Development and validation of a HPLC-MS/MSmethod for the determination of venlafaxineenantiomers and application to apharmacokinetic study in healthyChinese volunteersWen Liu,a* Ying-Chun Dai,a,b Nang Deng,a Xiang-Rong Liua and Yi Luoa

ABSTRACT: An HPLC-MS/MS method has been developed and validated for the determination of venlafaxine enantiomers inhuman plasma and applied to a pharmacokinetic study in healthy Chinese volunteers. The method was carried out on avancomycin chiral column (5 mm, 250 ¥ 4.6 mm) maintained at 25°C. The mobile phase was methanol–water containing30 mmol/L ammonium acetate, pH 3.3 adjusted with aqueous ammonia (8:92, v/v) at the flow rate 1.0 mL/min. A tandem massspectrometer with an electrospray interface was operated in the multiple reaction monitoring mode to detect the selected ionspair at m/z 278.0, 120.8 for venlafaxine enantiomers and m/z 294.8, 266.7 for estazolanm (internal standard). The methodwas linear in the concentration range of 0.28–423.0 ng/mL. The lower limit of quantification was 0.28 ng/mL. The intra-andinter-day relative standard deviations were less than 9.7%. The method was successfully applied for the evaluation of phar-macokinetic profiles of venlafaxine enantiomers in 18 healthy volnteers. Validation parameters such as the specificity, linear-ity, precision, accuracy and stability were evaluated, giving results within the acceptable range. Pharmacokinetic parametersof the venlafaxine enantiomers were measured in the 18 healthy Chinese volunteers who received a single regimen withvenlafaxine hydrochloride capsules. The results show that AUC(0–•), Cmax and t1/2 between S-venlafaxine and R-venlafaxine aresignificantly different (p < 0.05). Copyright © 2010 John Wiley & Sons, Ltd.

Keywords: venlafaxine; HPLC-MS/MS; enantiomers; pharmacokinetics

IntroductionVenlafaxine hydrochloride (VEN) is a phenethylamine bicyclicderivative, chemically known as 1-[2-(dimethylamino)-1-(4-methoxy-phenyl) ethyl]cyclohexanol hydrochloride. It is a novel,non-tricyclic antidepressant, and the mechanism of action inhumans is believed to be associated with potential of neurotrans-mitter activity in the central nervous system. VEN has two enan-tiomers, which are administered as a racemic mixture, exerting adual mechanism of action on the monoaminergic system (Muthet al., 1986, 1991; Andrews et al., 1996; Morton et al., 1995). Thetwo enantiomers exhibit different activities. R-VEN inhibits boththe noradrenalin and serotonine synaptic re-uptake whereasS-VEN inhibits only the serotonine one (Moyer et al., 1984; Holli-day and Benfield, 1995).

Different analytical methods have been reported for the deter-mination of venlafaxine enantiomers in biological samples. Cap-illary electrophoresis has been applied in the analysis of the chiralcompounds. The analytical methods of chiral separation of venal-faxine have included various capillary electrophoresis methods(Rudaz et al., 2000; Cherkaoui et al., 2001b; Desiderio et al., 2001).Rudaz et al. (2000) have established a method for the determina-tion of venlafaxine enantiomers in clinical samples by capillaryelectrophoresis using charged cyclodoextrin, which has a rela-tively low sensitivity, and the results demonstrated an important

inter-individual difference in the formation of ODV (the activeVEN metabolite); the enantiomeric ratio of both analytes indi-cated a stereoselective metabolism. While Desiderio et al. (2001)and Fanali et al. (2001) utilized the vancomycin silica stationaryphase in packed capillary electrochromatrography to separatethe venlafaxine enantiomers, and the two methods were success-fully applied to the analysis of the clinical patients under depres-sion therapy, showing a stereoselective type venlafaxinemetabolism. However, the two methods have a relatively highlower limit of quantitation (LLOQ; >20 ng/mL). HPLC is also agood analytical method for the determination of chiral drugs.

