BIRMINGHAM, ENGLAND – Patients with axial spondyloarthritis who currently smoke have been found to have worse disease activity than those who do not in an early analysis of data from the British Society for Rheumatology Biologics Register for Ankylosing Spondylitis (BSRBR-AS).

Smoking was associated with worse disease not only when comparing ever smokers with never smokers, but also when current smokers were compared to ex-smokers, and there was some evidence that the amount of smoking was important with heavier smokers having worse disease activity than light smokers.

The baseline analysis that revealed this association is “part of a bigger project to explore the effect of smoking” on anti–tumor necrosis factor treatment response in patients with axial spondyloarthritis (axSpA), Steven Sizheng Zhao, MD, said at the British Society for Rheumatology annual conference.

Dr. Zhao, of Aintree University Hospital in Liverpool, England, added that, as in previous studies, these data show that “smoking is associated with worse disease activity at baseline and this needs to be accounted for in the next stage of longitudinal analysis.”

An association between smoking and worse disease activity in patients with axSpA has been reported previously, Dr. Zhao acknowledged, but this is not as clear cut as in rheumatoid arthritis where smoking is known to have a pathogenic effect. The small number of earlier studies looking at the possible effect of smoking in AS have been limited by their size and varying methodology, he added, and the studies’ researchers were not able to see if there was any potential dose effect of smoking.

The BSRBR-AS, which started recruiting patients with an Assessment of SpondyloArthritis international Society (ASAS) classification of axSpA in 2012, offers a unique opportunity to explore the association between smoking and axSpA further, he said. More than 2,500 patients are included in the register at present, none of whom should have not been treated with biologic agents at the time of recruitment.

The aim of the present analysis that looked at data on 932 patients was to quantify the effect of smoking status and quantity on several disease outcomes as measured by the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Bath Ankylosing Spondylitis Functional Index (BASFI), and the Bath Ankylosing Spondylitis Metrology Index (BASMI). Spinal pain was also assessed, by way of a visual analog scale (VAS), and quality of life was determined via the disease-specific Ankylosing Spondylitis Quality of Life Instruments (ASQoL).

Most of the patients recruited were male (71%), and the mean age was 50 years. HLA-B27 data were available for 64% of the cohort, and 84% were positive.

Self-reported smoking status was recorded, with 19% saying they were current smokers, 37% saying they were ex-smokers, and 44% saying they had never smoked. If patients reported being ever smokers, the frequency with which they smoked (daily, weekly, monthly, once or twice, or never) was recorded, and if patients smoked daily, then the number of cigarettes smoked per day was obtained. Heavy smoking was defined as smoking 10 or more cigarettes per day and light smoking as fewer than 10 cigarettes per day. By this definition, around 37% of daily smokers were classed as heavy smokers.

In a comparison of ever smokers with never smokers, the mean BASDAI, BASFI, BASMI, spinal VAS, and ASQoL were significantly higher if patients had smoked at some point. All comparisons were adjusted for age, gender, body mass index, and HLA-B27 status.

The mean BASDAI score, for example, was more than 1 unit higher in a comparison of ever smokers with never smokers, with an adjusted regression coefficient of 1.04 and a 95% confidence interval (CI) of 0.72-1.36.

The adjusted regression coefficients for the other measures assessed were 1.34 (95% CI, 0.98-1.69) for BASFI, 0.61 (95% CI 0.36-0.87) for BASMI, 1.11 (95% CI, 0.74-1.49) for spinal VAS, and 2.71 (95% CI, 2.01–3.41) for ASQoL.

Similar findings were seen in the analysis comparing current with ex-smokers across all measures studied, and there was a trend for disease activity to be worse in heavier than in lighter smokers. The latter may not have reached significance because of the smaller number of patients (n = 172) involved in that part of the analysis. Nevertheless, these preliminary findings suggest that even being a light smoker can affect disease outcomes and so cutting down (i.e., to fewer than 10 cigarettes per day) may not be sufficient to reduce the effect that smoking has on disease activity.

“We should be actively encouraging our patients [with axSpA] to stop smoking,” Dr. Zhao concluded.

The BSRBR-AS is managed by a team of investigators based at the University of Aberdeen in collaboration with 83 centers throughout the United Kingdom. It is funded by the BSR, which in turn receives funding from the manufacturers of the biologic therapies included in this study (currently AbbVie, Pfizer, and UCB). Dr. Zhao and coauthors had no conflicts of interest to disclose.

Ankylosing spondylitis disease severity worsened by smoking