ABSTRACT
Cigarette smoking may be associated with the augmentation of pro-inflammatory cytokines including Tumor Necrosis Factor-alpha (TNF-alpha), which may affect the outcomes of pharmacological agents such as TNF-alpha inhibitors. The purpose of this study was to investigate the impact of smoking on the effectiveness of TNF-alpha inhibitors in patients with rheumatoid arthritis (RA) or Crohn's disease (CD). We used systematic literature review methods. A total of 1,147 articles were selected after exclusion of duplicates through a database search. Among them, 28 articles were finally selected through a review of titles and abstracts and a subsequent review of full articles. The effectiveness of TNF-alpha inhibitors in patients with RA or CD among the selected articles was summarized by their smoking status. Meta-analysis was performed with random effect model. When current smokers were compared with non-smokers for response after adjustments through meta-analysis among patients with RA, current smokers had 59% less response than non-smokers with statistical significance (Pooled adjusted OR=0.41, 95% CI=0.17-0.95). In patients with CD, current smokers tended to have lower clinical response than non-smokers, but statistical significance was not shown. In subgroup analyses for luminar CD or fistulizing CD, current smokers tended to have a lower response in luminar CD (Pooled OR=0.62, 95% CI=0.34-1.14), but smoking status was not associated with drug response in fistulizing CD. This study raises awareness of the adverse effects of smoking in terms of clinical response in patients treated with TNF-alpha inhibitors.