* Correspondence to: Wen Liu, The Pharmaceutical Department of the HunanProvincial People’s Hospital affiliated to Hunan Normal University, Chang-Sha 410002, People’s Republic of China. E-mail: Kaiwinliu@yahoo.com.cn

a The Pharmaceutical Department of the Hunan Provincial People’s Hospitalaffiliated to Hunan Normal University, ChangSha 410002, People’s Republicof China

b The Medical College of Hunan Normal University, ChangSha 410002, Peo-ple’s Republic of China

Abbreviations used: ME, matrix effect; VEN, venlafaxine hydrochloride.

Research Article

Received 19 February 2010, Accepted 13 April 2010 Published online in Wiley Online Library: 4 June 2010

(wileyonlinelibrary.com) DOI 10.1002/bmc.1464

412

Biomed. Chromatogr. 2011; 25: 412–416Copyright © 2010 John Wiley & Sons, Ltd.

Wang et al. (1992) determined the venlafaxine enantiomers byHPLC using a chiral reagent with low sensitivity. Liu et al. (2007)established an HPLC method coupled with single mass spectrom-etry for the measurement of the VEN enantiomers in therapeuticdrug monitoring (TDM), with a relatively high LLOQ and a linearrange of 5.0–400 ng/mL for both S-VEN and R-VEN. However, themethod was not suitable for the pharmacokinetic study of ven-lafaxine enantiomers in healthy volunteers for a single regimen,which requires a higher sensitivity. We developed an HPLC-MS/MS method with higher sensitivity and successfully applied itto the pharmacokinetic study of venalfaxine enantiomers inhealthy Chinese volunteers.

Experimental

Reagents and Chemical

S-VEN (>99.8%), R-VEV (>99.8%) were purchased from TRC (TorontoResearch Chemicals Inc., Canada). Estazolanm (internal standard, IS;99.8%) was obtained from Sigma (Steinheim, Germany). HPLC-gradereagents methanol and tert-butyl methyl ether were obtained from MerckCompany (Germany). Other AR-grade reagents such as aqueousammonia and ammonium acetate were purchased from the ChemicalReagent Factory of Hunan (Changsha, Hunan, China). Distilled water wasprepared by a laboratory purification system. Drug-free human plasmawas obtained from the Blood Center of Shanghai (Shanghai, China).

Apparatus and Chromatographic Conditions

A Waters Acquity Ultra Performance LC system (Waters, Milford MA, US)coupled with a Quattro Premier XE mass spectrometer was used. Theexperments were carried out on a vancomycin chiral column (5 mm, 250 ¥4.6 mm; Astec, USA) with column temperature 25°C. A security guardholder (4.0 ¥ 3.0 mm i.d.) was employed to protect the analytical column.The LC system was operated isocratically at 25°C using a mobile phase ofammonium acetate buffer 30 mmol/L (pH3.3)–methanol (8:92, v/v). It wasfiltered through a 0.45 mm membrane filter and run at a flow rate of1.0 mL/min and a postcolumn splitting ratio of 3:1. The injection volumewas 20 mL for both standards and samples.

The mass spectrometry was operated using an electrospray source inpositive mode and the ionization source parameters were: capillaryvoltage, 3.0 kV; source temperature, 120°C; desolvation gas temperature,350°C at a flow rate of 900 L/h (N2); cone gas flow rate, 50 L/h. Nitrogen(99.9% purity) and argon (99.9999% purity) were used as cone and colli-sion gases respectively. The multiple reaction monitoring (MRM) modewas used to detect the selected ion pair at m/z 278.0 → 120.8 for ven-lafaxine enantiomers and m/z 294.8 → 266.7 for estazolanm (IS). The finaloptimized detection conditions were as follows: capillary voltage, 3.0 kV;cone voltage, 22 V for VEN, 39 V for IS; collision energy voltage, 25 eV forVEN, 25 eV for estazolanm (IS).

Sample Preparation and Extraction Procedures

The plasma was prepared by a liquid–liquid extraction method. Frozenhuman plasmas were thawed at ambient temperature. One hundredmicroliters internal standard working solution (estazolanm, 2.4 mg/mL)was added to the sample (0.5 mL). The sample was alkalinized by adding30 mL aqueous ammonia then shaken for 1 min. Two milliliters of tert-butyl methyl ether were added to the sample. After 2 min vortex-mixing,the mixture was centrifuged at 3000g for 5 min at room temperature(25°C), the upper layer was carefully aspirated and the remainder wasextracted once again with 2 mL tert-butyl methyl ether. The upper layerwas put together with the former, and the tert-butyl methyl ether wasevaporated under a stream of nitrogen at 40°C. The dry residue wasdissolved in 100 mL mobile phase with vortex-mixing for 1 min and 20 mLsolution was injected for analysis through the auto-injector.

