RESUMO
OBJECTIVES: To determine cotinine levels in hair samples collected from smokers and estimate their relationship with nicotine intake and nicotine dependence class according to the Fagerström dependence questionnaire. DESIGN AND METHODS: Sixty-nine volunteers participated. They gave hair samples and answered questionnaires covering information related to smoking habits. Cotinine levels were determined by enzyme immunoassay after extraction from 20 mg of hair. Calibrators for the range 100 to 12,800 pg/mg were used. RESULTS: Mean cotinine concentration was 2070.9 pg/mg (range, 469.4-10,188.6 pg/mg). There were significant correlations between cotinine levels in hair and the questionnaire results (rs = 0.325, P = 0.018) and with the number of cigarettes smoked in a day (rs = 0.717, P < 0.001). Although the correlation between questionnaire results and the number of cigarettes smoked per day was significant (rs = 0.565, P < 0.01), it was weaker than the association between cotinine levels and number of cigarettes smoked. Cotinine levels were lower in the group that smoked from 1 to 5 cigarettes per day (1104.1 ng/mg) compared with the other groups. CONCLUSIONS: Cotinine levels were more highly correlated with nicotine exposure than with the Fagerström questionnaire scores. It was found that interference from exogenous sources of contamination, such as hair dyes, is a limitation to estimate nicotine intake from hair analysis.