Analytical Methods for Determination of Selective Serotonin Reuptake Inhibitor Antidepressants

Analytical methods for determination of selective serotonin reuptake inhibitor antidepressants

Z ü hre S¸ ent ü rk1 , Cafer Saka 2, * and I˙ brahim Te gˇ in 3 depression is an inadequate amount of serotonin, by prevent- 1 Department of Chemistry , Faculty of Science and Letters, ing the reuptake of serotonin with the presynaptic neuron. Y ü z ü nc ü Y ı l University, Van , Turkey Serotonin plays a part in the treatment of various disorders 2 School of Health , S ı ı rt University, S ı ı rt , Turkey , such as anxiety, depression, schizophrenia, pain, hyperten- e-mail: [email protected] sion, vascular disorders, and migraine (Asberg et al. 1976 ). 3 Department of Chemistry , Faculty of Science and Letters, SSRIs specifi cally prevent the reuptake of serotonin by S ı ı rt University, S ı ı rt , Turkey increasing the level of active serotonin in synapses. They have varying degrees of selectivity for the other monoam- *Corresponding author ine transporters, with pure SSRIs having only weak affi nity for the noradrenaline and dopamine transporter. Therefore, determination of SSRI pharmaceutical or biological samples Abstract are very important in pharmacokinetic, in therapeutic drug monitoring, and in bioequivalence studies. For these reasons, Selective serotonin reuptake inhibitors (SSRIs) are commonly reliable analytical methods are needed which reliably deter- used to treat depression. SSRIs are classifi ed as fl uoxetine, mine plasma levels of SSRI drugs in order to detect changes paroxetine, escitalopram, citalopram, sertraline, and fl uvox- (either desired or otherwise) as early as possible in the plasma amine. Several methods have been published for the determi- concentrations of the drugs themselves. SSRI drugs have side nation of SSRI drugs in pharmaceuticals, biological materials effects including sexual dysfunction, gastrointestinal effects, and environmental samples. This review covers the analyti- and disruption of the central nervous system (Khawam et al. cal methods described in the literature for the determination 2006 , Barlow and Durand 2009 ). of SSRIs and their main metabolites or some compounds In recent years, advanced analytical methods have been belonging to the same SSRI group. The analytical methods developed and optimized in the fi eld of pharmaceutical anal- are generally chromatography based methods coupled to dif- ysis, with the aim of improving precision and sensitivity, in ferent detectors, electroanalytical methods, capillary zone order to accurately quantify trace concentrations of pharma- electrophoretic methods, and spectrometric methods. ceuticals present in pharmaceutical formulations and biological fl uids. Keywords: analytical methods; antidepressant; pharmaceu- In this article, a review of the analytical methods for the tical and biological samples; selective serotonin reuptake determination of SSRI drugs is presented. SSRIs are clas- inhibitors. sifi ed as fl uoxetine (FL), paroxetine (PXT), escitalopram (ESCIT), citalopram (CIT), sertraline (SER), and fl uvoxamine (FLU). Analytical methods described in the literature for Introduction the determination of SSRIs and its main metabolites or some compounds belonging to the same SSRI group are generally Depression is a chronic or recurrent mood disorder that chromatographic based methods coupled to different detec- affects both economic and social functions of approximately tors, electroanalytical methods, capillary zone electrophoretic 121 million people worldwide. According to the World methods, and spectrometric methods. Additionally, analytical Health Organization, depression will be the second lead- methods such as the immunoassay and titrimetry methods ing contributor to the global burden of disease. Depression have also been described in the literature for the determina- can lead to suicide. Analyses of the risks of taking selective tion of SSRIs. However, these methods are not covered in serotonin reuptake inhibitors (SSRIs) have resulted in warn- the review. Figure 1 shows the chemical structures of SSRI ings about suicidality and aggression when these medications drugs. are used with children and adolescents. The US Food and Drug Administration (FDA) indicated that there is a 1.5-fold increase of suicidality in the 18 – 24 age group. This resulted in Sample preparation a black box warning on SSRIs and other antidepressant medications regarding the increased risk of suicidality in patients Several methods for sample preparation of SSRIs in phar- younger than 24 years (Levenson and Holland 2007 , Stone maceutical formulations and biological fl uids such as human and Jones 2007 , FDA 2008 , Wille et al. 2008 ). Antidepressants plasma, human urine samples, rat plasma, rat brain sam- are commonly used for the treatment of depression and obses- ples, fi sh tissue, hair samples, and human serum have been sive compulsive disorders. It is thought that one reason for described in the literature. The most common are liquid- 88 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

H N O NHCH3 O O

O CF3

F Fluoxetine Paroxetine

H CF NH 3 CH3

NH Cl N 2 O

OCH3 H Cl

Sertraline Fluxamine

CN CH 3 H3C O N CH 3 N F H3C

F N Citalopram Escitalopram

Figure 1 Chemical structures of SSRIs. liquid extraction (LLE) and solid phase extraction (SPE) with the multiple reaction monitoring (MRM) for the simultane- C18 and C8 columns. However, these methods present some ous determination of some drugs, including CIT, FL, FLU disadvantages and are laboriously time-consuming and use norﬂ uoxetine, and PXT in serum of 0.1 ml, which requires expensive solvents. Solid-phase microextraction (SPME), protein precipitation using acetonitrile/methanol. Acids such stir bar sorptive extraction (SBSE), protein precipitation and as trichloroacetic acid can also be used for protein precipita- direct injection of biological samples without sample prepa- tion prior to analysis of SSRI samples. Deproteinization using ration supported liquid membrane extraction (SLM), liquid- trichloroacetic acid 10 % (w/v) for the determination of FLU solid extraction (LSE), liquid phase microextraction (LPME), in plasma with recovery levels of 57.54% has been reported pressurized liquid extraction (PLE) have also been employed (Fernandes et al. 2006 ). This method demonstrated the effec- for SSRI determination. These methods show some disadvan- tiveness of simple precipitation of proteinaceous material tages being laborious and time-consuming. using an organic solvent.

Protein precipitation Liquid-liquid extraction (LLE)

Protein precipitation is the simplest means of bioanalytical LLE has been exploited as an extraction procedure for SSRIs sample pretreatment. It only involves the addition of a pre- from complex matrices. In a method published on the deter- cipitating solvent, subsequent homogenizing and centrifu- mination of SSRIs, human plasma or serum samples were gation. Deproteinization is rarely used in the extraction of prepared using a three-step LLE (Ulrich 2003 ). Massaroti SSRIs from biological matrices. Reubsaet and Bjergaard et al. ( 2005 ) presented a high-performance liquid chroma- (2004) used a protein precipitation method for screening of tography coupled with tandem mass spectrometry (LC-MS/ more than 70 central nervous system-stimulating drugs in MS) method for PXT quantiﬁ cation in human EDTA plasma human plasma, including CIT, PXT, and SER. Acetonitrile based on LLE using a mixture of ethyl acetate/hexane (50:50, was used to precipitate the proteins. Kirchherr and Kuhn - v/v) with retention times of 1.6 and 1.7 for PXT and FL, Velten (2006) developed another high-performance liquid respectively. Several papers have reported extraction with chromatography-mass spectrometry (HPLC-MS) method in acetonitrile methanol or prior to clean-up of the extracts by Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 89

