RESUMO
BACKGROUND: Cigarette smoking is the most well-established risk factor for developing bladder cancer. OBJECTIVE: To investigate the role of smoking status on the clinical outcome of patients with non-muscle-invasive bladder cancer. DESIGN, SETTING, AND PARTICIPANTS: Data obtained during a prospective phase 3 study with three schedules of epirubicin were used for statistical analysis. Smoking status (obtained when entering the study), other prognostic variables, and clinical outcome measures of 718 patients were analyzed. Mean follow-up was 2.5 yr. MEASUREMENTS: The primary outcome measure was recurrence-free survival (RFS). RESULTS AND LIMITATIONS: Demographics were similar for nonsmokers versus ex-smokers and current smokers, except for gender (p<0.001) and grade (p=0.022). In univariate analyses, RFS was significantly shorter in male patients (p=0.020), in patients with a history of recurrences (p<0.003), in patients with multiple tumors (p<0.004), in patients with a history of intravesical therapy (p=0.037), and in ex-smokers and current smokers (p=0.005). In multivariate analyses, a history of recurrences, multiplicity, and smoking status remained significant factors for predicting RFS. Gender and initial therapy were no longer a significant influence on RFS. Because progression was uncommon (n=25) and follow-up was short and focused only on recurrences, no conclusion can be drawn on progression-free survival. A limitation of the study were the questionnaires. They were only used when entering the study, and there were no questions about passive smoking and other causal factors. CONCLUSIONS: In this prospective study, the significance of known factors (history of recurrences and number of tumors) in predicting RFS was confirmed. Another significant factor that appears to predict RFS is smoking status: ex-smokers and current smokers had a significantly shorter RFS compared with nonsmokers.