Smoking Impact on Radiographic Progression in An Early Rheumatoid Arthritis Cohort

Abstract

Smoking is a known predictor of rheumatoid arthritis (RA), but its effect in radiographic progression is still controversial.

To investigate the effects of cigarette smoking on radiographic progression in an early RA cohort two years after the beginning of DMARD therapy.

Method:

One hundred and fifty six patients (83% female) with early RA (disease duration < 2 years) treated with a similar therapeutic strategy with DMARDs and low-dose glucocorticoids were included. Several demographic, genetic (DRB genotype), clinical, laboratory and radiographic data were obtained at baseline and after one and two years of follow-up. Smoking was evaluated at study entry and patients were classified as past smokers, current smokers and non-smokers. Radiographic progression after one and two years of follow-up was evaluated in hand and feet radiographs using the erosion joint count (EJC) and Larsen-Scott score. Multivariate lineal regression analysis was performed.

Result:

The frequency of current, past and non-smokers were 30.1%, 12.2% and 57.7% respectively. Only 14% of patients were heavy smokers (> 20 pack/year). Radiographic damage at one and two years was higher in current smokers compared with non-smokers, although only in the EJC at two years the difference was statistically significant (1.2±1.7 vs. 0.7±1.7, p=0.04). The Larsen score at two years was higher in current smokers than in non-smokers although the difference was not significant (7.4±12.9 vs. 4.2±6.7, p>0.05). In the univariate analysis taking Larsen-Scott score as the measure of radiographic damage, female sex, higher HAQ at disease onset, the presence of anticitrullinated antibodies, the shared epitope (SE) and the HLA-DRB*04 had a significant effect on radiological score. In the multivariate regression analysis adjusted for baseline radiographic damage, female sex, HLA-DRB*04 genotype and current smoking were independently associated with two year radiographic damage measured by the Larsen Score as well as with the EJC (Table). No significant differences were seen in clinical disease activity or rates of EULAR response at one and two years of follow-up between smokers and non smokers.

Conclusion:

In our early RA cohort, current smoking emerges as an independent factor for radiographic progression after two years of DMARDs.