LLE with hexane. In some cases, this procedure was followed 300 µl human plasma by using a SPE procedure with mixed- by SPE. SLM extraction and/or enrichment for selective sepa- mode cation exchange cartridges. The SPE cartridges were ration and concentration of various metal ions from aqueous activated by passing 1 ml of methanol through the cartridge dilute solutions is similar to LLE. SLM uses the relatively three times, and then conditioned by passing 1 ml of a pH small volume of organic components in the membrane and 6.0, 25 mm phosphate buffer three times. CIT was used as simultaneous extraction and re-extraction. An automated the internal standard. Linearity ranged in the plasma over the sample pretreatment of human blood plasma for liquid chro- 5.0 – 160.0 ng/ml for each FLU isomer. Precision was better matographic determination of three antidepressant drugs than 4.0 % . (dibenzepine, reboxetine, fl uvoxamine), based on SLM for In recent years, much attention has been focused on the unsurpassed sample clean-up and analyte enrichment has development of on-line techniques. For such on-line sys- been developed by Barri and J onss ö n (2004) . The entire ana- tems, SPE is generally preferred over LLE as the isolation lytical procedure revealed good linearity and low detection technique, because it is less laborious, uses smaller amounts limits of 5, 15, and 20 ng/ml for dibenzepine, reboxetine, and of organic solvent, yields better analyte enrichment and is fl uvoxamine, respectively. easier to couple on-line to the chromatographic technique to be used. As compared to off-line SPE, on-line SPE offers Solid-phase extraction (SPE) a series of advantages such as analysis of the total amount of analytes extracted, small sample volumes are suffi cient SPE is a separation process that is dissolved or suspended in to obtain enough sensitivity, matrix effects, ionic suppres- a liquid mixture. SPE uses the affi nity of solutes dissolved or sion or enhancement in MS spectrometry, less fl exibility, suspended in a liquid for a solid through which the sample is and most systems do not allow the combined use of differ- passed to separate a mixture into desired and undesired com- ent cartridges, automatization and minimal sample handling ponents. SPE involves liquid-solid partition. This technique which translates into better precision and accuracy, direct has been used extensively to extract trace organic materi- and fast elution of the sample after preconcentration, mini- als from samples. Sample pretreatment procedures adopted mal consumption of organic solvents and reduced analysis in most analytical methodologies for determining SSRIs time and high throughput. The use of on-line SPE techniques in pharmaceutical formulations and biological fl uids are has made possible the development of faster methods by based on solvent extraction, followed by clean-up by SPE. reducing the sample preparation time and thus increasing the Currently, the analyst can choose different SPE formats, such sample throughput (Bones et al. 2006 , Rodriguez -Mozas et as cartridges, disks, and in-tube capillary columns. The SPEs al. 2007). with either off-line or on-line preconcentration procedures, Saber (2009) presented a capillary liquid chromatography- coupled with various single element or multi-element analyt- electrospray ionization-mass spectrometry method (LC- ical techniques are frequently used in the modern analytical ESI-MS) including on-line SPE for quantifi cation of FL laboratory. The main sample pretreatment methods and tech- hydrochloride in human plasma with metronidazole (inter- niques currently being utilized for SSRIs are summarized nal standard). The within assay and between assay precisions in Tables 1 – 4 . For SPE, many commercial cartridges are were between 8.5 % and 11 % and 6.6 % and 7.5 % , respec- available. The sorbents used include the apolar C8, extrac- tively. For spiked plasma samples, the lower limit of quanti- tion disk or C18 and the mixed modes Bond Elut Certify or fi cation (LLOQ) value was ∼ 5.0 ng/ml. The limit of detection Oasis MCX. Oasis HLB, a hydrophilic-lipophilic balanced (LOD) value of the method was ~3.0 ng/ml. column, is also used in sample preparation before HPLC The column-switching system presents major advan- analysis of SSRIs. tages over earlier off-line methods because of elimination Wille et al. (2005) presented gas chromatography-mass tedious manual extraction and minimizes manipulation of spectrometry (GC-MS) and HPLC-diode array detection the biological samples. Liu et al. (2008) reported a fully (DAD) methods for 13 antidepressants, including FL, FLU, automated column-switching ion-pair HPLC-UV (254 nm) CIT, SER, and PXT, together with eight of their metabolites method for the analysis of FLU in rat plasma. The linearity from plasma by using the SPE procedure. This procedure for FLU ranged between 5 and 5000 ng/ml (r> 0.999). LOQ consisted of a conditioning step with 3 ml of eluent, 2 ml of and LOD values were 5 and 1.5 ng/ml. This method applies methanol, and 3 ml of the eluting phosphate buffer. The SPE the integrated sample clean-up confi guration using a RAM sorbents applied in this study can be divided into four different SPE pre-column connected via the electrically driven six-port categories: namely, apolar sorbents (Bond Elut C18, Empore switching column valve from a programmable autosampler to HD C8, and RPselect B Lichrolut), polymeric sorbents (Focus, a reversed-phase analytical column. The plasma sample was Strata X, and Oasis HLB), ion-exchange sorbents (strong and injected onto a pre-column packed with Shim-pack MAYI- weak cation exchangers), two mixed modes combining ion- ODS (50 µm), where the drug was automatically purifi ed exchange properties with C8 or a styrene-divinylbenzene and enriched by on-line SPE. Also, Souverain et al. (2003) polymer (Bond Elut Certify and Strata XC). Recoveries reported a LC-ESI-MS method for the simultaneous deter- ranged between 70 % and 109 % for all antidepressants. mination of FL and its primary metabolite (norfl uoxetine) in Saracino et al. (2006) reported a high-performance liquid plasma based on a column-switching approach with a pre- chromatography-ultraviolet detection (HPLC-UV) method column packed with large size particles in < 4 min. Flow rate for the simultaneous determination of FLU and quetiapine in was 4 ml/min. Linearity ranged between 25 and 1000 ng/ml 90 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs 2004 ) al. ) 1993 (Eap et al. 1996 ) (Addison et al. (Addison 1998 ) (Kim et al. 2002 ) (Ulrich 2003 ) (Wille et al. 2005 ) (Fernandes et et (Fernandes al. 2008 ) (Lai et al. ) 2000 2002 ) uoxetine) cients of varia- of cients 0.99 and0.99 precision of > 2 15 % 0.999 cient of < 66 % . Inter-day for coeffi % to 10 % tion: ranged from 4 FL for % to 13 % FLU and from 4 response was obtained in the range ng/ml10.0 with a correlation – 0.2of coeffi Intra-day precision: between 5.4 % % Intra-day precision: between 5.4 25, of at plasma levels % and 12.7 and100, 200 ng/ml the for four precision: Inter-day enantiomers. at ng/ml 100 % and 9.1 % between 5.3 Mean recoveries: % 91 (FL) and % 87 (norfl RSD Linearity: 0.469 – 120 ng/ml120 – Linearity: 0.469 (Leis et al. Linearity r of 2 ng/ml to % Recoveries: ranged from 50 SensitivityLOD values: 1.5 and 6 ng/ml Other parameters References 1.0 ng/ml1.0 ng/ml.100 – Linearity: 1 LOQ and LOD thermal with desorption: and 37 0.46 pg/ml LOQ: 28.4 ng/ml,LOQ: 28.4 LOD: 8.5 ng/ml LLOQ: 0.469 ng/ml with liquid liquid with desorption: 30.0 pg/mland 10.0 GC-MS Sub-ng/ml (Lamas et al. GC-MS electron (EI) impact mode SIM GC-MS SIM ng/ml 0.1 . The linear % 85 – Recovery: 80 and detector and GC-nitrogen- phosphorus detection GC-MS and HPLC-DAD GC-MS SIM with an electron impact GC-electron GC-electron capture detector GC-MS SIM LOD: ng/ml, 0.2 GC-EI-MS SIM LOQ and LOD lm lm rst rst m Varian µ c, Folsom, c, 0.53 mm0.53 i.d. × 0.33 mm0.33 fi m, Agilent × µ 0.25 mm i.d., 1.0 0.25 mm 1.0 i.d., × 0.2 mm i.d., 0.11 mm mm0.2 0.11 i.d., 0.25 mm 0.25 i.d., × × 0.25 mm 0.25 mm i.d., × lm 0.2 mm0.2 i.d. 0.25 mm 0.25 i.d., m fi lm thickness, Restek, Sulzbach, × × µ 0.25 mm coated i.d.) with a 0.25 mm m Macherey-Nagel, Oensingen,m Macherey-Nagel, lm thickness) lm lm thickness of 5 % phenylmethyl phenylmethyl % lm thickness 5 of lm thickness, ThermoQuest, Vienna, column (15 mcolumn (15 µ Switzerland) with crosslinked methyl silicone m (Ultra-1, 17 fi × Columncapillary) m (15 fi mm Germany) Separation mode factor Four ms VF-5 column (Varian, Middelburg, The Netherlands) 0.25 thickness capillary HP-5 column preceded a 0.5 m by deactivated methyl silica guard column fi silicone (J&W Scientifi CA, USA) umn (15 mumn (15 fi Austria) m, 0.25 mm, 0.25 Technologies) preparation LLE fused silica Rtx-1 capillary (fi SPME CP-SIL 8 CB 30 m, 0.25 mm i.d., Human plasma LLE GC-MS 1 ng/ml (Reymond et Plasma LLE Fused silica Optima 5 capillary Human plasma plasma Human or serum Human plasmaHuman SPE Environmental water m 30 GC methods. uoxetine CIT and its metabolites FLU and norfl FL SPE m 25 SER Human plasma LLE Fused silica capillary column coated PXT Human plasma LLE Fused silica capillary m column (15 FL and major metabolites Table 1 Table Analyte Matrix Sample FL, FLU, CIT, CIT, FLU, FL, SER, PXT and other drugs CIT, FL, FLU, SER PXT Human plasma LLE fused DB5-MS silica capillary col- FL Human plasma SBSE fused HP-5MS silica column (30 Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 91 (Nevado et al. 2005 ) (Fontanille et al. ) 1997 (Nevado et al. 2006a ) (Wille et al. 2007 ) (Nevado et al. 2006b ) (Wille et al. 2009 ) (Lefebvre et al. ) 1999 (Goodnough (Goodnough ) 1995 et al. (Pujadas et al. 2007 ) 0.45 for FL, for 0.45 ± 0.58 for FLU for 0.58 ± Recovery: 100.37 98.59 Recoveries ranged between 98.1 % % Recoveries ranged between 98.1 % and 102.7 Recovery ranged between 44.5 % % Recovery ranged between 44.5 % and 97.7 g/l FL, for µ g/g and liver brain Sensitivity Other parameters References FLU LOD values: 0.3 and 2 ng/ml 5.7 ng/l FL for 5.7 to 12.5 ng/ml 12.5 to 3.6 – 41.5 mg/l Brain to plasma from ranged ratios to 17. 0.8 Hair concentrations ranged from to 0.4 ng/mg 2.5 ing from 0.56 to mg/l)1.22 and liver (ranging from 0.04 ng/mg) 3.58 to LOD: 1.32 and 1.36 andLOD: 1.36 1.32 µ tissue 0.9 to 44.20.9 ng/ml LOQ values: 33.5, 300.0 ame ioni- ame and detector and GC-nitrogen- phosphorus detection GC-MS ng/l CIT, 0.7 for GC-MS LOQ ranged from 5 GC-MS LOQ: LOD, cal ionization mode GC-MS LOD: blood (rang- GC-nitrogen- phosphorus detection GC-MS SIM LOQ ranged from GC-fl detector zation m µ m, m, lm lm µ lm lm m m fi µ µ m (Lara-Spiral, µ 0.2 mm i.d., 0.11 mm0.2 0.11 i.d., × 0.32 mm0.32 with i.d., a fi 0.25 mm 0.25 i.d., 0.25 mm and i.d., 0.25 0.25 mm 0.25 i.d., × × × × lm thickness of 5 % phenylm- % lm thickness 5 of 0.32 mm coated0.32 i.d.) with a 0.52 × lm thickness, Agilent Technologies) Column Separation mode 25 m25 thickness 0.25 of Couternon, France) 15 m Supelco, Barcelona, Spain) J&W-5 ms column fromJ&W-5 Agilent USA) PA, (Avondale, Technologies apparatus equipped with an on-column injection and a 30 m, mm 0.25 column DB-5 i.d., m pm fi siliconeethyl (Ultra-1, 16.5 m(Ultra-1, 16.5 fi HP-5 (5 % phenylmethyl silicone, phenylmethyl % (5 HP-5 15 m column thickness) preparation SPE GC-MS chemi- LLE Varian Saturn II Ion-trap GC-MS LLE Fused silica capillary column (25 uid SPE capillary Methylsilicone column Human plasma LLE fused silica capillary OV-1 column, Human plasmaHuman SPE m 30 Pharmaceutical formulations Postmortem blood, brain tissue, and hair regions, blood, blood, regions, andliver hair of male rats Liver andLiver brain tissue formulations PXT, SERPXT, Oral fl Analyte Matrix Sample FL and major metabolites CIT, FLCIT, Urine samples SPE silicone, phenylmethyl % (5 Equity-5 CIT, FL, FLU, FLU, FL, CIT, andSER, PXT, other drugs FL, FLU, CIT, CIT, FLU, FL, SER and PXT FL, FLU, CIT, CIT, FLU, FL, SER and PXT and other drugs FL Discrete brain FL and other drugs (Table 1 (Table continued) FL, FLUFL, Pharmaceutical 92 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs li et et li (Bagli et al. 1997 ) 2000 ) 2000 al. 2007 ) 2004a ) 2009 ) 2008 ) al. 2006 ) al. ) 1998 et al. 2007 ) 2007 et al. Jonss ö n 2004 ) 1997 ) 2008 ) References 0.25 (Esrafi 1.0 (Melo et al. 1.0 et al. (Silva 1.2 et (Saracino 1.0 (Wirth et al. 1 (Liu et al. (ml/min) g/l µ g/l µ 2 LOQ: 5 ng/ml et al. (Maya LOD values ranged from 0.5 to 0.7 LLOQ: nmol/l 10 0.3 (Li et al. LOQ values ranged from 20 ng/ml to 50 ng/ml LOQ values ranged from 20 to 50 ng/ml LOQ and LOD: 15.0 ng/ml LOD: 3 ng/ml (Meineke et LOQ: 25 ng/ml 1.0 (Fernandes LOD: 20 ng/ml 0.18 (Barri and LOQ: 5 ng/ml and LOD: ng/ml 1.5 Sensitivity Flow rate ame ame HPLC-UV LLOD values 5 and of HPLC-UV HPLC-UV (226 nm) HPLC-UV HPLC-UV nm) (215 HPLC-UV HPLC-UV (226 nm) LC-UV LC-UV (230 nm) LC-UV LC-UV (230 nm) HPLC-UV HPLC-UV nm)(245 HPLC-UV HPLC-UV (225 nm) HPLC-UV HPLC-UV (227 nm) Ld-UV Ld-UV (230 nm) ionization detec- NMR tion HPLC-UV (260 GC-fl nm) HPLC-UV HPLC-UV (254 nm) and detector and

m containing 2 4 , pH 3.5) , pH 3.5) m PO 3 AcOH solution of pH 4.0 pH 4.0 of solution AcOH phosphate buffer pH of m m sodium 1-octanesulfonate m (58:19:23, v/v/v) (58:19:23, and MeOH (54:46, v/v) 0.02 67 mmol/l67 potassium phosphate buffer and (67:33, (pH 3.0) ACN v/v) ACN-0.1 % ACN-0.1 H Phosphate buffer solution mol/l0.05 and pH 3.8, of v/v) (53:47, ACN Phosphate buffer (0.05 solution v/v) mol/l, pH 3.8)-ACN (53:47, ACN (30 % ) and phosphate % (30ACN m buffer (10.5 containing % 0.12 triethylamine (70 ) % 6.0 adjusted6.0 triethylamine, ACN, MeOH (13:65:22, v/v/v) 0.01 0.01 ACN:acetate buffer 25 mmol/l with triethylamine 25 mmol/l (70:30)pH 4.6 triethylamine aqueous solution (adjusted phos- to pH 2.90 by phoric acid) Gradient mobile phase (water- ACN-TFA) m 35 % ACN and 65 % 10 m 10 % and ACN 65 % 35 (36:64, v/v) 4 × 2 m; m; 4 × µ ×

50 mm, × × m) m; Waters, m) µ µ µ l; l; Agilent, µ m) µ 3 mm 5 i.d., × m; Merck, Darmstadt, µ m particle size; Merck, m particle size) µ µ 2.1 mm,2.1 3.5 2.1 mm 5 2.1 i.d., 250 mm, 5 g particle size; 4.6 mm4.6 5 i.d., 4.6 mm,4.6 3 5 × × × × × × Zorbax SB-C8 column 0 (4.6 mm column (125 (125 column Merck) Wilmington, DE, USA) Australia) mm analytical column (4.6 (4.6 column analytical Tokyo, Japan) mm mm Germany) Darmstadt, (C18) column (250 mm(C18) 4 mm, 5 Germany) mm mm, 5 mm USA) Stationary phase Mobile phase Separationmode mm, packed with 3 LPME Zorbax column (100 Extend C18 LLE LiChrospher CN 100 (150 preparation Drug sub- Drug and stances formulated products Human serum LSE Ultrasep ES CN-column 100 ACN-MeOH-phosphate buffer Human plasma LLE LiChrospher 60 RP-Select B Urine, plasma and tap water samples Plasma samples SBSE LichroCART mm RP (125 18 Rat plasma SPE L-column octadecylsilane (ODS) Serum SPE Symmetry Waters C8 (150 Human plasma SPME LiChrospher 60 RP-select B Human plasma SPE Varian ResElut C8 column (150 Plasma samples SPME Phase Sep column (150 C18 HPLC methods with UV and DAD detection. and DAD detection. HPLC methods with UV FL FLU and PXT FL SER and other drugs CIT, PXT, FL, SER and other drugs FLU FL and major metabolites CIT, PXT, FL, SER and other drugs FLU and quetiapine FL and metabolites FL and its metabolite Table 2 Table Analyte Matrix Sample FLU Human plasma SLM column (250 Nucleosil C18 Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 93 al. 2003 ) al. 2003 ) al. 2003 ) al. 2007 ) References al. 2009 ) ) et al. 2008 al. 2007 ) al. ) 1995 1997 ) Ö ztun ç 2006 ) 1(Frahnert et et 1(Frahnert 1 (Ulu 2007 ) 1.0 (Malfara et (ml/min) 1.01.2 (Chaves et 1.0 (Cruz -Vera (Chaves et LOD: and 1 ng/ml 1.4 and LOQ: 5 and 2 ng/ml plasma for and urine LOD: 10.0 ng/mlLOD: 10.0 1LLOQ: 5 ng/ml 1.0 (Zainaghi et (Zainaghi et LOQ: 5 ng/ml CIT, for ng/ml FL,10 PXT, for SER Sensitivity Flow rate LOQ ranged between and 25 ng/ml16 LOD: ng/ml 90.1 for FL ng/ml 40.0 and ng/ml LLQ: 6 ng/ml 1 (Knoeller et LOQ: 5 ng/ml 0.35 (Foglia et al. LOD: 2 ng/ml 1.0 ( Ö nal and HPLC-UV HPLC-UV (230 nm) (450 nm) HPLC-UV HPLC-UV (230 nm) HPLC-UV HPLC-UV (254 nm) HPLC-UV HPLC-UV (205 nm) and detector and LC-UV LC-UV (230 nm) HPLC-UV (254 nm) LC-UV LOQ: between 10.0 HPLC-UV HPLC-UV nm)(295 HPLC-UV HPLC-UV (205 nm) (567 nm)(567

n l/l l/l PO µ 3 glacial AcOH m potassium dihydrogen Potassium dihydrogen m m phosphate of pH 7.0-ACN phosphate pH 7.0-ACN of (60:40, v/v) 65 % 0.05 phosphate buffer 85 % H % phosphate buffer 85 ACN/water (80:20, v/v) HPLC-UV 25 m25 (adjusted to pH 4.5 with 1 35 % of a mixture of ACN/ of % 35 of % MeOH (92:8, and v/v) 65 mol/l0.25 sodium acetate buffer 0.5 % Potassium dihydrogen phosphate (pH 2.5)-ACN (75:25, v/v) 0.05 mol/l0.05 sodium phosphate and buffer ACN pH 5.0 of (50:50, v/v) 0.02 of pH 2.5, ACN and pH 2.5, of ACN 125 Phosphate buffer (0.05 mol/l, pH v/v) and (57:43, 3.8) ACN phosphate dihydrogen Sodium buffer ACN-MeOH (pH 3.0) v/v/v) (70:25:5, different mobile phasesTwo including acetate buffer solution: ACN:MeOH NaOH) containing 2 g tetrabu- tylammonium sulfate hydrogen ACN-ethanol (3:2, v/v) % and 35 octylamine (62:38, v/v) ACN-water (70:30)ACN-water HPLC-UV × m m; m; m µ µ µ m, m, 4 µ × m; m; Thermo m; m; MASIS, m; m; Thermo µ µ m) µ 4 mm, 5 4 mm, 5 4 mm, 5 µ × × × – 5 – column 4 4.6 mm4.6 5 i.d., m particle size; Merck, × µ 4.6 mm4.6 5 i.d., 2 mm 5 i.d., 4.6 mm4.6 5 i.d., 4.6 mm4.6 5 i.d., 150 150 mm) × × × × × mm; Merck, Darmstadt, Germany) Phenomenex C18 column (250 C18 Phenomenex mm Separation, USA) (4.6 umn (250 Macherey and Duren, Nagel, Germany) (150 particle size; Merck) Owani, Japan) Owani, column (250 mm Merck) (150 Stationary phase Mobile phase Separationmode 4 mm, 5 Darmstadt, Germany) Darmstadt, column (250 mm particle size; Merck) column (250 mm (250 Separation, USA) preparation LLE LiChrospher 60 RP-Select B Human plasmaHuman SPE C Grand Pack and urine Urine samples SPEPlasma samples SBSE Eclipse column X-DB-C8 LichroCART mm RP (125 18 Serum SPE Nucleosil 100-Protect 1 col- Plasma samples SPME LiChrospher 60 RP-Select B PXT Human plasma LLE RP-Select B column (250 FLU, major FLU, metabolites (Table 2 (Table continued) PXTPXT Human plasma LLE Human plasma LLE LiChrospher 60 RP-Select B Beckman, Ultrasphere C column FLU plasma Human Analyte Matrix Sample Trazodone FL and SER, FL, CIT, PXT, and other drugs FLU,PXT, FL, SER, and CIT FL, SER, CIT andPXT, other drugs PXT Human plasma LLE column C18 Phenomenex CIT, PXT, FL, SER and other drugs 94 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs al. 2002 ) (Shah et al. 2007 ) al. 2003 ) et (Skibinski al. ) 2000 2004b ) et al. 2006 ) 2006 et al. al. ) 1998 al. 2006 ) al. 2006 (Gatti et al. ) 2003 al. ) 1997 References (Molander et al. ) 2001