Preparation of Stock Solutions and Calibration Standards

The primary stock solutions of R-VEN (80 mg/mL), S-VEN (120 mg/mL) andestazolanm (IS, 2.47 mg/mL) were prepared by dissolving an appropriateamount of the pure substance in methanol. Working solutions wereobtained by diluting the stock solutions with methanol. All the standardsolutions were stored at 0°C.

Routine daily calibration curves were prepared by spiking the appro-priate volume of working solutions to 0.5 mL in drug-free serum. Effectiveconcentrations in plasma were 0.28, 2.82, 17.62, 70.50, 176.25, 282.0 and423.0 ng/mL for S-VEN and R-VEN. The quality control (QC) samples wereseparately prepared by adding a suitable amount of venlafaxine enanti-omers to drug-free plasma to make concentrations of 2.8, 70.5 and282.0 ng/mL. The spiked plasma samples (standards and quality controls)were then treated following the ‘Sample Preparation and Extraction Pro-cedures’ on each analytical batch along with the unknown samples.

Assay Validation

Assay specificity and matrix effect. Specificity was assessed byextracting samples of six batches of blank plasma, and comparing theresults for plasma samples spiked with estazolanm (IS) and venlafaxineenantiomers. The chromatograms were also inspected visually for inter-fering chromatographic peaks from endogenous substances. Sincechromtatographic conditions may cause co-elution of a member ofendogenous compounds that were undetected by the MS but mightaffect the ionization efficiency, the effect of matrix on the response of theanalytes were also evaluated. The matrix effect (ME) on the ionization ofanalytes was evaluated by comparing the peak area of analytes resolvedin blank sample (the final solution of blank plasma after extraction andreconstitution) with that resolved in mobile phase. Three different con-centration levels of S-VEN enantiomers (2.8, 70.5 and 282.0 ng/mL) and2.47 mg/mL of internal standard were evaluated by analyzing five samplesat each level. The corresponding peak areas of the analyte in spikedplasma post-extraction (A) were then compared with those of theaqueous standards in mobile phase (B) at equivalent concentrations. Theratio (A : B ¥ 100) was defined as the ME. An ME value of 100% indicatesthat the responses for venlafaxine in the mobile phase and in the plasmaextracts were the same and that no absolute ME was observed. A value of>100% indicates ionization enhancement, and a value of <100% indicatesionization suppression. The results showed there was no absolute ME inthis study, and the variability was acceptable, with RSD values <5.4% atdifferent concentrations of S-VEN and internal standard.

Linearity. Plasma samples were quantified using the ratio of the peakarea of venlafaxine enantiomers to that of IS as the assay response. Cali-bration curves were prepared by spiking the appropriate standard solu-tion to 0.5 mL of blank plasma. Effective concentrations in plasmasamples were 0.28. To evaluate the linearity, plasma calibration curveswere prepared and assayed in triplicate on three consecutive days overthe range of 0.28–423 ng/mL encompassing the therapeutic range of thisantidepressive drug. Calibration curves were calculated using the peak-area ratio vs analyte concentrations. The seven concentrations of thestandard solutions were subjected to regression analysis by the least-squares method to calculate the calibration equation and correlationcoefficient (r). Linearity was expressed as correlation coefficient; the valuemust be >0.9990.

