Table 1.

Impact of smoking on main antidepressants (SSRI and SNRI) pharmacokinetics

Study	Drug	Country	N total/N smokers	Age	Men (%)	Main findings	Study limitations
Lundmarck et al. [12]	Fluoxetine	Sweden	291/138	43	39	No significant correlation between serum levels of FLX and smoking habits was found	Possibility of interactions with other medications
Koelch et al. [13]	Fluoxetine	Germany	65/13	14.6	37.0	Serum concentrations of the active moiety and NORFLX were significantly correlated with smoking status	Possibility of interactions with other medications
						Serum levels of the active moiety of FLX were 44% lower in smokers than in nonsmokers	Due to the ethnic polymorphism of CYP2D6 [6], the current results are not transferable to races other than Caucasians
						No significant correlation between serum levels of FLX and smoking habits was found	Population younger than 19
Lundmarck et al. [14]	Sertraline	Sweden	319/89	54.4	31	Smokers had significantly lower concentration-to-dose (C/D) mean ratios of serum sertraline and its main metabolite desmethylsertraline than nonsmokers	Possibility of interactions with other medications
Taurines et al. [15]	Sertraline	Germany	85/5	14.8	45.1	No significant correlation between serum levels of Sertraline and smoking habits was found	Possibility of interactions with other medications
							No standardization of timing of blood withdrawal and length of treatment as well as patient compliance
							Low number of subjects taking high doses of sertraline
							Population younger than 19
Reis et al. [16]	Escitalopram	Sweden	130/31	51	32	No significant correlation between serum levels of Escitalopram and smoking habits was found	76% of the patients took one or more drugs in addition to escitalopram
Reis et al. [17]	Citalopram	Sweden	19/10	<21	18.8	No significant correlation between serum levels of Citalopram and smoking habits was found	Possibility of interactions with other medications
							Population younger than 21
Spigset et al. [18]	Fluvoxamine	Sweden	24/12	36.7	58.3	Cmax, and AUC were significantly lower in the smokers than in the nonsmokers	Only 50 mg/day of FLV was tested
						There were no group differences in elimination half-life	Small sample size
Carrillo et al. [19]	Fluvoxamine	Sweden	14/6	33.9	50	Among extensive metabolizers (CYP1A2 e CYP 2D6) there was no difference in FLV kinetics between smokers and nonsmokers	Only 50 mg/day of FLV was tested
							Small sample size
							Limited to extensive metabolizers
Yoshimura et al. [20]	Fluvoxamine	Japan	30/11	52	36.7	Serum levels of FLV were significantly higher in nonsmokers than smokers	Small sample size
							Some subjects also took benzodiazepines
Gerstenberg et al. [21]	Fluvoxamine	Japan	49/15	49.9	69.0	No significant difference between nonsmokers and smokers in the Css of FLV and FLA and FLA/FLV ratio was found	Thirty-nine patients also took benzodiazepines
							Small sample size
Sugahara et al. [22]	Fluvoxamine	Japan	49/13	w.i.	w.i.	The mean C/D ratio of FVX in smokers was reduced by more than 30% in comparison with that in nonsmokers	No information about age or sex of patients
Katoh et al. [23]	Fluvoxamine	Japan	32/6	39.0	46.9	The steady-state plasma C/D ratio of FLV in patients who smoked was significantly lower than that in nonsmoker patients	Possibility of interactions with other medications
							Small sample size
Suzuki et al. [24]	Fluvoxamine	Japan	87/22	36.6	65.5	Heavy smokers had significantly lower FLV concentration than nonsmokers in the FLV 50 mg/d dose group	Possibility of interactions with other medications
						At 150 mg/day and 200 dose groups, no significant differences in FLV concentration were observed between nonsmokers and heavy smokers	Small sample size
Reis et al. [25]	Venlafaxine	Sweden	141/58	61.3	33	The steady-state plasma C/D ratio of ODV and DDV in patients who smoked was significantly lower than that in nonsmoker patients	Possibility of interactions with other medications
						No differences in C/D VEN values or in any of the metabolite/VEN ratios were found
Unterecker et al. [26]	Venlafaxine	Germany	227/87	49.1	36.4	In smokers, mean serum levels of ODV were 21% lower than in nonsmokers	Possibility of interactions with other medications
						No differences in C/D VEN values between two groups
Fric et al. [27]	Duloxetine	Germany	23/8	47.3	75	Smokers show significantly lower duloxetine serum level than nonsmokers	Possibility of interactions with other medications
							Small sample size
Lobo et al. [28]	Duloxetine	USA	594/123	48.8	26	Serum levels of duloxetine were significantly lower in smokers than in nonsmokers	Possibility of interactions with other medications
						Nonsmokers have a 43% higher Css than smokers
Ishida et al. [29]	Trazodone	Japan	43/16	43	44.2	Smokers show significantly lower duloxetine serum level than nonsmokers	Small sample size
						No differences in mCPP concentrations between two groups	Some subjects also took benzodiazepines
Lind et al. [30]	Mirtazapine	Sweden	56/36	50	34	Smokers show significantly lower S-mirtazapine and R-N-desmethylmirtazapine serum levels than nonsmokers	No information about the number of cigarettes consumed daily
Sirot et al. [31]	Mirtazapine	France/Switzerland	45/17	51	18.9	Smokers show significantly lower mirtazapine, S-mirtazapine and R-N-desmethylmirtazapine serum levels than nonsmokers	Possibility of interactions with other medications
Hsyu et al. [32]	Bupropion	USA	34/17	26.2	52.9	No significant correlation between serum levels of bupropion and smoking habits was found	Almost every subjects are caucasian
							Small sample size
							Only evaluates 150 mg daily