-1 l µ 1.5 (Berzas et 5 min 0.3 (Li et al. 1.2 (Mandrioli 0.8 (El -Gindy et 1.0 (Holladay et (ml/min)

-2 -3 10 10 g/ml × g/ml g/ml × µ µ g/ml g/ml µ g/ml g/ml and pg/ml,

µ µ -3 -2

10 10 m × × )citalopram )citalopram µ + -(-)citalopram -( -( R S LOD values: 5.51 LOD values: 5.51 LOD and LOQ ranged andfrom 10 1.0 and 3.3 and 33.3 and and 3.3 33.3 to 0.26 LOQ: 14 nmol/lLOQ: 14 LLOQ PXT for and 0.6FLU: ng/ml 10 (Llerena et LLOQ serum for FLX and N-FLX 10 was (on-column nmol/l amount 200 of fmol) LOQ and LOD: 7.5 ng/ml and 2.5 ng/ml LOD: 13.37 LOD: 13.37 LOQ: 40.53 and 4.35 LOD: ng/ml 5.0 1.2 (Alvarez et LOQ values: 1.84 and 1.45 and 1.45 for for and LOQ: ng/ml 10 for enantiomer each Sensitivity Flow rate LC-UV LC-UV (240 nm) HPLC-PAD HPLC-PAD (230 250 – nm) LC-UV LOQ ranged from 0.05 HPLC-UV HPLC-UV (226 nm) HPLC-UV HPLC-UV and(293 nm) 253 LC-UV (226 nm) HPLC-UV HPLC-UV (220 nm) LC-UV LC-UV nm) (233 HPLC-UV HPLC-UV (227 nm) LC-UV LC-UV (227 nm) LC-UV LC-UV (226 nm) and detector and , pH 3.0 , pH 3.0 m aqueous -cyclodextrin -cyclodextrin m uorophosphate/ -phosphoric β

o m ammonium formate m potassium dihydro- m 10, v/v) + ACN-pH 2.5 phosphateACN-pH buffer (40:60, v/v) ACN-45 m v/v) pH 4 (25:75, of 0.1 % 0.1 triethylammonium acetate (adjusted pH 4.0 buffer, with acetic acid), and con- ACN, m 12 taining (90 ACN ACN (30 ), % water ), % (67 ), and 400 ml % acetate buffer (3 of dimethyloctylamine MeOH-tetrahydrofuran- phosphate buffer at pH 2.65 (0.0657 mol/l) (53:5:42, v/v/v) 67 mmol/l67 potassium phosphate buffer and (67:33, (pH 3.0) ACN v/v) phosphate buffer containing triethylamine v/v) (35:65, % 0.1 0.05 phosphategen buffer (pH 5.6 adjusted with acid)-ACN (50:50, v/v) Acidic aqueous (con- Acidic solution taining ml perchloric of 0.1 acid tetramethylammo- g of and 1.5 nium perchlorate per and liter) (58:42, v/v) ACN acetonitrile (55 % (55 % ACN-45 distilled water mcontaining 10 triethylamine) ACN and mACN a 12.3 4.6 300 × × 0.46 cm × m) m) µ µ m particle; µ ve micron ve particle m, Mid Glamorgan, µ end-capped column m; Waters, Milford,m; Waters, 8 µ m) µ 4.6 mm4.6 3 i.d., 4.0 mm4.0 i.d.; particle size, 2.1 mm 5 2.1 i.d., 150 mm,150 5 × × × m) m) fromm) Merck × µ µ -cyclodextrin Nova Pack C18 column (3.9 mm column (3.9 Pack C18 Nova i.d. Millipore,Waters Milford, MA, USA) Varian, C MA, USA) MA, Shim-pack (fi size) cyanopropyl column with β (100 column mm reversed-phase narrow bore reversed-phase column mm, 3.5 LiChrospher 100 RP-180, C18 C18 LiChrospher RP-180, 100 column (250 mm, mm, 4.0 i.d., 5 (125 4 (Rainin, Woburn, MA, USA) (15 column octa decyl 5 i.d., mm 4 i.d., Stationary phaseUK) Mobile phase Separationmode LLE Chiralcel column ODR Potassium hexafl preparation Pharmaceutical formulations Serum or homogenate from brain areas Plasma SPE Kromasil column (0.32 C18 Human plasma LLE Hypersil BDS C column Rat serum SPE Symmetry Waters C8 (150 Human plasma SPE Genesis C8 RP column (150 Pure powder powder Pure and tablet formulations uoxetine uoxetine (Table 2 (Table continued) CIT enantiomers FLU, PXTFLU, Human plasma FL,CIT, FLU SPEand other drugs Nova-Pak, LiChrolut RP-18 FLX and N-FLX CIT, FL,CIT, PXT and their metabolites FLX and N-FLX FL FL enantiomers FL Serum LLE Shimadzu, Microsorb MV FLX and norfl (N-FLX) SER and its main metabolite Analyte Matrix Sample FL and olanzapine Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 95 al. ) 2001 al. 2004 ) al. 2008 ) (Ferretti et al. ) 1998 et al. 2003 ) 2003 et al. et al. 2003 ) 2003 et al. References 2007 ) 1 et (Tournel 1.0 (Sabbioni et 1(Unceta et et 1(Unceta 1.0 (Vivekanand 1.0 (Duverneuil (ml/min) 1.2 (Reddy et al. g/ml g/ml g/ml, g/ml, µ µ g/ml µ LOD and LOQ: 2 and 6 ng/ml LOQ values ranged from ng/ml 15 and 50 ng/ml LOQ and LOD: 30 and 15 ng/ml LOD values close to LOD values close mg/l0.01 LOQ and LOD: 2 and 7.5 LOD values ranged from 2.5 to 5 ng/ml SensitivityLOQ: 0.001 LOD: 0.003 for FL Flow rate HPLC-PAD HPLC-PAD nm)(296 HPLC-PAD HPLC-PAD (200.4 nm) HPLC-PAD HPLC-PAD (230 nm) HPLC-PAD HPLC-PAD (230 nm) LC-PAD LC-PAD (265 nm) LC-UV (220, 240, and 290 nm) and detector and HPLC-PAD (225 nm) 0.1 0.1 ± , pH m sodium dihydrogen phos- dihydrogen sodium m -Hexane-ethanol-diethylamine n v/v/v) (94:6:0.5, 3.80) ACN and 17 mmol/l andACN 17 tetramethylammonium perchlorate pH 3.0 (50:50, v/v) TMACl (pH 4; 0.15 % ):ACN ):ACN % TMACl (pH 4; 0.15 (50:50, v/v) (50 % , v/v) ACN in a sodium ACN , v/v) % (50 phosphate buffer (0.05 Hexane, isopropanol, and dieth- v/v/v) (96:0.4:0.3, ylamine Acetonitrile-phosphate buffer 3.8 pH of 9.5 m phate (pH adjusted to 6.8 with triethylamine), and ACN v/v/v) (40:30:30, MeOH 0.4 4.6 m) × × µ 4.6 mm, 4.6 10 m) × µ m; m; Shimadzu, m; m; Beckman, µ µ 4.6 mm4.6 i.d.) × 4.6 mm4.6 5 i.d., m) andm) Chiralcel OJ × µ m; Tracer, Barcelona,m; Tracer, 4.6 mm, 4.6 10 µ 4.6 mm,4.6 5 × 4.6 mm,4.6 5 × × m), Chiralpak AD (250 (150 mm Carbamate derivative-based column (Chiralpak AD) cm, 5 Spain) µ mm, 10 (250 Inertsil RP column (150 C18 mm cm France) Gagny, (ThermoHypersil) analytic analytic (ThermoHypersil) column (250 Stationary phaseKyoto, Japan) Mobile phase Separationmode SPME cm Spherisorb (15 ODS2 LLE 5-microm Hypurity C18 preparation Human plasma SPE RP Microsorb column MV C18 samples and pharmaceutical formulations Pharmaceutical formulations Serum LLE column Beckman (25 ODS C18 FL and its metabolite CIT and FL Human urine PXT Raw material PXT LLE Chiralcel OD (250 CIT, PXT, FL, SER, FLU and other drugs (Table 2 (Table continued) FL and olanzapine PXT, FLU, SER, FL, otherCIT, drugs Analyte Matrix Sample 96 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs Mohammadi Mohammadi 2007 ) 1998 ) ) 2003 1998 ) 1995 ) (Higashi et al. 2005 ) 2003 ) 2003 2001 ) 2001 (Peyton et al. 1991 ) 2000 ) 2000 2008 ) ) 1999 References al. 2004 ) 1.2 (Guo et al. 1.5 (Lucca et al. 1 (Millan et al. 1.2 (Raggi et al. (ml/min) l µ g/ml µ in 100 (23 fmol) (Fukushima et m m of plasma LOQ: 10 n LOQ: 10 LOD: 5 ng/ml FLU, for and FL,PXT, and 10 ng/ml SER for LOQ: 0.5 ng/mlLLOD and LLOQ: 0.008 and 0.015 (Bahrami and thanLOD values lower ng/ml2.1 and ng/g 42.8 LOD: ng/ml 0.2 1.2LOQ: ng/ml 1.5 (Shin et al. LOD: 1 ng/ml 0.5 ng/mlLOQ: 2.5 (Raggi et al. LOQ: 5 ng/mlLOQ: 2.0 ng/ml 1 (Kosel et al. (Matsui et al. Sensitivity Flow rate LOD: 17 n 17 LOD: LOQ: ng/ml 0.96 (Macek et al. : 537 nm : 537 : 490 nm nm : 537 nm : 470 : 540 nm nm : 313 nm : 336 : 306 nm nm: 350 nm: 238 : 294 nm : 296 nm : 302 nm : 478 nm : 478 : 366 nm nm : 470 : 540 nm nm : 470 : 285 nm : 224 nm : 206 nm nm: 295 : 300 nm : 230 nm : 240 nm : 249 nm

ex em ex em ex em ex em ex em ex em ex em ex em ex em ex em ex em ex em ex em

HPLC-FLD λ HPLC-FLD λ HPLC-FLD λ λ HPLC-FLD HPLC-FLD λ and detector and λ λ λ λ λ λ λ λ λ λ λ λ HPLC-FLD λ HPLC-FLD λ HPLC-FLD λ HPLC-FLD λ λ λ λ LC-FLD λ LC-FLD λ CN CN 3 uoro O) ; pH 2.8) contain- 2 m phosphate buffer pH of tetramethylammonium potassium phosphate m m m ACN:aqueous tetramethyl ammonium perchlorate pH of (40:60,1.9 v/v) 17 m perchlorate with pH 1.9 of perchloric acid and ACN % 7 (45:55, v/v) m20 (70:30,4.6-ACN containing v/v) % 0.1 diethylamine Isooctane:THF (70:30, v/v) HPLC-FLD Gradient mobile phase consisting and ACN potassiumof phosphate mmol/l,buffer (10 pH 7.2) and in H TFA ACN (600ACN ml)-triﬂ in v/v) % (400AcOH ml) (0.1 water 10 m (40:60,buffer-ACN v/v) % adjusted with to pH 3.2 80 phosphoric acid ing 1 ml/l triethylamine (72:28) MeOH and sodium phosphate buffer (0.05 ACN-water (55:45, v/v) (55:45, ACN-water HPLC-FLD Hexane-isopropanol (82:18, v/v)Hexane-isopropanol (82:18, HPLC-FLD pH 6.0 buffer, ACN-phosphate (50:50, v/v) MeOH-AcOH-triethylamine (99.9:0.055:0.060, v/v/v) m; m; m 4.6 µ 0.46 µ × × m, m, m, µ m; m; 4.6 4.6 mm, µ m; m; µ × µ m, m, m particle µ 4 mm; Merck, µ × 3.0 mm 5 3.0 i.d., 4.6 mm,4.6 5 × × m, Littleton, CO, 4 mm, 10 µ 4.6 mm4.6 i.d. with 5 4.6 mm4.6 5 i.d., 4.6 mm4.6 5 i.d., 15 cm i.d., 5 cm i.d., 15 4.6 mm;4.6 Astec, Basel, × × × × × × m particles) µ Darmstadt, Germany) Darmstadt, umn (250 Torrance, USA) CA, mm Varian, Creek, Walnut CA, USA) Littleton, USA) CO, (4.6 ThermoQuest, Runcorn, UK) Runcorn, ThermoQuest, umn (250 (250 umn Stationary phase Mobile phase Separationmode (150 (250 mm particles; Kanto Chemical, Tokyo, Japan) size; Merck) Shimadzu, Japan) Shimadzu, mm i.d., 5 µm) from Tosoh Co. mm Co. 5 µm) i.d., from Tosoh Japan) (Tokyo, cm, 5 USA) 5 Switzerland) 150 SPELLE col- Phenomenex Luna C18 ResElut C8 RP column (150 LLE Apes Jones Silica column preparation LLE (25 column Silica Apex LLE LiChrosorb column RP-8 LLE (45 column Cyano LLE 5 Chirobiotic V, Rat brain column C18 Gradient mobile phases (CH Human plasma Human plasma Plasma Direct mm (4.0 RP-18 Human plasma Plasma col- Hypersyl ODS C18 Rat plasma packing material C18 column Rat plasma and brain tissue Human plasma Human serum LLE Shimpack (150 CLC-C18 Human plasma Human plasma HPLC methods with FLD detection. FL CIT FL and its main metabolite CIT FL FLU, PXT, PXT, FLU, SER, and FL FLU CIT and metabolites PXT FLU CIT CIT Table 3 Table Analyte Matrix Sample FL Rat plasma (250 ODS-80Ts TSKgel Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 97 Gauthier Gauthier 2005 ) (Vlase et al. 2005 ) Athanasiou et ) al. 2007 al. 2007 ) References al. 2002 ) 2007 ) and Dominguez ) 1999 et al. 1999 ) 1999 et al. (Lacassie et al. ) 2000 (ml/min) 1.5 (Waschgler et 1 (Unceta et al. 1.0 (Lopez -Calull 1.2 (Kristoffersen Sensitivity Flow rate LOD values: and 3.2 ng/ml2.1 in plasma, ng/g and 26.1 and 31.5 in brain tissue LOQ: ng/ml 12 LOQ and LOD: 0.71 and ng/ml 0.9 LOQ: 7 ng/ml (Meng and LOD: < 5 ng/ml LLOQ: ng/ml 10 1.0LOD: ng/ml 1.2 - (Vergi 0.5 (Mandrioli et LOQ: 0.025 mmol/l for PXT withCIT, FLD mmol/lLOQ: 0.10 for UV with FL LOD values and LOQ values ranged from 2.5 to 5 mg/l and from 10 mg/l 20 to : 305 nm nm: 325 nm : 312 nm: 350 nm: 350 nm: 330 : 227/300 nm : 227 nm : 250 nm : 230 nm nm: 295 nm: 295 : 294 nm : 260 nm

ex em ex em ex em ex em ex em ex em ex em

HPLC-FLD λ and detector and λ λ λ λ λ λ λ HPLC-FLD λ HPLC-FLD λ HPLC-FLD λ λ HPLC-FLD and UV HPLC-FLD λ LC-FLD λ HPLC-FLD and GC-nitrogen- phosphorus

