nodular disease (84/89 (94%) vs. 141/199 (71%), OR = 6.9 (95% CI 2.66-17.91), p < 0.00001). ACPA levels were also significantly raised in nodular smokers compared to non-nodular smokers (median ACPA 250 (IQR 145-426) vs 116 (1-257.5), p < 0.00001), as were ACPA positivity rates (83/89 (93%) vs 123/191 (64%), OR = 7.65 (95% CI 3.17-18.4), p < 0.0001). These pilot results support the hypothesis that nodular RA represents a distinct disease subtype initiated by cadmium inhalation, which we suggest induces pulmonary rheumatoid nodule formation and generation of RA-associated autoantibodies.