Precision, accuracy and recovery. The precision of the assay wasdetermined from QC samples. The intra-day precision was determined byrepeating the analysis of the standard five times a day, and the inter-daycoefficients of variance (CVs) were determined by repeating the analysison three consecutive days. The precision was then calculated using one-way ANOVA. In this assay, the intra-run precision was 9.7% or less, and theinter-run precision was 9.2% or less for each QC level of venlafaxine. Theaccuracy was within �10.4%. The results above demonstrated thatthe values were within the acceptable range and the method wasaccurate and precise. The recoveries of R-VEN and S-VEN, determined at

413

Determination of venlafaxine enantiomers

Biomed. Chromatogr. 2011; 25: 412–416 wileyonlinelibrary.com/journal/bmcCopyright © 2010 John Wiley & Sons, Ltd.

three concentrations (2.8, 70.5 and 282.0 ng/mL), were 92.7 � 6.8, 16.3 �

5.3 and 102.7 � 6.1%, and 89.6 � 7.2, 102.8 � 4.9 and 103.5 � 6.7%,respectively.

Stability. The stability experiments aimed at testing all possible condi-tions which the samples might be exposed to during sample shippingand handling. To test the stability of venlafaxine enantiomers in plasma,QC samples were stored under different conditions. Standard solutions ofR-VEN (80 mg/mL) and S-VEN (120 mg/mL) in methanol were stored at 0°C.Three batches of quality control samples spiked with the two analytes(R-VEN: 2.8, 70.5 and 282.0 ng/mL; S-VEN: 2.8, 70.5 and 282.0 ng/mL) werestored at -20°C for 3 months and at 4°C or room temperature for 24 h. Theresults of stability experiments showed that no significant degradationoccurred at -20°C for 3 months and at 4°C or room temperature for 24 h.

Pharmacokinetic Study Design

Subject. The clinical study protocol was approved by the Ethical Com-mittee of College of Hunan Provincial People’s Hospital (ChangSha, PRChina). Eighteen healthy Chinese male volunteers, aged 22–25 years,were selected for this study after clinical assessment of their health status(physical examination, electrocardiogram), hematological and biochemi-cal tests, and urinalysis. No subject had a history or evidence of renal,gastrointestinal, hepatic or hematologic abnormality or any acute orchronic disease, or allergies to any drugs. All the subjects were non-drinkers and non-smokers. Subjects who had used drugs of any kindwithin 2 weeks before the study were excluded. No tobacco, alcohol ordrink containing caffeine was allowed. Informed consent was obtainedfrom all subjects and the nature and purpose of the study had beenclearly explained. The volunteers had the following clinical characteristics[expressed as means � SD (range)]: age, 23.5 � 1.5 years; height, 167.1 �

6.5 cm (156–175); body weight, 59.5 � 5.6 kg (46–72).

Drug administration and blood sampling. This was a single-period experiment and the subjects were administered 75 mg venlafax-ine hydrochloride orally. Before the administration they were subjected tophysical examination and an overnight fast. The heparinized bloodsamples (4 mL) were collected immediately from a suitable forearm veinusing an indwelling catheter into heparin containing tubes before takingthe drug and then at 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0, 10.0, 12.0, 15.0, 24.0and 36.0 h after the administration. The blood samples were centrifugedat 3000g for 5 min, and plasma samples were separated and then storedat -20°C until analysis. The study was conducted at the Clinic PharmacyResearch Laboratory, Hunan Provincial People’s Hospital, which wasequipped with standard emergency drugs and equipment. A doctor withGCP training and two experienced nurses were recruited as part of theresearch staff to monitor changes in vital signs and adverse reactionsduring the study.

Pharmacokinetic analysis. Plasma concentrations of venlafaxineenantiomers were analyzed using the developed HPLC-MS/MS method.Pharmacokinetic parameters of the two enantiomers were calculated byone-compartment analysis using Drug and Statistics (version 2.0, China).The maximum plasma concentration (Cmax) and their time of occurrence(Tmax) were obtained directly from the measured data. The area underthe plasma concentration–time curve from time zero to the last mea-surable concentration (AUC0–t) was calculated using the linear trapezoi-dal rule and was extrapolated to infinity (AUC0–•) according to therelationship: AUC0–• = (AUC0–t + Ct)/Ke, where Ct is the last concentrationevaluated in plasma greater than the limit of quantification (LOQ); theterminal elimination rate constant (Ke) was estimated by linear least-squares regression of the terminal portion of the plasma concentration–time curve, and the corresponding elimination half-life (t1/2) was thencalculated as 0.693/Ke; Cl/F = Ke·Vc. Vc is the apparent volume of distri-bution of the center compartment, which was estimated by the modelafter calculation. V/F is the apparent volume of distribution. Differencesin pharmacokinetic parameters (AUC0–•, Tmax, T1/2 and Cmax) between

S-VEN and R-VEN were compared using a paired (two-tailed) t-test witha significant p-value of 0.05.