4 PO 2 , pH m potassium -ACN (2:1, v/v) v/v) (2:1, -ACN m 4 3.5) (30:70, v/v) (30:70, 3.5) phosphate buffer PO = 2 m ammonium formate (pH ; pH KH (45:55, v/v) (45:55, 4 4) m m m = PO 3 (pH 0.04 dihydrogen phosphate buffer of v/v) pH 2.3 (31:69, ACN andACN 40 m buffer (KHACN-phosphate ACN-10 m of pH 3.2 adjusted pH 3.2 of with ortho H 45 m 4.0)-ACN (70:30, v/v) 66.7 % aqueous phosphate at pH % 66.7 ACN % 2.5 and 33.3 10 m 75 % 75 tetramethyl ammonium chloride in CAN Different mixtures sodium of acetate (0.005 solution 4.5) and MeOH 3.9 × m, m, m; m; µ µ m bonded 0.40 mm, 3.0 3.0 mm µ × × m; Waters, Milford,m; Waters, m; Phenomenex, USA) m; Phenomenex, µ µ 3.0 mm 3.5 3.0 i.d., 3.9 mm i.d.) 3.9 × × m column m particle size; Tracer µ µ Stationary phase Mobile phase Separationmode mm Agilent) 5 150 mm, 5 USA) MA, i.d., 5 i.d., umn packed with 5 silica (Beckman, USA) 5 Spain) Barcelona, Anal., i.d. (3 mmi.d. (3 dp) column preparation LLELLE Zorbax column (150 SB-C18 SPE Zorbax Eclipse XDB-C18 NovaPak C column (4 SPE Extrasil cm CN (15 SPE Symmetry C column (150 SPE Ultrasphere RP silica C18 col- SPE (250 mm Luna C18 Human plasma Human plasma Human plasma Human serum LLE column C18 v/v) (30:70, ACN-buffer HPLC-FLD Plasma and brain tissue Whole blood plasma and Deproteinized Deproteinized plasma Human plasma Human serum LLE Hypersil mm ODS, 15034.6 Analyte Matrix Sample FL and its active metabolite PXT PXT FL,CIT, and other drugs FL CIT, FL,CIT, and PXT, metabolites CIT PXT and its three main metabolites FLU, PXT, PXT, FLU, SER, FL, and other drugs (Table 3 (Table continued) 98 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs et al. 2005 ) 2005 et al. 2009 ) and and Schober ) 2003 2002 ) (Chen et al. 2006 ) Metcalfe 2007 ) et al. 2006 ) 2006 et al. and Kuhn - and Kuhn Velten 2006 ) et al. 2009 ) 2009 et al. al. 2007 ) et al. 2003 ) 2003 et al. and and Bjergaard 2004 ) References 0.150.05 (Massaroti (Saber 4(Kollroser 4(Kollroser 0.2 (Chu and 0.2 (Fernandes 1.0 (Kirchherr 0.2 (Franceschi 0.8 (Queiroz et (ml/min) 5.0 ∼ g/ml for µ 3.0 3.0 ng/ml ∼ g/ml and LLOQ: ng/ml LOD: 5 analytes all LOD: 2 ng/ml 1.2 (Shen et al. LLOQ CIT, for andFLU, PXT: 20, 20, and 10 µ LLOQ: 0.10 ng/ml LOQ values: and 0.24, 0.07, ng/g wet 0.14 PXT, for weight FL and its active metabolite LOQ and LOD: and 3 ng/ml10 LOD and LOQ: 0.06 and 0.17 ng/ml LOQ: 5.00 LOQ: 5.00 ng/ml Sensitivity Flow rate LC-ESI/MS LOD: LC-APCI-MS/ MS MRM LC-ESI-MS/MS MRM LC/APCI-MS/ MS MS-M SRM LC-APCI-MS/ MS LC-ESI-MS MIM HPLC-ESI- MS/MS MRM HPLC-ESI-MS MRM LC-ESI/MS SIM and detector and LC-ESI/MS 0.2 (Reubsaet

m ammo- m acetate m ammonium uoroacetate m ammonium m ammonium acetate in water ammonium acetate pH 5.0) of m m Water-ACN (95:5, v/v)Water-ACN LC-ESI/MS LOQ: 25 ng/ml 4 (Souverain ACN-0.05 mol/lACN-0.05 ammonium formate v/v) buffer (25:75, HCOOH 0.1 % in water ACN: (6:4, % HCOOH 0.1 v/v) ACN-water-HCOOH with 2 m ACN-water-HCOOH weight) ammoniumweight) trifl v/v/v) Formic acid (0.1 % ) mobile phase % Formic acid (0.1 HPLC-APCI- acetate) MeOH-acetate buffer 5 mmol/l pH of (60:40,4.8 v/v) ammonium acetate v/v/v) (68:32:0.1, acetate) and B (95 % ACN and 5 % % and ACN 5 % acetate) and B (95 water containing m 10 Gradient mobile of phases A (water mcontaining 10 buffer at pH 3.9 (40 % ) MeOH gradient and 5 m Gradient mobile phases (mobile m in 10 ACN % phase A: 5 nium acetate with AcOH; pH 5.0 of acetonitrile in % mobile phase B: 90 10 m 1 % Glacial AcOH in ACN (60 % ) and % Glacial (60 in AcOH ACN % 1 m20 2.1 2.1 × 2 × 2.0 mm, 4.6 mm4.6 (by % MeOH containing 0.075 m; m; Varian × × 3.9 mm,3.9 5 µ × 2.1 mm 3 2.1 i.d., × m) µ m; Brockville, ON, 150 mm150 HPLC i.d., µ × 4.6 mm;4.6 VWR/Merck, m) column × µ 2.1 mm, 5 mm,2.1 5 2.0 mm 100 Å , 3 Å 2.0 mm 100 50 mm; Agilent, Palo Alto, × × m particle size; Varian) m, m, 3.0 × µ µ m; Supelco, Bellefonte, PA, USA) m; Supelco, PA, Bellefonte, m particle size; Dublin, Waters, µ i.d. USA CA, column (50HS C18 5 mm column 5 mm, 3 mm 4 i.d., Canada) Darmstadt, Germany) Darmstadt, Chromolith Speed C18 ROD mm(50 Intersil column ODS-3 (50 mm Norway) Oslo, Holger, Stationary phase Mobile phase Separationmode µ Ireland) switching LLE column (50 Polaris C18 SBSE (100 Luna Phenomenex C18 precipitation precipitation preparation Human plasmaHuman Column- Human plasma Symmetry (Waters, USA) C18 plasma samples Fish tissue SPE mm column (150 Genesis C18 Serum Protein Human plasma plasma Human Protein HPLC methods with MS detection. FLFL and its metabolite Human plasma SPE columns Zorbax mm 0.3 C18 FLSER Human plasma LLE Human plasma LLE 250 CHIROBIOTIC V, Zorbax column Eclipse XDB C18 MeOH/water/formic acid (75:25:0.1, FL Serum samples LLE column Beckman (ODS-150 C18 CIT, FLU, FLU, CIT, PXT PXT Human EDTA FL plasma Human PXT, FL andPXT, its active metabolite FL Serum samples SPME mm column (150 C18 CIT, FL, FLU, FLU, FL, CIT, PXT and other drugs CIT, PXT, SER and other drugs Table 4 Table Analyte Matrix Sample Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 99 2009 ) 2005 ) Neto et al. 2008 ) (He et al. 2005 ) al. 2002 ) 2002 ) (Jia et al.(Jia 2007 ) References et al. 2001 ) 2001 et al. al. 2003 ) Neirinck 2002 ) et al. 2006 ) 2006 et al. Eerkes Eerkes ) 2003 al. 2004 ) 0.85 (Juan et al. (ml/min) 0.35 (Sutherland 0.25 (Segura et 0.22 (Zhu and 1.0 (Kovacevic 0.5 and (Weng g/l µ g/l; µ LLOQ: 0.5 ng/ml 0.35 (Patel et al. LOD values: 0.5, ng/ml and0.3, 0.1 LOQ: 1 ng/ml 0.05- (Santos LOQ: 0.5 ng/ml LLOQ: 0.5 ng/ml 0.6 (Green et Sensitivity Flow rate LOD: 0.334 ng/ml LLOQ: 0.15 ng/ml LOD and LOQ: g/l0.20 and 0.70 LLOQ: 0.2 ng/ml LOQ LC-MS: for 2.5 LOQ for LC-FLD: 20 LOQ: 0.50 ng/ml 0.15 (Pistos et LC-TIS-MS/MS MRM HPLC-ESI/MS SIR LC-ESI-MS/MS MRM HPLC-ESI-MS SIM LC-TIS-MS/MS MRM and detector and LC-ESI-MS/MS SRM MRM MRM LC-MS/MS MRM LC-MS-SIM LC-MS-SIM LC-FLD and LC-ESI-MS SIM LC-ESI-MS/MS MRM ammonium m ammonium acetate in ammonium formate l/min m µ m Ammonium formate – m CN-water (274:276, v/v)CN-water (274:276, LC-TIS-MS/MS ng/ml LOD: 0.1 0.2 (Li et al. 3 uoroacetate trifl 750 ml MeOH–250750 ml deionized water–2.5 ml, 1.0 20 mmol/l Ammonium acetate- % with bufferAcOH 70 pH 5.4 of at 5 ACN , ammonium % (HCOOH Water 0.6 acetate: 30 mmol/l)-ACN (35:65, v/v) Methanol-10 mmol/l ammonium acetate solution-acetonitrile v/v/v) (62:28:10, Solvent A: 0.05 % HCOOH in MeOH % A: 0.05 Solvent formic acid % B: 0.05 Solvent CAN-5 m (50:50) CAN-0.02 % HCOOH (340:660, v/v)% CAN-0.02 LC-TIS-MS/MS ACN-0.02 % HCOOH (66:34, v/v)% ACN-0.02 HPLC-ESI-MS (4:3, v/v) ACN-water-triethylamine (35:65:0.4, v/v/v) ACN-40 m 10 m HCOOH (pH 4.5) and (30:70, ACN v/v) CH water-TFA (94:6:0.05, v/v/v) water-TFA 2 × 2.1 2.1 × 2.1 2.1 2.1 2.1 , 5 × ° , × 2.1 2.1 mm, m × µ 4.6 4.6 mm, × m; m; µ 2.1 mm 3.5 2.1 i.d., m; m; Germany) × µ 4.6 mm4.6 i.d.) m; Chadds Ford, PA, × MAX-RP 2 mm, 4 µ m) µ × µ m) m, column; Phenomenex, c, Bellefonte, PA, USA) PA, Bellefonte, c, µ µ 4.6 mm,4.6 5 m, m, 150 3.0 mm Keystone 3.0 i.d., × m) m; m; Zorbax) µ × µ µ m; YMC Europe, Schermbeck/ m particle size; Thermo-hypersil, rial column (pore size 120A µ Weselerwald, Germany) (5 mm, 5 mm mm, 3.5 mm, 5 Torrance, USA) CA, 80A (150 80A Phenomenex, Aschaffenburg,Phenomenex, Germany) 4 Stationary phase Mobile phase Separationmode Scientifi 50 column (250 mm µ USA) mm 5 i.d., USA) 5 preparation Plasma packing YMC mate- ODS-AQ Human plasma LLE Betasil mm C8 column (100 Human plasma LLE Luna Phenomenex (150 C 18(2) Human plasmaHuman LLE 4 Synergi Human plasma SPE column Xterra (50 MS C18 Human plasma LLE Zorbax column cm (5 SB-C18 FL and other drugs SER Human plasma Shimadzu column ODS C18 FL, PXT CIT, Human plasma SPE column (250 C18 Macherey-Nagel SER Human plasma column Discovery C18 Formic acid-acetonitrile % 0.1 SER and its metabolite FL and its metabolite PXT and its metabolite PXT Human plasma LLE column (50 Genesis C18 Analyte Matrix Sample FL and its metabolite PXT Human plasma LLE Betasil silica column (5 CIT Human plasma LLE Hypersil C8 micro-bore BDS (Table 4 (Table continued) FL Human plasma LLE Eclipse XDB C8 (150 FL and its metabolite 100 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs al. 2006 ) al. 2007 ) al. 2004 ) 2005 ) et al. 2005 ) 2005 et al. References (Hattori et et (Hattori al. 2005 ) al. 2006 ) ) 2008 et al. (Shinozuka ) 2006 et al. ) 2007 et al. al. 2008 ) ) 2007 et al. (Jawecki 2007 ) 0.5 (Smyth et 0.5 (Jain et al. (ml/min) 0.2 et (Sauvage 0.4 Castro (de 0.05 (Frison et 0.5 (MacLeod , -9 10 × , 9 , and -7 -7 mol/l uid,and -8 10 10 × × 10 × uent of 1.6 1.6 uent of 6 ng/luent of 1.56 LOQ: 2.5 ng/ml 0.8 (Djordjevic Sensitivity Flow rate 7.8 ng/ml serum and ng/ml10 plasma LLOQ: 2 ng/ml in oral fl ng/ml2, 4 or 10 0.03 and 0.63 mg/ml LOQ: 5 ng/ml (Castaing LOD: pg/mg 10 LOD values infl ng/land and 1.7 LOQ values 0.1 infl LOD values: 6.25

