Application of response surface methodology to vortex-assisted dispersive liquid-liquid extraction for the determination of nicotine and cotinine in urine by gas chromatography-tandem mass spectrometry

A method of vortex-assisted dispersive liquid-liquid extraction coupled with gas chromatography and tandem mass spectrometry for the determination of nicotine and cotinine in urine was developed. Response surface methodology was applied to obtain the optimum extraction conditions. In this method, Plackett-Burman design was utilized to evaluate the impact of five selected factors on pretreatment procedure. Then, three main factors were optimized using a Box-Behnken design. The optimized method showed good linearities at 1-2000 μg/L with correlation coefficients of 0.9998 for nicotine and 0.9986 for cotinine. Recovery was 91.4-106 and 91.7-108% for nicotine and cotinine, respectively. The intraday relative standard derivations of determination were 1.47-4.06% for nicotine and 0.41-3.16% for cotinine, and interday relative standard derivations were 3.03-6.70% for nicotine and 1.64-6.38% for cotinine. The method detection limits for nicotine and cotinine were 0.33 and 0.34 μg/L, respectively. A total of 87 urine samples from smokers and nonsmokers were tested with the proposed method. Urinary nicotine and cotinine were 23.0-6.67 × 103 and 18.4-4.17 × 103  μg/(g·cr) for smokers and 1.31-286 and 1.39-131 μg/(g·cr) for nonsmokers, respectively. The method is sensitive, suitable and reliable for the determination of nicotine and cotinine in urine and meets the requirements for evaluating short-term tobacco exposure.