Results and Discussion

Chromatrography and Specificity

To develop this HPLC-MS/MS-based method to quantify venlafax-ine enantiomers in human plasma, electrospray ionization (ESI)source were evaluated in positive ion mode. The selected ion pairwere m/z 278.0 → 120.8 for S-VEN and R-VEN, m/z 294.8 → 266.7for the internal standard.

The composition of mobile phase was critical factor for achiev-ing the chiral separation of venlafaxine. In the present study,methanol–30 mmol/L ammonium acetate pH 3.3 (85:15, v/v) wasselected as an isocratic mobile phase. Selectivity was assessed bycomparing the chromatograms of the blank plasma with the cor-responding spiked plasma. Figure 1 shows the typical chromato-grams of blank plasma, spiked plasma sample with S-VEN, R-VENand IS, and plasma sample from the subject after oral administra-tion. Interferences from the matrices at the expected retentiontimes of the target ions were not observed. The retention times ofS-VEN, R-VEN and IS were 6.02, 6.72 and 5.26 min, respectively.

Linearity and Lower Limit of Quantification

Calibration standards at seven venlafaxine enantiomers concen-tions were extracted and assayed. Least-squares linear regressionwas used to determine the plasma concentration from the peakarea ratios. The response was linear for venalfaxine enantiomersthroughout this concentration range and the correlation coeffi-cients (r) were >0.9990 for all standard curves using a 1/xweighted linear regression model. The typical equations were y =2.83 + 15.93 ¥ c (ng/mL; r = 0.9992), y = 3.63 + 15.64 ¥ c (ng/mL; r= 0.9994) for S-VEN and R-VEN, respectively. The lower detectionlimit (LOD), defined as signal-to-noise ratio (S/N) > 3 was0.02 ng/mL and the LLOQ of S-VEN and R-VEN, defined as S/N>10was 0.28 ng/mL. The precision and accuracy of the LLOQ were<9.7% and within �0.5%. During routine analysis, each analyticalrun included a set of calibration samples, a set of QC samples induplicate and the unknowns.

Precision, Accuracy and Extraction Recovery

The extraction recovery, inter- and intra-day precision and accu-racy values are shown in Table 1. The extraction recovery of the ISwas >89.6%, and both the intra-day and inter-day CVs were<9.7%.

Stability Studies

The results of the stability experiments showed that no signifi-cant degradation occurred at 4°C for 24 h, at room temperaturefor 24 h and at -20°C for 3 months. The accuracy values of low(2.82 ng/mL), medium (70.5 ng/mL) and high(282.0 ng/mL) con-centrations of S-VEN were 92.1, 97.7 and 106.6% at -20°C for 3months, 103.1, 97.8 and 99.6% at 4°C for 24 h, and 95.8, 101.7 and104.6% at room temperature for 24 h. The QC plasma samples ofR-VEN were also stable in these conditions (Table 2). The stocksolutions of S-VEN and R-VEN in methanol were stable at 4°C for aweek.

Pharmacokinetic Study

This method was successfully applied to the pharmacokineticstudy of venlafaxine enantiomers in a single dose. The mean

414

W. Liu et al.

Biomed. Chromatogr. 2011; 25: 412–416wileyonlinelibrary.com/journal/bmc Copyright © 2010 John Wiley & Sons, Ltd.

(�SD) plasma concentration–time profiles for healthy volunteersafter the oral administration of venlafaxine at dosages of 75 mgare shown in Fig. 2. The values of the main pharmacokineticparameters are shown in Table 3.

Discussion

Although we have established an HPLC-ESI/MS method for thesimultaneous stereoselective determination of S-VEN and R-VEN

Figure 1. Chromatography of compounds and IS, A: Control human plasma B: Compounds and IS in thecontrol human plasma C: Compounds and IS in the volunteer’s plasma. Figs 1–3 represent S-VEN, R-VEN andIS respectively.