n ow LC-MS/MS3 ng/ml – LOD: 1 (Hattori et LC-ESI/MS LLOQ: ng/ml 1.0 1.0 (Singh et SIM and detector and LC-MS/MS LOD values: 5 LC-ESI-MS/MS MRM LC-SSI-MS LOD between LC-ESI-MS MRM LC-ESI-MS/MS LLOQ: 25 pg/mg HPLC-ESI-MS/ MS MRM HPLC-ESI/MS HPLC-ESI/MS ESI-QToF-MS/ MS MRM LC-TIS-MS SRM HPLC, MS HPLC, (227 nm) PAD Turbulent-fl LC-MS ) m ammonium ammonium acetate m m OAc, 0.1 % HCOOH, pH 4 % 0.1 OAc, 4 ammonium acetate m ammonium acetate (pH m NH uoroAcOH) uoroAcOH in water, and 0.01 % % in and water, uoroAcOH 0.01 m Solvent A: 7.5 m A: 7.5 Solvent acetate B: aqueous solvent solution; HCOOH in ACN % 0.05 2.0 m 2.0 (pH 5.0)-ACN (54:46, v/v) Acetate buffer (40:60, and ACN v/v) LC-ESI/MS Gradient and ACN ammonium of formate 2 m (pH 3, (pH 5.0):ACN (70:20:10,(pH 5.0):ACN v/v/v) mate buffer mmol/l, (4 pH 3.2) Gradient mobile phases (solvent: HCOOH, ammonium % 0.1 water, formate 2 mmol/l pH 3 and of solvent HCOOH, ammonium % 0.1 B: ACN, formate 2 mmol/l) (90:10) (solvent:MeOH/water/HCOOH/ trifl 5 m (20:80,3):ACN v/v) Gradient mobile phases (0.01 % % Gradient mobile phases (0.01 trifl in ACN) TFA MeOH:nanopure water with 20 m Gradient of mobile phase A (0.1 % % Gradient mobile of phase A (0.1 HCOOH in and water) mobile phase HCOOH in ACN) % B (0.1 4.6

m, m, × µ µ 4.6 4.6 mm 1 mm, 5 × × m) ) 4 mm 3 4.6 mm)4.6 Gradient mobile phases µ µ × m) × µ m particle size; Whippany, µ 2.1 mm,2.1 3.5 4.6 mm, 4.6 5.0 4.6 mm,4.6 5 re C18 IS column (20 re C18 2.1 mm2.1 i.d.; Milford, Waters, × × × m column (Merck) × µ m; Phenomenex, Torrance, CA, mm i.d.; Shimadzu, Japan) Kyoto, µ USA) mm LiChroCART (55 (55 LiChroCART particles; Merck) 5 Stationary phase Mobile phase Separationmode i.d., 5 i.d., USA) NJ, XTerra MS C18 column (5 XTerra MS C18 50 USA) MA, mm mm LLE LiChrospher RP-8 100 E 250-4, preparation SPE Chirobiotic V (250 mm Direct SPE Sunﬁ uid and uid Hair samples LLE (150 Luna column C18 Hair samples column (150 C18 Human serum Direct Shim-pack MAYI-ODS (10 samples Pharmaceuticals wastewater in Human serum or plasma plasma Blank serum sample ﬂ Oral Human plasma LLE column XTerra RP18 Gradient ACN/ammonium of for- CIT and other drugs FL, CIT, PXT, SER, and other drugs SER, FLU, PXT Escitalopram Human plasma LLE ODS YMCTM column (150 AQ FL Rat liver Purosphere RP-18, STAR FL plasma Human Analyte Matrix Sample CIT, FL,CIT, and other drugs (Table 4 (Table continued) SER,PXT, FLU CIT, FL, PXT, SER and other drugs FL, PXT, SER, FLU, and CIT FLU, PXTFLU, Human plasma SPE Inertsil C8 column m Methanol:10 SER, PXT, FL, and CIT, FLU, other drugs SER Human plasma SPE Beta C8 column Basic Type (100 Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 101

with a determination coefﬁ cient higher than 0.99. LOQ values were 25 ng/ml for FL and norﬂ uoxetine. al. 2007 ) (Gutteck and Rentsch ) 2003 (Moraes et al. ) 1999 References ) 2007 et al. Solid-phase microextraction (SPME)

SPME can be thought of as a very short GC column turned inside out. SPME involves the use of a ﬁ ber coated with an 0.5 (Doherty et (ml/min) extracting phase that can be a liquid (polymer) or a solid (sor- bent), which extracts different types of analytes (including both volatile and non-volatile) from different types of media g/l (de Castro

µ that can be in liquid or gas phase (Somenath 2003 ). The attrac- tion of SPME is that the extraction is fast and simple and can be done without solvents, and detection limits can reach parts LLOQ values: between and 1.2 54 nmol/l the for different drugs LOQ: 0.15 ng/mlLOQ: 0.15 and 0.50 ng/ml SensitivityLOQ: 10 Flow rate per trillion (ppt) levels for certain compounds. Lamas et al. (2004) developed a SPME-GC-MS method for the analysis of venlafaxine, FLU, FL, CIT, and SER

n on direct SPME at 100° C by using polydimethylsiloxane

ame ame (PDMS)-divinylbenzene ﬁ bers in environmental water with LOD values in the sub-ng/ml level. In these experiments, the MRM LC-ESI-MS SIR HPLC-ESI/MS – MS ESI GC-ﬂ ionization detection LC-ESI-MS/MS MRM and detector and sample mode [direct extraction SPME and headspace (HS) mode (HS-SPME)], the introduction of an in situ acetylation step to transform the analytes in less polar compounds, and ° °

0.1 % 0.1 the extraction temperature (25 C and 100 C) were studied. + Fernandes et al. (2007) reported another HPLC-UV method by using SPME for determination of FL and its metabolite norfl uoxetine in plasma samples with a heated liquid fl ow through the interface. SPME conditions were optimized employing a factorial design. The sampling step was performed using a PDMS-DVB fi ber and desorption was carried

0.1 % HCOOH] B and solvent % 0.1 out in a novel homemade heated interface. LOD was 10 ng/ml + ammonium carbonate hydrogen

m for FLU and 5 ng/ml for nor-FLU. The range was evaluated Gradient mobile phase including A [MeOH-water (20:80,solvent v/v) v/v)[MeOH-water (90:10, HCOOH] Gradient mobile phases including 10 m and(pH 10) ACN from 25 ng/ml to 500 ng/ml. Recovery values were different for concentrations of 25, 100, and 500 ng/ml because plasma

4.6 proteins progressively adsorb on the ﬁ ber coating, decreas- × ing the extracted amount. Because the assessment had started with the lowest concentration (25 ng/ml) the recovery was lower for the highest concentration (500 ng/ml). m; Phenomenex,

µ In-tube solid-phase microextraction (in-tube SPME) has been successfully applied to the analysis of drugs in biological ﬂ uids. In-tube SPME has been used in HPLC as an ) µ

2.0 mm, 5 effi cient and simple preparation method, and it offers several × advantages over the fi ber SPME syringe-LC approach. In-tube Torrance, USA) CA, mm mm, 5 Stationary phase Mobile phase Separationmode SPME is similar to fi ber SPME, but the extraction device has a piece of fused silica GC capillary column in place of a fi ber. An inner surface coated capillary is commonly used as the extraction phase, such as the commercial GC capillaries

preparation and the newly developed ones, such as polypyrrole coatings. Organic compounds in aqueous samples are directly extracted and concentrated into the capillary columns stationary phase by repeated draw/eject cycles of the sample solution, being further transferred to the liquid chromatographic column (Wang et al. 2004 , Queiroz et al. 2007 ). Serum LLE column Reversed-phase C18 ACN-ammonium acetate buffer pH 4 HPLC-ESI-MS Human plasma Human Human plasma SPEHair samples Gemini guard column C18 (4 mm column (150 Luna C18 Queiroz et al. (2007) reported a LC-MS method at ion monitoring mode with immunoafﬁ nity in-tube SPME for analysis of FL in serum samples with LOQ of 5.00 ng/ml. Linearity ranged between 5.00 and 50.00 ng/ml with correlation coef-

uoxetine ﬁ cients better than 0.998. Silva et al. (2008) presented an in- CIT, FLU, and FLU, CIT, other drugs FLX and norﬂ SER, FL, CIT, FLU,PXT, and other drugs Psychoactive drugs (Table 4 (Table continued) Analyte Matrix Sample tube SPME-LC-UV (230 nm for most of the drugs) method 102 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

for simultaneous determination of mirtazapine, CIT, PXT, at pH 2 were poorly recovered and remained principally in duloxetine, FL, and SER in human plasma with quantifi cation the organic phase. For these drugs, two-phase LPME could be limits ranging between 20 and 50 ng/ml. Linearity for most of used as an alternative technique, where the aqueous acceptor the drugs ranged from 50 to 500 ng/ml with correlation coef- phase is replaced by an organic solvent. In the solubility range fi cients higher than 0.9985. of 1 – 5 mg/ml, most drugs were effectively extracted (recovery >30 % ), whereas drugs belonging to the solubility range of Stir bar sorptive extraction (SBSE) 5– 150 mg/ml were all extracted with recoveries above 30% by three-phase LPME. SBSE is a new solventless sample preparation method for the Esrafi li et al. (2007) presented a hollow fi ber-based LPME extraction and enrichment of organic compounds from aque- combined with a HPLC-UV method for extraction and deter- ous matrices. The method is based on the same principles as mination of three antidepressant drugs (amitriptyline, imip- SPME. Compared with SPME, a relatively large amount of ramine, and SER) in urine, plasma, and tap water samples. extracting phase is coated on a stir bar. It uses a small amount The extraction was performed owing to pH gradient between of solvent and is based on the use of a stir bar incorporated in the inside and outside of the hollow fi ber membrane with cali- a glass tube coated with a PDMS layer. The stir bar is placed bration curves obtained in the range of 5 – 500 µ g/l. in the aqueous sample and the analytes distribute between this matrix and the PDMS phase during the stirring. The Pressurized liquid extraction (PLE) amount of PDMS can vary depending on the bar length and fi lm thickness. The technique has been successfully applied to PLE involves the use of liquid solvents at elevated tempera- trace analysis in environmental, biomedical, and food appli- tures (40 – 200° C) and pressures (1000– 2500 psi). Under these cations. This technique is a very sensitive tool for the deter- conditions, solvents have enhanced solvation power and have mination of volatile and semi-volatile compounds. However, increased the extraction rates. A LC-atmospheric pressure the combination of SBSE with liquid desorption and HPLC chemical ionization (APCI)-MS/MS method for the determi- is also possible, providing an attractive approach for the nation of residues of PXT and FL, and its active metabolite, analysis of thermolabile solutes as well as higher molecular norfl uoxetine, in fi sh tissue including extraction of tissue by mass compounds (Baltussen et al. 1999 , Tienpont et al. 2002 , PLE is presented by Chu and Metcalfe (2007) . The LOQ val- David et al. 2003 ). A SBSE/LC-UV method for the determi- ues were 0.24, 0.07, and 0.14 ng/g wet weight for PXT, FL, nation of antidepressants (mirtazapine, CIT, PXT, duloxetine, and norfl uoxetine, respectively. The procedure for sample FL, and SER) in plasma samples with extractions based on preparation includes extraction of tissue by PLE, followed by both adsorption polypyrrole and sorption PDMS mechanisms clean-up on a mixed-mode SPE cartridge, Oasis MCX. With have been developed by Melo et al. (2009) . The LOQ values the optimized method, matrix interferences were reduced and ranged from 20 ng/ml to 50 ng/ml. recoveries > 85 % were obtained.

Liquid phase microextraction (LPME) Liquid-solid extraction (LSE)

LPME has been combined with LC and CE, in addition to the In LSE, a solvent is added to a solid. Insoluble material can be general way used by coupling to GC, and has been applied separated by gravity or vacuum fi ltration, and soluble mate- to various matrices, including biological, environmental, and rial is extracted into the solvent. A sequence of solvents, of food samples. In recent years, a porous-walled polypropylene varying polarity or pH, can be used to separate complex mix- hollow fi ber has been used to support the organic phase in tures into groups. the pores of the wall while holding the second aqueous phase Reymond et al. (1993) presented methods based on GC and in the lumen. According to the hollow fi ber-LPME method, GC-MS for the determination of levels of CIT, desmethylci- a porous polypropylene hollow fi ber acts as the interface talopram, and didesmethylcitalopram in plasma using LSE. between the donor contaminated sample and micro-volumes The demethylated amines are derivatized with trifl uoroacetic of the acceptor extractant phase solution. The advantages anhydride, and the acid metabolite with methyl iodide. The of using this microporous membrane include protection of LOQ values were 1 ng/ml for CIT and desmethylcitalo- the acceptor phase as well as effi cient sample microfi ltra- pram and 2 ng/ml for the other metabolites. Also, an iso- tion through the pores of the hollow fi ber (Zhu et al. 2001 , cratic reversed-phase HPLC for simultaneous determination Psillakis and Kalogerakis 2003 , Pedersen -Bjergaard et al. of FLU and PXT in human serum involving a pre-column 2005 , Ratola et al. 2008 ). technique for the on-line LSE was reported by Bagli et al. Pedersen-Bjergaard et al. (2005) presented the extract- (1997) with direct injection of serum samples and for their ability of 58 different basic drugs, including SSRIs, by three- preconcentration. phase LPME. The basic drugs were extracted from 1.5 ml In summary, there are many different ways to extract SSRIs water samples (pH 13) through approximately 15 µ l of dode- from pharmaceutical or biological sample matrices. However, cyl acetate immobilized within the pores of a porous polypro- sometimes compromising issues such as screening methods, pylene hollow fi ber (organic phase), and into 15 µl of 10 m m time and cost issues are required so that a simple extraction HCl (acceptor solution) present inside the lumen of the hollow system might be more suitable than a more complex extrac- fi ber. Compounds with a calculated solubility below 1 mg/ml tion with higher recoveries. Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 103

