Background Systemic sclerosis (SSc) is a rare, multisystem autoimmune disorder. It is characterised by generalized microangiopathy, in which hypoxia and oxidative stress have been implicated in its pathogenesis. Tobacco inhalation increases free radicals and strongly promotes vascular damage. So far, data available with regards to a role of tobacco exposure with SSc severity and progression are scarce.

Objectives We aimed to assess the effects of smoking on the speed of worsening of organ manifestations, namely lung involvement (forced vital capacity, FVC; forced expiratory volume, FEV1/FVC ratio; diffusing capacity for carbon monoxide corrected for alveolar volume, DLCO/VA), skin involvement (modified Rodnan skin score; mRSS), and digital ulcers (DU) in the European scleroderma trials and research (EUSTAR) database.

Methods Adult SSc patients from the EUSTAR cohort with a follow-up visit 12±4 months after their baseline visit and available data on their smoking habits were included.

The associations of smoking behaviour (never smokers vs ex-smokers vs current smokers) with the disease manifestations at follow up were assessed after adjusting for potentially confounding covariates using multivariable linear or logistic regression analyses.

Missing data were imputed using multiple imputations.

Results Of the 3,023 patients included (mean age 57 years, SD 13; 85% female), 66% stated that they never smoked, 23% were ex-smokers and 11% were current smokers. On average, ex-smokers had smoked for 19.5 years (SD 12) while current smokers smoked for 29.1 years (SD 13). Ex-smokers had smoked on average 17.3 pack-years (SD 20) and current smokers 29.3 pack-years (SD 36). The mean time since smoking cessation in ex-smokers was 15.8 years (SD13).

The FEV1/FVC ratio changed from 96.3 (SD 14) at baseline to 96.2 (SD 13) at follow up. Taking into account the effect of age, sex, autoantibody status, disease duration and SSc subset as well as the baseline values of the outcome, the FEV1/FVC ratio decreased faster in currently smoking SSc patients than in never smokers (β=-2.8, p=0.001). Similarly, the DLCO/VA diminished faster in current smokers (β=-3.7, p=0.006) while on average the DLCO/VA decreased from 78.1 (SD 18) to 76.5 (SD 17). Changes in FVC were however not associated with smoking behaviour (p=0.13).

The mRSS changed between baseline and follow-up from an average of 7.6 (SD 7) to 7.3 (SD 7); on average, in smokers the mRSS improved faster than in never smokers (β=-0.65, p=0.015). The presence of DUs at follow-up was not associated with current smoking (p=0.9).

Having previously smoked was not significantly associated with the progression of any of the assessed outcomes.

Conclusions Although the adverse effects of smoking on bronchial airways that is known in the general population could be replicated in the SSc population, the lack of a measurable adverse effect of smoking on SSc organ manifestations argues against a major role of tobacco associated free radicals and vasoconstriction in the pathogenesis of SSc vasculopathy and fibrosis.

Disclosure of Interest V. Jaeger: None declared, G. Valentini: None declared, E. Hachulla: None declared, L. Czirják: None declared, E. Siegert: None declared, O. Distler Grant/research support from: 4 D Science, AbbVie, Actelion, Active Biotec, Bayer, BiogenIdec, BMS, Boehringer Ingelheim, ChemomAb, EpiPharm, espeRare foundation, Genentech/Roche, GSK, Inventiva, iQone Healthcare, Lilly, medac, Mepha, MedImmune, Mitsubishi Tanabe Pharma, Pharmacyclics, Pfizer, Sanofi, Serodapharm and Sinoxa, Consultant for: 4 D Science, AbbVie, Actelion, Active Biotec, Bayer, BiogenIdec, BMS, Boehringer Ingelheim, ChemomAb, EpiPharm, espeRare foundation, Genentech/Roche, GSK, Inventiva, iQone Healthcare, Lilly, medac, Mepha, MedImmune, Mitsubishi Tanabe Pharma, Pharmacyclics, Pfizer, Sanofi, Serodapharm and Sinoxa, L. M. Bambara: None declared, E. Rosato: None declared, Y. Allanore: None declared, M. Matucci-Cerinic: None declared, P. Airό: None declared, U. Walker: None declared