Table 1. Extraction recovery, inter- and intra- day precision and accuracy of VEN enantiomers

Added drug Addedconcentration

Recovery Inter-day Inra-day

(ng/mL) Mean CV% Foundconcentration

Accuracy,RSD

Foundconcentration

Accuracy,RSD

(%) (ng/mL) (%) (ng/mL) (%)

S-VEN 2.8 92.7 6.8 3.0 7.7 2.6 5.370.5 106.3 5.3 63.4 6.4 74.9 2.7

282.0 102.7 6.1 292.7 5.2 289.6 4.6R-VEN 2.8 89.6 7.2 3.0 4.4 2.5 9.7

70.5 102.8 4.9 65.4 9.2 72.5 5.2282.0 103.5 6.7 289.7 6.2 291.9 8.3

n = 5 at each concentrations.

Table 2. Stability for VEN enantiomers

Added drug Concentration -20°C 3 months 4°C 24 h Room temperature 24 h(ng/mL) Average (%) CV % Average (%) CV % Average (%) CV %

S-VEN 2.8 92.1 9.5 103.1 7.3 95.8 9.770.5 97.7 7.1 97.8 8.1 101.7 8.5

282.0 106.6 7.2 99.6 5.7 104.6 4.2R-VEN 2.8 90.8 9.7 100.8 9.0 92.6 6.7

70.5 101.9 9.0 98.9 6.2 103.3 7.9282.0 97.8 8.1 105.8 6.7 96.4 5.1

n = 3 per test and each concentration.

415

Determination of venlafaxine enantiomers

Biomed. Chromatogr. 2011; 25: 412–416 wileyonlinelibrary.com/journal/bmcCopyright © 2010 John Wiley & Sons, Ltd.

in human plasma, the LLOQ of this method was relatively high(5.0 ng/mL) and the linear range was 5.0–400.0 ng/mL, which wasnot suitable for pharmacokinetic study. Because the lowest con-centration of VEN enantiomers in human plasma was 2.9 ng/mLin the present study (the concentration at 36 h after administra-tion of venlafaxine hydrochloride capsules), we had to establish amore sensitive method for the pharmacokinetic study. In thepresent study the LLOQ was 0.28 ng/mL for both S-VEN andR-VEN, which met the sensitive requirement.

In a single-dose study with nine subjects, S-VEN/R-VEN ratiosmeasured between 2 and 6 h did not suggest a stereoselectivedisposition of VEN enantiomers (Wang et al., 1992), while in Eapet al.’s (2000) report, at two blood samplings, the S-VEN concen-trations were approximately twice as high as those of R-VEN,strongly suggesting a stereoselective disposition. To our knowl-edge, no studies have so far measured the venlafaxine enanti-omers in the Chinese volunteers after a single regimen. AlthoughZhang et al. (2007) reported the pharmacokinetics of venlafaxinein healthy Chinese volunteers, they did not measure the twoenantiomers and research their pharmacokinetic parameters. Inpresent study, we found that the Cmax of the S-VEN was higherthan the R-VEN and there was a significant difference (p < 0.05)between them, which agrees with Eap et al. (2000). The AUC(0–•)

and t1/2 between S-VEN and R-VEN were also significantly differ-ent (p < 0.05).

ConclusionsIn conclusion, the described HPLC-MS/MS method is a sensitive,accurate and selective assay for the determination of venlafaxineenantiomers in human plasma and provides complete and reli-able measurements of the pharmacokinetics of the two enanti-omers studied. Comparing the pharmacokinetic parameters ofS-VEN with R-VEN, the AUC(0–•) of S-VEN and R-VEN were signifi-cantly different (p < 0.05). The Cmax of the S-VEN was higher thanthat for R-VEN and significantly different (p < 0.05). It is suggestedthat the R-VEN is metabolized faster than S-VEN.

ReferencesAndrews JM, Ninan PT, Nermeroff CB. Venalfaxine: a novel antidepres-

sant that has a dual mechanism of action. Depression, 1996; 4: 48–56.