Analytical methods Nevado et al. (2005) reported a capillary gas chromatographic using fl ame ionization detector method for the Many analytical methods for the detection and quantitation of determination of FL, FLU, and clomipramine without deriva- SSRIs and their metabolites in pharmaceutical formulations tization steps in several pharmaceutical formulations. LOD and biological fl uids have been reported. When classifi ed and LOQ values were 10.1, 105.3, and 40.0 µ g/l and 33.5, upon separation methodology, the analytical methods can be 300.0, and 80.0 µ g/l for FL, FLU, and clomipramine, respec- categorized into chromatographic and non-chromatographic. tively. Recoveries were 100.37± 0.45 for FL, 98.59± 0.58 for Included within the chromatographic methods are GC, HPLC FLU, and 100.13± 0.32 for clomipramine. coupled with various detections by UV, fl uorescence detec- Ulrich (2003) developed direct enantioselective assay of tion (FLD), mass MS, DAD, and capillary electrophoresis FL and norfl uoxetine in human plasma or serum by two- (CE). Included within the non-chromatographic methods are dimensional capillary gas-liquid chromatography by using electroanalytical methods and spectrometric methods. three-step LLE for sample preparation. LOD values ranged between 1.5 and 6 ng/ml, respectively. Fluvoxamine and Chromatographic methods nisoxetine were used as internal standards. Linearity ranged between 5 and 250 ng/ml for (R )- and (S )-FL as well Several analytical methods have been developed for the as 15 and 250 ng/ml for (R )- and (S )-norfl uoxetine. Intra- analysis of SSRIs and in some cases also their metabolites day precision ranged between 5.4 % and 12.7 % at plasma in biological fl uids. Most of these methods are based on levels of 25, 100, and 200 ng/ml for the four enantiomers, reversed-phase HPLC coupled to UV, MS detection and on and inter-day precision ranged between 5.3 % and 9.1 % at FLD. GC is also applied in SSRI quantifi cation coupled to 100 ng/ml. nitrogen-phosphorous, fl ame ionization detector, and MS. A method for determination of PXT levels in human plasma by using GC-electron capture detection is presented by Lai et al. (2000) . LOD and LOQ values for PXT were 8.5 ng/ml and Gas chromatographic (GC) methods GC provides high 28.4 ng/ml. Fontanille et al. (1997) reported a GC method separation effi ciency, fast analysis, automation capabilities with nitrogen-phosphorus detection for direct analysis of FL and generally requires a small sample injection volume with and norfl uoxetine in plasma including an extraction with a commercially available specifi c or universal detectors. In mixed organic solvent and injection into a capillary GC with typical GC analysis, a defi ned sample volume is injected into an OV-1 fused-silica column coupled to a nitrogen-phospho- the preheated inlet and almost instantaneously volatilized. The rus detector. The calibration curves were linear over the range volatilized compounds are moved by the carrier gas towards 5– 3000 ng/ml with detection limits of 0.3 and 2 ng/ml for FL the column for separation. The separation process is based and norfl uoxetine, respectively. on their relative affi nity to the stationary phase of the column GC-based methods provide high resolution and low LOD (capillary or packed) under a thermal gradient program values, but they are labor-intensive and costly. A typical (Korytar et al. 2002 , Santos and Galceran 2002 , Rompa approach to overcome this limitation is the application of et al. 2003 , Marriott 2005 ). GC is a technique frequently derivatization protocols using various reagents in organic used for the determination of SSRIs. Determination of low medium. In this way, the analytes volatility and thermal sta- volatile compounds can be improved by derivatization. For bility are increased and the mass fragmentation pattern is derivatization, acid anhydrides, benzyl halides and alkyl improved when MS detectors are employed. chloroformates are preferred. Reagents most often used A capillary GC-MS in selected-ion monitoring (SIM) mode for derivatization of SSRIs are trifl uoroacetic anhydride for the analysis of CIT, FL, and all of their metabolites in (Reymond et al. 1993 ), ( S )-(-)-N -trifl uoroacetylprolyl chloride urine samples including an optimized SPE with LOD values (Eap et al. 1996 ), trifl uoroacetic anhydride (Addison et al. between 0.7 ng/l for CIT and 5.7 ng/l for FL was reported by 1998 ), heptafl uorobutyric anhydride (Lai et al. 2000 , Kim Nevado et al. (2006a) . The GC-MS method with SIM shows et al. 2002 ), pentafl uorobenzyl carbamate (Leis et al. 2002 ), signifi cantly lower detection limits than all of the methods ethyl chloroformate (Fernandes et al. 2008 ), N -methyl- N - previously discussed, by approximately 1000. Imipramine (trimethylsilyl)trifl uoroacetamide trimethylsilyl (Pujadas et hydrochloride was used as internal standard. The biological al. 2007 ), and heptafl uorobutyrylimidazole derivatives (Wirth samples were treated to achieve the extraction of FL, norfl u- et al. 1997 ). oxetine (NFL), CIT, and its metabolites by a SPE procedure Prior to the application of LC-MS, GC-MS had been the using a reversed-phase cartridge C18. The best recoveries most commonly used technique for the analysis of SSRIs and were obtained with phosphate buffer, methanol:water (30:70, their metabolites in biological fl uid. Even now, high-resolu- v/v) and subsequent elution with methanol. An enrichment tion GC directly combined with the wide range of low-cost factor of 10 was provided by this extraction-preconcentration bench-top MS instruments remains the most attractive tech- procedure for the analytes in the biological samples (5 ml of nique. There are GC methods which include several detection urine samples/0.5 ml of fi nal extract). systems available for the determination of SSRIs. They can Eap et al. (1996) developed a GC-MS for the simultaneous be divided into four different groups according to their use: determination of the plasma concentrations of FLU and of the nitrogen-phosphorous detection, fl ame ionization detector, enantiomers of FL and norfl uoxetine after derivatization with electron capture detector, GC-MS or GC-MS/MS. the chiral reagent, (S )-(-)-N -trifl uoroacetylprolyl chloride. 104 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

LOQ values were 2 ng/ml for FLU and 1 ng/ml for the (R )- performed on moderated as nonpolar stationary phases using and ( S )-enantiomers of FL and norfl uoxetine. almost exclusively packed columns with silica gel chemically A GC-MS method for the determination of FL, and its bonded with OS or ODS groups. Most platforms use reversed- major metabolite, norfl uoxetine involving SPE followed by phased columns with either C8 or C18 material. Additionally, acetylation with trifl uoroacetic anhydride and analysis of the a CN column, a phenyl-hexyl column and a polar reversed- derivatives using SIM with LLOQ of 1.0 ng/ml is presented phase are described (Tables 2 – 4 ). by Addison et al. (1998) . Currently, separation and analyses of SSRIs by HPLC are Leis et al. (2002) presented an improved sample work-up carried out with the mobile phase of methanol (MeOH) or and derivatization procedure to the pentafl uorobenzyl car- acetonitrile (CH3 CN, ACN), phosphate buffer solution and bamate derivative in one step for the quantitative determi- water containing a small amount of different additives (acid, nation of PXT negative ion chemical ionization in human base, neutral salts), such as formic acid (HCOOH), trifl uoroa- plasma by GC-MS subsequently analyzed without any further cetic acid (TFA), acetic acid (AcOH), diethylamine (DEA), purifi cation with the lower limit of detection (LLOD) of 0.2 triethylamine (TEA), phosphoric acid (H3 PO4 ), ammonium ng/ml plasma. LLOQ was 0.469 ng/ml plasma for phar- acetate (NH4 AcO), and ammonium formate. The pH range macokinetic measurements. In addition, Kim et al. (2002) covers (close to) neutral (pH 6 – 7) and acidic (in the pH range described a method for the determination of SER in human of 1.9 – 5) conditions depending on the detection method and plasma by using GC-MS with the SIM mode using single on the SSRI drugs of interest (Tables 2 – 4 ). LLE at alkaline pH after deproteinization of plasma protein For preparative purifi cation of SSRIs, a column size of and perfl uoroacylation with heptafl uorobutyric anhydride. 150 ×4.6 mm i.d. can be a good compromise between cost and The LOD value was 0.1 ng/ml and recovery was 80 – 85 % . sample-loading capacity. HPLC columns with i.d. ranging A method obtained in the order of pg/ml of LOD and LOQ from conventional (4.6 mm), narrow-bore (2.1 mm) are used was presented by Fernandes et al. (2008) with in situ deriva- for SSRIs analyses. Tables 2 – 4 show representative HPLC tization by using ethyl chloroformate on-line coupled to chromatograms of SSRIs. The fl ow of the eluting solvents GC-MS for the analysis of FL by using SBSE in plasma sam- is usually 0.2 – 1.5 ml/min, either in isocratic or in gradient ples. The developed method using liquid desorption showed modes, but fl ows up to 4.0 ml/min have also been used. linearity of r 2 >0.99 and precision of RSD < 15 % . The LOQ and LOD values with liquid desorption were 30.0 and 10.0 pg/ml, HPLC methods with UV and DAD A growing number respectively. LOQ and LOD values with thermal desorption of publications describe the use of GC-MS, HPLC-MS and were 0.37 and 0.46 pg/ml, respectively. Thermal desorption HPLC-MS/MS for the determination of SSRIs and their also demonstrated precision (RSD< 12 % ). metabolites, showing that these are techniques of choice. Pujadas et al. (2007) developed a GC-MS method for However, satisfactory results can also be obtained using identifying and quantifying psychoactive drugs in oral fl uid HPLC-UV, which is cheaper and less complicated. HPLC- including antidepressant drugs (amitryptiline, PXT, and SER) UV methods are very popular for the determination of using the SPE procedure and derivatization with N -methyl- SSRIs in biological samples. UV detection (Knoeller et al. N -(trimethylsilyl) trifl uoroacetamide. LOQ was between 0.9 1995 , Foglia et al. 1997 , Holladay et al. 1997 , Alvarez et al. and 44.2 ng/ml oral fl uid for the different analytes. Wille et al. 1998 , Meineke et al. 1998 , Maya et al. 2000 , Skibinski et al. (2007) reported a GC-MS method for the simultaneous deter- 2000 , Molander et al. 2001 , Frahnert et al. 2003 , Gatti et al. mination of antidepressants (mirtazapine, viloxazine, venla- 2003 , Llerena et al. 2003 , Ohkubo et al. 2003 , Zainaghi et al. faxine, trazodone, CIT, mianserin, reboxetine, FL, FLU, SER, 2003 , Barri and Jonss ö n 2004 , Li et al. 2004a,b , El -Gindy maprotiline, melitracen, and PXT) and their active metabo- et al. 2006 , Mandrioli et al. 2006 , Ö nal and Ö ztun ç 2006 , lites in plasma by using different ionization modes with Saracino et al. 2006 , Chaves et al. 2007 , Esrafi li et al. 2007 , sample preparation consisting of a strong cation exchange Fernandes et al. 2007 , Malfara et al. 2007 , Shah et al. 2007 , mechanism and derivatization with heptafl uorobutyrylimi- Ulu 2007 , Cruz -Vera et al. 2008 , Liu et al. 2008 , Silva et al. dazole. LOQ ranged between 5 and 12.5 ng/ml in electron 2008 , Chaves et al. 2009 , Melo et al. 2009 ) or DAD (Ferretti impact and positive ionization and 1 and 6.25 ng/ml in nega- et al. 1998 , Tournel et al. 2001 , Berzas et al. 2002 , Duverneuil tive ionization. et al. 2003 , Vivekanand et al. 2003 , Sabbioni et al. 2004 , Table 1 shows the distribution of the principal GC determi- Reddy et al. 2007 , Unceta et al. 2008 ) has been used in most nation (Reymond et al. 1993 , Goodnough et al. 1995 , Eap et analytical methods for the determination of SSRIs. SSRIs al. 1996 , Fontanille et al. 1997 , Wirth et al. 1997 , Addison et have a commonly absorption maximum in the range between al. 1998 , Lefebvre et al. 1999 , Lai et al. 2000 , Kim et al. 2002 , 220 nm and 300 nm, depending on the composition of mobile Leis et al. 2002 , Ulrich 2003 , Lamas et al. 2004 , Nevado et phase, especially in the presence of buffers of various pH al. 2005, 2006a,b, Wille et al. 2005, 2007, 2009, Pujadas et values. The wavelength applied most often has been 230 nm, al. 2007 , Fernandes et al. 2008 ) strategies established for the then 205 nm or 254 nm (Table 2 ). Generally, HPLC-UV quantifi cation of SSRI drugs. assays for the determination of SSRIs in biological materials were developed with LOQ values ranging between 2 and 10 HPLC methods Different stationary phases have been ng/ml. Mobile phases generally include acetonitrile, methanol, employed for the bioanalysis of SSRIs. In contrast to “ normal water, and acetate tampons. One limitation of UV detection is phase ” , the “ reversed phase ” HPLC mode of separation is the relatively poor sensitivity and the requirement that good Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 105

