The effect of smoking on cumulative damage in systemic lupus erythematosus: An incident cohort study.

Abstract

To investigate the relationship between smoking history and pack-year exposure on the rate of end-organ damage in systemic lupus erythematosus (SLE). The SLE incident cohort included patients who met American College of Rheumatology (ACR) 1997 or SLE International Collaborating Clinics (SLICC) 2012 SLE criteria and had rheumatology encounters at a US academic institution (2008-16). The primary outcome was median time to SLICC/ACR damage index (SLICC/ACR-DI) increase or death. Main explanatory variables were smoking status and pack-years. Covariates included age, sex, race, ethnicity, receipt of Medicaid, neighborhood area deprivation index, and baseline SLE damage. Damage increase-free survival was evaluated by smoking status and pack-years using Kaplan-Meier and Cox proportional hazards methods. Patients of Black race and Medicaid recipients were more commonly current smokers (p's < 0.05). Former smokers were older and more likely to have late-onset SLE (54% versus 33% of never and 29% of current smokers, p = 0.001). Median time to SLICC/ACR-DI increase or death was earlier in current or former compared to never smokers (4.5 and 3.4 versus 9.0 yrs; p = 0.002). In multivariable models, the rate of damage accumulation was twice as fast in current smokers (HR 2.18; 1.33, 3.57) and smokers with a >10 pack-year history (HR 2.35; 1.15, 3.64) versus never smokers. In this incident SLE cohort, past or current smoking predicted new SLE damage 4-5 years earlier. After adjustment, current smokers and patients with a pack-year history of >10 years accumulated damage at twice the rate of never smokers.