Cherkaoui S, Rudaz S and Veuthey JL. Nonaqueous capillaryelectrophoresis–mass spectrometry for separation of venlafaxine andits phase I metabolites. Electrophoresis 2001b; 22: 491.

Desiderio C, Aturki Z and Fanali S. Use of vancomycin silica stationaryphase in packed capillary electrochromatography II. Enantiomerseparation of basic compounds. Electrophoresis 2001; 22: 535–543.

Eap CB, Berte-Laubscher R, Zullino D, Amey M and Baumann P. Markedincrease of venlafaxine enantiomers concentrations as a consequenceof metabolic interactions: a case report. Pharmacopsychiatry 2000; 33:112–115.

Fanali S, Rudaz S, Veuthey J-L and Desiderio C. Use of vancomycin silicastationary phase in packed capillary electrochromatography II. Enan-tiomer separation of venlafaxine and O-desmethylvenlafaxine inhuman plasma. Journal of Chromatography A 2001; 919: 195–203.

Holliday SM and Benfield P. Venlafaxine, a review of its pharmacology andtherapeutic potential in depression. Drugs 1995; 49: 280–294.

Liu W, Wang F and Li HD. Simultaneous stereoselective analysis of venal-faxine and O-desmethylvenlafaxine enantiomers in human plasma byHPLC-ESI/MS using a vancomycin chiral column. Journal of Chroma-tography B. Analytical Technology in Biomedicine and Life Sciences 2007;850: 183–189.

Morton WA, Sonne SC and Verga MA. Venlfaxine: a structurally uniqueand novel antidepressant. Annals of Pharmacotherapy 1995; 29: 387–395.

Moyer JA, Muthe EA and Haskins JT. In vivo antidepressant profiles of thenovel bicyclic compounds. WY-45,030 and Wy-45881. Society for Neu-roscience 1984; 10: 261.

Muth EA, Haskins JT, Moyer JA, Husbands GE, Nielsen ST and Sigg EB.Antidepressant biochemical profile of the novel bicyclic compoundWy-45,030, an ethyl cyclohexanol derivative. Biochemical Pharmacol-ogy 1986; 35: 4493.

Muth EA, Haskins JT, Moyer JA, Husbands GE and Andree TH. Biochemical,neurophysiological and behavioral effects of Wy-45,233 and otheridentified metabolites of the antidepressant venlafaxine. Drug Devel-opment Research 1991; 23: 191.

Rudaz S, Stella C, Balant-Gorqia AE, Fanali S and Veuthev JL. Simultaneousstereoslecetive analysis of venlafaxine and O-desmethylvenlafaxineenantiomers in clinical samples by capillary electrophoresis usingcharged cyclodextrins. Journal of Pharmaceutical and BiomedicalAnalysis 2000; 23: 107–115.

Wang CP, Howell SR, Scatina J and Sisenwine SF. The disposition of ven-lafaxine enantiomers in dogs, rats, and humans receiving venlafaxine.Chirality 1992; 4: 84–90.

Zhang W, Xiang BR and Wan CY. Liquid chromatography–mass spectrom-etry method for the determination of venlafaxine in human plasmaand application to a pharmacokinetic study. Biomedical Chromatog-raphy 2007; 21: 266–272.

Figure 2. Mean plasma concentrations-time curves of S-VEN and R-VENafter a single oral dose of 75 mg venlafaxine. n = 18, Mean � SD.

Table 3. One compartment of pharmacokinetic parametersof a single dose (75 mg) in Chinese volunteers

Pharmacokinetic S-VEN R-VENparameters Mean SD Mean SD

tmax (h) 2.5 0.9 2.3 0.9AUC(0-•) (mg h/L) 769.4 259.1 516.6 273.2Cmax (mg/L) 48.1 15.9 33.1 17.7Ka (1/h) 4.4 5.1 5.2 5.1Ke (1/h) 0.08 0.01 0.06 0.02t1/2 (h) 8.7 1.9 11.8 4.2Cl/F 3.13 1.33 3.6 1.5V/F (L) 37.6 14.3 57.2 21.9

416

W. Liu et al.

Biomed. Chromatogr. 2011; 25: 412–416wileyonlinelibrary.com/journal/bmc Copyright © 2010 John Wiley & Sons, Ltd.