detection limits require a high-absorbing UV complex for (dibucaine) with chloroform. Intra- and inter-day variation good sensitivity in the direct detection mode. coeffi cients were in the range of 1.9 – 9.4 % and 2.3 – 13.3 % , Applications of HPLC-UV and HPLC-DAD for the respectively. LOQ and LOD values were 0.5 and 0.2 ng/ml. deter mination of SSRI drugs over recent years are listed in Another HPLC method for the determination of CIT and two Table 2 . metabolites was reported by Raggi et al. (2003) in human plasma by using SPE with 1 ml MeOH. LOQ was 1.5 ng/ml HPLC methods with FLD The necessity of determining for CIT and desmethylcitalopram and 2.0 ng/ml for didesm- very low levels of the SSRIs and their native fl uorescence ethylcitalopram. Also, Raggi et al. (1999) presented a HPLC led us to the choice of HPLC with FLD as a sensitive and method for the determination of FL and its main metabo- selective method for this purpose. The possibility of using lite norfl uoxetine in human plasma with LLE and the inter- HPLC-FLD for SSRI determination therefore depends on nal standard (PXT). The LOD value was 1 ng/ml for both the availability of suitable derivatizing reagents. The amino fl uoxetines. group in the structure of SSRIs such as FL facilitates the A chiral LC method for separation of the enantiomers of derivatization with acyl chloride reagents such as dansyl CIT and its two N -demethylated metabolites by using alpre- chloride. The nonchiral fl uorogenic reagents having a 2,1, nolol as internal standard in human plasma was reported 3-benzoxadiazole (benzofurazan) moiety possess long excita- by Kosel et al. (1998) . LOQ values were 5 ng/ml for each tion and emission wavelengths, which minimize interference enantiomer of CIT and demethylcitalopram. Matsui et al. by native fl uorescence from endogenous substances so that (1995) presented a HPLC method with a column-switch- highly selective and sensitive detection can be achieved ing technique for simultaneous determination of CIT and (Guo et al. 2003 ). Pre-column derivatization with dansyl its four metabolites in plasma by direct plasma injection. chloride (Lucca et al. 2000 ), 4-chloro-7-nitrobenzofurazan Unceta et al. (2007) developed a reversed-phase LC method (Bahrami and Mohammadi 2007 ), 4-fl uoro-7-nitro-2,1,3- for determination of FL and norfl uoxetine racemic mix- benzoxadiazole (Fukushima et al. 2004, Higashi et al. 2005), tures. LOD values were 3.2 and 2.1 ng/ml in plasma, and 4-(N -chloroformylmethyl- N-methyl)amino-7-nitro-2,1,3- 31.5 and 26.1 ng/g in brain tissue for FL and NFL, respec- benzoxadiazole (Guo et al. 2003 ) as fl uorescence label has tively. Also, a HPLC method for the simultaneous deter- been reported for HPLC-FLD analysis. FLD has also been mination FL and the n-desmethyl metabolite, norfl uoxetine applied to the chiral separation and analysis of the enantiomers in rat brain microdialysis samples by using pre-column of SSRIs. The pre-column chiral derivatization with (-)-(R )-1- derivatization with 4-fl uoro-7-nitro-2,1,3-benzoxadiazole (1-naphthyl)ethyl isocyanate has been reported for HPLC-FD was reported by Fukushima et al. (2004) . The LOD values analysis (Peyton et al. 1991, Millan et al. 2008). for FL and norfl uoxetine were approximately 17 and 5 n m When the pre-column derivatization of SSRIs was per- (23 and 7 fmol). formed using dansyl chloride as a fl uorescent reagent a large Applications of HPLC combined with FLD including chro- sample volume (1 ml) was required. 4-(N -Chloroformyl- matographic conditions and validation data for analysis of methyl- N-methyl)-amino-7-nitro-2,1,3-benzoxadiazole as the SSRI drugs reported in recent years are summarized in Table reagent was adopted to measure levels of SSRIs in a volume 3 (Peyton et al. 1991 , Matsui et al. 1995 , Kosel et al. 1998 , of rat plasma as small as 100 µ l (Guo et al. 2003 ). However, Shin et al. 1998 , Kristoffersen et al. 1999 , Lopez-Calull and a disadvantage of this method is the reaction time, which Dominguez 1999, Raggi et al. 1999 , Lacassie et al. 2000 , takes approximately 2 h at 60 ° C derivatization and was time- Lucca et al. 2000 , Macek et al. 2001 , Waschgler et al. 2002 , consuming. Also, 4-fl uoro-7-nitro-2,1,3-benzoxadiazole as Guo et al. 2003 , Raggi et al. 2003 , Fukushima et al. 2004 , the reagent was adopted to measure levels of SSRIs in rat Higashi et al. 2005 , Meng and Gauthier 2005 , Vlase et al. plasma in 100 µ l of plasma at 60° C for 5 min and injected 2005 , Bahrami and Mohammadi 2007 , Mandrioli et al. 2007 , into HPLC. Retention times of FLU and an internal standard Unceta et al. 2007 , Vergi -Athanasiou et al. 2007 , Millan et (propafenone) derivative were 15.5 and 13.5 min, respec- al. 2008 ). tively (Guo et al. 2003 ). The detection wavelengths are commonly carried out by FD using nm in the range excitation of HPLC methods with mass spectrometric detection 224 – 470 nm and emission of 300 – 500 nm. The HPLC-FLD HPLC-MS working with SIM acquisition mode in positive method using 4-chloro-7-nitrobenzofurazan as a pre-column electrospray ionization (ESI) and MRM are currently derivatization agent and an internal standard (FL) could be preferred for detection of SSRIs in biological materials. The very convenient for bio-analytical purposes with its LOQ more reliable identifi cation can be obtained by LC coupled to value of 0.5 ng/ml. tandem MS with a triple quadrupole. For mass spectrometric A HPLC-FLD method for the quantitation of citalopram in detection, the analytes have to be ionized fi rst. A variant of human plasma was reported by Macek et al. ( 2001) involv- ESI using a heated gas for desolvation of the eluent is called ing LLE of CIT with hexane-isoamyl alcohol (98:2, v/v) and turbo ion spray (TIS) or heated electrospray ionization. back-extraction of the drug to 0.02 m hydrochloric acid with APCI ionization is based on a principle generating ions via LOQ of 0.96 ng/ml using 1 ml of plasma. chemical ionization in the gas phase. APCI-MS, including Shin et al. (1998) developed a reversed-phase HPLC electrospray (ESI) and APCI in selected reaction monitoring method for the measurement of PXT in human plasma by (SRM) or MRM mode, has been established as a sensitive using only one-step extraction of PXT and internal standard and selective analytical technique for SSRIs. 106 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

Different ionization techniques were used for MS analyses drugs. Linearity ranged between 25 and 2000 pg/mg. Flow of SSRIs in combination with reversed-phase HPLC, includ- rate was 50 µl/min and the injection volume was 8 µ l. The ing APCI (Shen et al. 2002 , Kollroser and Schober 2003 , Chen acceptance range for recovery was set at 70 – 120 % for both et al. 2006 , Chu and Metcalfe 2007 ), ESI (Moraes et al. 1999 , the internal standard and the analyte. Zhu and Neirinck 2002 , Gutteck and Rentsch 2003 , Segura ESI-MS is currently the most common LC-MS method et al. 2003 , Souverain et al. 2003 , Weng and Eerkes 2003 , in SSRI analyses. Compared to APCI-LC-MS, ESI-MS has Pistos et al. 2004 , Reubsaet and Bjergaard 2004 , Singh et al. a fairly high sensitivity and is associated with lower back- 2004 , Djordjevic et al. 2005 , He et al. 2005 , Juan et al. 2005 , ground. Table 4 presents a summary of some representative Massaroti et al. 2005 , Fernandes et al. 2006 , Kirchherr and LC-MS methods for SSRI analyses, and the general thought Kuhn -Velten 2006 , Smyth et al. 2006 , Castaing et al. 2007 , de is that for MS, acidic eluents are most appropriate as they Castro et al. 2007, 2008 , Doherty et al. 2007 , Jia et al. 2007 , are able to donate a proton. Therefore, acidic eluents with MacLeod et al. 2007 , Queiroz et al. 2007 , Frison et al. 2008 , ammonium acetate, ammonium formate, formic acid, acetic Santos -Neto et al. 2008 , Franceschi et al. 2009 , Saber 2009 ), acid, and even TFA are often described for the bioanalysis of and thermo ion spray (Sutherland et al. 2001 , Green et al. SSRIs. 2002 , Li et al. 2002 , Jain et al. 2005 , Patel et al. 2009 ). Franceschi et al. (2009) reported an isocratic reversed- APCI is described for PXT, FL and its active metabolite, phase HPLC-ESI-MS/MS method in MRM for simultaneous norfl uoxetine in fi sh tissue including extraction of tissue by quantifi cation of FL and its major active metabolite in serum PLE by Chu and Metcalfe (2007) . A specifi c and sensitive samples by using extraction in a simple three-step LLE ml/min HPLC-APCI-MS/MS method in SRM mode was developed with run time of less than 5 min for each injection. Linearity by Kollroser and Schober (2003) for the rapid identifi cation for FL and its major active metabolite ranged between 0.3 and and quantitative determination of CIT, FLU, and PXT in 50 ng/ml. The LOQ value was 0.17 and 0.18 ng/ml for FLU human plasma. After dilution with 0.1% formic acid, plasma and N-FLU, respectively, and the LOD value was 0.06 ng/ml samples were injected into the LC-MS/MS system. The inter- for both analytes. and intra-assay coeffi cients of variation for all compounds Zhu and Neirinck (2002) presented a LC-MS/MS method were < 11 % . The total analysis time was 6 min per sample. for the determination of PXT in human plasma involving a LLOQ values for CIT, FLU, and PXT were 20, 20, and 10 LLE with cyclohexane-ethyl acetate at a fl ow rate of 0.22 µ g/ml. LOD values were 5 µ g/ml for all analytes. ml/min with total run time of 2.2 min. The LLOQ value was Shen et al. (2002) reported a high-throughput sample pre- 0.2 ng/ml. The standard curve was linear over a working paration procedure using LLE in a 96-well plate format in range of 0.2– 50 ng/ml and absolute recovery was 70.8% for conjunction with LC-APCI-MS/MS in the MRM mode for PXT and 84.1 % for the internal standard with accuracy of quantifi cation of FL enantiomers in human plasma. After inter-assay and intra-assay accuracy ranging between 24.8 % addition of internal standard and ammonium hydroxide, and 20.5% and 23.4% and 4.8% , respectively. LC-MS/MS samples were extracted with ethyl acetate. The organic equipped with an ion spray source was selected and used as extract was evaporated to dryness and reconstituted in meth- Q3 ion to be monitored. anol. Adequate separation of FL enantiomeric pairs (resolu- A narrow-bore LC-MS method for the quantifi cation of CIT tion of 1.17) was achieved on a vancomycin column eluted in human plasma using internal standard (imipramine) and with methanol containing 0.075 % (by weight) ammonium LLE with a mixture of hexane-heptane-isopropanol (88:10:2, trifl uoroacetate. v/v/v) was developed by Pistos et al. (2004) with run time Chen et al. (2006) developed a more sensitive LC-APCI- of 10.0 min. Linearity ranged between 0.50 and 250 ng/ml MS/MS method for the determination of SER with a LLOQ (r2 > 0.996) with LOQ of 0.50 ng/ml. value of 0.10 ng/ml using only 0.25 ml of plasma. The analyte Kirchherr and Kuhn -Velten (2006) developed another and internal standard (diphenhydramine) were extracted with HPLC-MS method in the MRM mode for the simultaneous 3 ml of diethyl ether/dichloromethane (2:1, v/v) from 0.25 ml determination of some drugs including CIT, FL, FLU norfl u- plasma, then separated on a Zorbax Eclipse XDB C18 col- oxetine, PXT, and SER in a small sample volume of 0.1 ml umn using methanol/water/formic acid (75:25:0.1, v/v/v) as of serum which requires only protein precipitation and step- the mobile phase. The triple quadrupole mass spectrometry wise dilution for sample preparation. LOQ values for CIT, was applied via an APCI source for detection. The method FL, FLU, PXT, and SER were 1.00, 2.17, 1.17, 1.07, and 0.70 was linear over the concentration range of 0.10 – 100 ng/ml. ng/ml. Once the ions have entered the mass spectrometer, detection Li et al. (2002) also developed a LC-TIS-MS/MS method in takes place. This can be accomplished by means of selecting MRM mode (m/z 314 – 44) for the determination of FL and its the molecular ion in SIM, but preferably a typical fragment main active metabolite norfl uoxetine in human plasma using ion is selected in MRM. The methods using SIM describe a deuterated FL as internal standard with a run time 5 min. The lower LOQ than the methods using MRM (Table 4 ). Frison et method was validated and concentration ranged from 0.27 to al. (2008) developed a LC-MS method SIM with positive ion 22 ng/ml with coeffi cients of correlation > 0.999. The LOD ESI for the determination of CIT, escitalopram, and their dem- value was 0.1 ng/ml for plasma FL and norfl uoxetine. ethylated metabolites in 10 mg hair samples including LLE Sutherland et al. (2001) presented another LC-MS/MS with diethyl ether/dichloromethane using clomipramine-d3 method in the MRM mode for the simultaneous determination as an internal standard with a LOQ value of 25 pg/mg for all of FL and its major active metabolite norfl uoxetine in plasma Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 107

with extraction from alkalized plasma with hexane-isoamyl ionization quadrupole-time-of-fl ight tandem mass spec- alcohol (98:2, v/v) and back-extraction into formic acid (2 % ). trometry (ESI-QToF-MS/MS) behavior of antidepressant TIS ionization was used for ion production with run time of drugs (CIT, FL, mirtazapine, PRX, SER, and venlafaxine). 2.60 min and retention times of 2.35 and 1.97 min for FL and Following Soxhlet extraction, the presence of SER in a hair norfl uoxetine, respectively. The mean recoveries for FL and sample was confi rmed using HPLC/ESI-MS2 analysis based norfl uoxetine were 98% and 97% , respectively, with LLOQ on its retention time on the HPLC column (17.71 min) and its set at 0.15 ng/ml for the analyte and its metabolite. fragmentation characteristics using mass spectrometry (m / z Green et al. (2002) developed a LC-MS method with MRM 306 – 275). The concentration of SER in this hair sample was using the TIS interface operating in positive ion mode. For calculated to be 1.90 ng/mg. This procedure was repeated for the determination of FL and norfl uoxetine in human plasma, the identifi cation and quantifi cation of paroxetine in a hair automated SPE on Oasis HLB cartridges with run time of sample and yielded a result of 0.25 ng/mg. 4 min was used. The method was linear over the range of Doherty et al. (2007) developed methods on HPLC-ESI/MS, 0.5– 50 ng/ml (using a sample volume of 0.5 ml) with a LLOQ ESI – MSn , GC-fl ame ionization detection and polarographic value of 0.5 ng/ml. behavior in the determination of selected drug compounds in Patel et al. (2009) developed a LC-MS/MS method in posi- hair samples. GC-fl ame ionization detection was carried out tive ion and MRM mode with TIS for simultaneous determi- on a Zebron ZB-5 (Phenomenex, Macclesfi eld, Cheshire, UK) nation of SER its primary metabolite ( N -desmethyl SER) in column (30 m× 0.25 µ m). In polarographic analysis, the current human plasma using FL as internal standard by applying LLE axis was set from 0 to -160 nA and cathodic scans were initiated in methyl tert-butyl ether. Run time was 2.5 min. Linearity from -0.4 V to -1.6 V using 0.1 m acetate buffer (pH 4.47) as ranged from 0.5 to 150 ng/ml for SER and N-desmethyl background electrolyte and a standard Ag/AgCl 3 m KCl elec- sertraline with mean correlation coeffi cient of 0.9993 and trode system. Applications of HPLC combined with various MS 0.9980, respectively. LLOQ was 0.5 ng/ml. detectors including subjected sample, analyte, chromatographic Santos -Neto et al. (2008) reported a capillary restricted conditions and validation data for analysis of SSRIs reported in access media LC-MS/MS in MRM mode with positive ion recent years are summarized in Table 4 (Moraes et al. 1999 , ESI method for simultaneous analysis of fi ve antidepressant Sutherland et al. 2001 , Green et al. 2002 , Li et al. 2002 , Shen drugs (FL, imipramine, desipramine, amitriptyline, and nor- et al. 2002 , Zhu and Neirinck 2002 , Gutteck and Rentsch 2003 , triptyline) by using direct injection of biofl uids with a total Kollroser and Schober 2003 , Segura et al. 2003 , Souverain et run time of 8 min. al. 2003 , Weng and Eerkes 2003 , Pistos et al. 2004 , Reubsaet Juan et al. (2005) presented a HPLC-ESI/MS in the selected and Bjergaard 2004 , Singh et al. 2004 , Djordjevic et al. 2005 , ion recording (SIR) mode method for simultaneous determi- Hattori et al. 2005 , He et al. 2005 , Jain et al. 2005 , Juan et al. nation in human plasma of FL, CIT, PXT, and venlafaxine, 2005 , Massaroti et al. 2005 , Chen et al. 2006 , Fernandes et al. and intra- and inter-day variation coeffi cients were less than 2006 , Kirchherr and Kuhn-Velten 2006, Kovacevic et al. 2006 , 15.0 % . The LOD values were 0.5, 0.3, 0.3, and 0.1 ng/ml for Sauvage et al. 2006 , Shinozuka et al. 2006 , Smyth et al. 2006 , FL, CIT, PXT, and venlafaxine, respectively. Castaing et al. 2007 , Chu and Metcalfe 2007 , de Castro et al. He et al. (2005) developed a HPLC-ESI/MS assay for the 2007, 2008, Doherty et al. 2007 , Hattori et al. 2007 , Jawecki determination of SER in human plasma using zaleplon as the 2007 , Jia et al. 2007 , MacLeod et al. 2007 , Queiroz et al. 2007 , internal standard, SER is extracted from the alkalized plasma Frison et al. 2008 , Santos -Neto et al. 2008 , Franceschi et al. with cyclohexane. The organic layer is evaporated and the res- 2009 , Patel et al. 2009 , Saber 2009 ) . idue is re-dissolved in the mobile phase of methanol-10 mm ammonium acetate solution-acetonitrile (62:28:10). An ali- HPLC methods with other detection Greiner et al. quot of 20 µ l is chromatographically analyzed on a Shimadzu (2007) developed a method for the determination of citalo- ODS C18 column by means of SIM mode of MS. The calibra- pram and escitalopram together with their active main tion curve of SER in plasma exhibits a linear range from 0.5 metabolites desmethyl (es-)citalopram in human serum by to 25.0 µ g/l. The LOQ value was 0.5 ng/ml. column-switching HPLC and spectrophotometric detection To obtain high selectivity and sensitivity, MS/MS tech- (210 nm) with haloperidol as internal standard. Sample clean- niques employ specifi c SRM conditions, which are conve- up was carried on a LiChrospher CN 20 µ m pre-column using nient, especially in bioanalytical applications. Jia et al. (2007) 8% acetonitrile in deionized water. Then, separation was reported a LC-MS/MS method for the determination of SER performed on the analytical column (LiChrospher CN 5 µ m) hydrochloride in human plasma, using PRX as internal stan- using phosphate buffer (8 mmol/l, pH 6.4)-acetonitrile (50:50, dard. SER hydrochloride was chromatographed by using a v/v) as mobile phase at a fl ow rate of 1.5 ml/min. LOD for (es-)- Discovery C18 column. The mobile phase consisted of 0.1 % citalopram was 6 ng/ml with inter-day variation of < 9.05 % formic acid-acetonitrile (50:50). ESI source was applied and for citalopram and < 14.88 % for desmethylcitalopram. operated in the positive ion mode. SRM mode with the transi- Schatz and Saria (2000) reported a HPLC method with tions of m / z 306.0 – 274.9 and m / z 330.1 – 191.9 was used to coulometric detection for the simultaneous determination of quantify SER hydrochloride and the internal standard, respec- PXT, risperidone, and its main metabolite 9-hydroxyrisperi- tively. Detection limit was 0.334 ng/ml. done in human plasma involving a multistep SPLE. LLOD Smyth et al. (2006) reported the electrospray ionization was 1 ng/ml in the range of 5– 500 ng/ml. The drugs were and ion trap mass spectrometry (ESI-MSn ) and electrospray separated on a cyano column. 108 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

