Objective
To examine whether smoking is associated with autoantibody production in systemic lupus erythematosus (SLE) patients, unaffected first-degree relatives (FDR) of individuals with SLE - a group at increased risk of developing SLE, or unaffected, unrelated controls.
Methods
Detailed demographic, environmental, clinical, and therapeutic information was collected by questionnaire on 1,242 SLE patients, 981 FDRs, and 946 controls in the Lupus Family Registry and Repository; a blood sample was obtained. All sera were tested for multiple lupus autoantibodies by immunofluorescence and luminex bead-based assays. Generalized estimating equations, adjusting for age, gender, and ethnicity and accounting for correlation within families, were used to assess smoking status with the dichotomous outcome variables of positivity for SLE status, positivity of ANA by immunofluorescence (≥ 1:120), positivity for ≥ 1 autoantibody by the luminex assay, and positivity for each of the 11 autoantibodies.
Results
Current smoking was associated with being positive for ≥ 1 autoantibody (excluding ANA) (adjusted OR=1.53, 95% CI 1.04–2.24) in our subjects with SLE. No association was observed in unaffected FDRs or healthy controls. Former smoking was associated with anti-Ro/SS-A60 in our unaffected FDRs. There was an increased association with anti-nRNP A seropositivity, as well as a decreased association with anti-nRNP 68 positivity, in current smokers in SLE subjects.
Conclusions
No clear association between smoking status and individual autoantibodies was detected in SLE patients, unaffected FDRs, nor healthy controls within this collection. The association of smoking with SLE may therefore manifest its risk through mechanisms outside of autoantibody production, at least for the specificities tested.