Thin-layer chromatography (TLC) method TLC was an Capillary electrophoresis method Andersen et al. (2003) earlier chromatographic method. It is simple, cheap, and presented a chiral CE system for simultaneous enantiomer allows the screening of a larger number of samples; however, it determination of CIT and its pharmacologically active meta- has lower sensitivity, precision, accuracy, and reproducibility bolite desmethylcitalopram with LPME based on a rod-like than other methods. porous polypropylene hollow fi ber. The LOQ value was LC and high-performance thin-layer chromatography < 11.2 ng/ml. (TLC) in the absorption mode methods for the simultaneous A CE method for separation of SER FLU and FL with estimation of FL and olanzapine in pure powder and tablet fully integrated SPE in situ within a fused silica capillary formulations was reported by Shah et al. (2007) . Linearity from either butyl methacrylate-co -ethylene dimethacrylate or ranged between 100 and 800 and 400 and 3200 ng/spot, 3-sulfopropyl methacrylate-co -butyl methacrylate-co -ethyl- respectively. LOD values for olanzapine and FL were 3.429 ene dimethacrylate was developed by Halvorsen et al. (2001) and 13.37 µ g/ml, and LOQ values were 10.392 and 40.53 developed a LPME and capillary electrophoresis of CIT and µ g/ml with LC, respectively. LOD values for olanzapine and its main metabolite N-desmethylcitalopramin human plasma FL were 33.13 and 132.08 ng/spot, and LOQ values were with extraction from 1 ml plasma samples through hexyl ether. 100.42 and 400.25 ng/spot with TLC, respectively. Prior to extraction, the samples were made strongly alkaline Gondova et al. (2008) presented a TLC method using den- and CIT and its main metabolite were enriched by a factor of sitometric detection in the refl ectance mode at 240 nm for 25– 30. LOQ values for CIT and its main metabolite in plasma the simultaneous analysis of CIT, SER, FL, and FLU. The were 16.5 ng/ml and 18 ng/ml, respectively. separation was achieved on silica gel 60 F 254s TLC plates A CE method with a laser-induced fl uorescence detection using acetone-benzene-ammonia (50:45:5, v/v/v) as mobile ( λ = 488 nm) for the analysis of SER together with its main phase. Linearity was in the range of 500– 5000 ng/spot for metabolite N-desmethylsertraline in human plasma was devel- all compounds. The LOD value for FL, FLU, and SER was oped by Musenga et al. (2007) . It is based on a SPE procedure 40 ng/spot for CIT and 50 ng/spot, respectively. Sharma et employed for biological sample pretreatment, followed by a al. (2007) developed a HPTLC densitometric method for the derivatization step with fl uorescein isothiocyanate isomer I determination of PXT hydrochloride in tablet dosage forms (FITC). In addition, similar CE assays for the separation of with LOD and LOQ for PXT hydrochloride of 60 ng/spot and SSRIs have been studied by other authors (Halvorsen et al. 160 ng/spot, respectively. Chromatographic separation was 2001 , Nevado and Salcedo 2002 , Andersen et al. 2003 , Flores carried out using precoated silica gel 60 F254 aluminum sheets et al. 2004a,b , Schaller et al. 2006 , Musenga et al. 2007 ). (10 ×10 cm, E. Merck, Cat. No. 1.05554.0007) with mobile phase consisting of ethyl acetate-acetic acid-water (7.5:1.5:1, Micellar electrokinetic chromatographic method v/v). Quantifi cation was carried out at 296 nm. Linearity for (MEKC) CE based techniques – employing MEKC, a buffer PXT hydrochloride ranged between 160 and 960 ng/spot with containing sodium dodecyl sulfate and an organic modifi er a correlation coeffi cient of 0.995. – have been developed for the successful separation of SSRIs (Lucangioli et al. 2000 , Labat et al. 2002 , Pucci et al. 2002 , Non-chromatographic methods Chen et al. 2004 , Flores et al. 2008 , Su and Hsieh 2008 ). Su and Hsieh (2008) presented a cation-selective exhaus- Electrophoretic methods CE is an attractive approach for the separation of SSRIs, because of its high effi ciency and rapid tive injection and sweeping MEKC method for the analysis of separation. The main advantages of capillary electrophoretic SER, FL, PXT, FLU, and CIT with LOD values ranging from techniques include high separation effi ciencies, minute amounts 0.056 to 0.22 ng/ml using an on-line preconcentration method. of sample are required; it is easily automated and consumes Chen et al. (2004) reported another MEKC system using limited amounts of reagents, generating low volumes of waste. a pH 11.5 borate buffer containing sodium deoxycholate and β In recent years, capillary zone electrophoresis (CZE) has gained hydroxypropyl- -CD with a detection wavelength of 210 nm popularity as an analytical tool for SSRIs. Although some was used. Optimum separation was achieved using a buffer successful examples have been reported, most CE applications (pH 11.5) of 35 mm sodium borate containing 30 mm sodium β in SSRI analyses require sensitivity to be enhanced by using deoxycholate and 20 mm hydroxypropyl- -cyclodextrin. more specifi c detection systems (e.g., MS) and on-line or Pucci et al. (2002) developed a MEKC method involving off-line sample preconcentration to increase sample solute a sodium dodecyl sulfate for the determination of CIT, FL, concentration. Table 5 contains a summary of the applications FLU, PXT, and SER with the most favorable MEKC system of electrophoretic techniques to the analyses of SSRIs. consisting of 20 mmol/l sodium dodecyl sulfate in a phosphate The different CE modes used for the separation of SSRIs are buffer (pH 7.5) with 30% methanol. Separation was achieved CZE and micellar electrokinetic chromatography (MEKC). on an uncoated fused silica capillary with spectrophotometric Table 5 outlines the optimal electrophoretic conditions for the detection at 200 nm. analyses of SSRIs in various biological samples. For ionic SSRIs, MEKC separations are based on both degree of ion- Electroanalytical methods The main problems encoun- ization and hydrophobicity. MEKC results are consistent with tered in using HPLC, GC methods are either the need for results obtained using HPLC or LC-MS for the determination derivatization or the need for time-consuming extraction of SSRIs. procedures. Because these techniques have slightly expensive Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 109 (Flores et al. 2004b ) (Nevado (Nevado and Salcedo 2002 ) (Schaller et al. 2006 ) al. 2008 ) (Su and Hsieh 2008 ) (Chen et al. 2004 ) (Pucci et al. 2002 ) (Andersen ) 2003 et al. ) 2001 et al. al. 2004a ) al. 2007 ) Pharmaceutical formulations Human urine (Flores et Bulk drug, tablets and capsules Human plasmaHuman (Halvorsen Human urine (Flores et Human plasmaHuman et (Musenga 3.7, 3.7, g/l g/l < µ g/l all for µ 95.6 % 95.6 > ve antidepressants ve 97.1 % , 97.1 precision was > /11.2 ng/ml. Intra-day/11.2 precision: < g/l µ 12.8 % RSD, and RSD, inter-day precision: % 12.8 all for RSD, enantiomers % 14.5 ranged mg/l from to 0.3 0.7 and % Recoveries obtained between 98 the of nominal content % 103 the compounds. LOD 12.5 values were and 25 ng/l LOD values obtained between and 3.0 7.1 The linearity FL, for SER, for 0.9962 PRX, for 0.9977 0.9964 and FLU, for the CIT; for LOD values 0.71, were 0.9914 and ng/l, 0.35 0.88, 1.5, 0.90, respectively Linearity ranges were 0.4 – 5.0 5.0 – Linearity ranges 0.4 were LOQ CIT values for and its main metabolite in ng/ml plasma were 16.5 and 18 ng/ml, The LOD respectively. values were 5 ng/ml and ng/ml, 5.5 respectively LOQ: < < for PXTfor and its metabolites accuracy (recovery) was LOD values and between 23.1 9.3 Linearity: 3.0 500 – ng/ml Extraction:

m m µ

m × m i.d.; 75 m µ × 50 borate buf- borate × m sodium dodecyl -cyclodextrin, -cyclodextrin, g/ml each analyte m sodium borate borate sodium β µ m -cyclodextrin β nal consisted BGE of -methyl)- sodium and deoxycholate O m -cyclodextrin as chiral selector m o.d.) β ammonium acetate and 1.1 % % ammonium acetate and 1.1 µ m 375 375 75 m capillary75 using a non-aqueous × phosphate buffer adjusted to pH 2.5 LOD the values for fi hydroxypropyl- carbonate with 2.5 m pH 9.0, buffer, Phosphate buffer in pH 3.5 of ACN GLC and 20 % v/v acetone % GLC and 20 ammonium acetate and 1 % acetic acid % ammonium acetate and 1 × m m m m m m m i.d. 50 m 40 m 40 (20:80, with v/v) UV detection 200 of nm and sample concentration 20 of Fused silica capillary tube (60 cm acetic acid in 80:20 methanol-ACN (v/v) Polymicro Phoenix, Technologies, AZ, USA) m 35 of A buffer (pH 11.5) m20 20 mmol/l sodium dodecyl sulfate in a phos- methanol on % withphate 30 buffer (pH 7.5) an uncoated fused silica Mobile phase consisting m 15 of containing (pH 9.2) fer 20 m 2-propanol through (v/v) a % sulfate and 12 fused silica capillary cm total (57 length 1 % 1 Sulfated- in 25 ACN m % in combination with 12 containing m 30 µ Tris-acetic acid pH 4.6, 3 % (w/v) Tween-20 Tween-20 (w/v) % Tris-acetic 3 acid pH 4.6, as the mg/l separationand 75 FC-135 buffer phosphate 2.5 pH A 30-cm length effective capillary was utilized (40.2 cm total length) m with 75 procedure,A SPE a derivatization by followed step with FITC. The fi m20 heptakis (2,6-di- preconditioned C18 cartridge,preconditioned cm C18 a 57 50 m buffer methanol-ACN 9:1 system of containing m25 capillary using a non-aqueous buffer consist- m ing 18 of ber capillary 64.5 cm total in situ in × m i.d. 488 nm) µ = λ uorescencedetection length) Capillary zone electrophoresis method grated SPE (50 exhaus- Cation-selective injectiontive and sweeping method MEKC MEKC system with a detec- nm 210 of tion wavelength MEKC method capillary spectrophotometric with detection at 200 nm MEKC method with a DAD (240 nm) after an extraction- preconcentration step with a cartridgepreconditioned C18 (Waters, Milford, MA, USA) Chiral CE system LPME based on a rod-like porous fi hollow polypropylene CE-LPME with extraction from 1 ml plasma samples through ether hexyl CE with a laser-induced fl ( Non-aqueous CE method Extraction-preconcentration step with a Non-aqueous CE method cm 57 Isomers and Electrophoretic methods. -desmethylsertraline N -desmethylcitalopramin FL, trazodone, CIT, FLU, and nefazodone SER, and FLU, FL CE method with fully inte-SER, FL, and FLU, PXT, CIT cis-trans enantiomers SER of CIT, FL, andCIT, PXT, FLU, SER Breast cancer (letrozole), (CIT) antidepressant an Table 5 Table AnalyteCIT and desmethylcitalopram Methods conditions Experimental Results Applications References CIT and N SER and its main metabolite PXT, tamoxifen,PXT, and their metabolites main PXT and three metabolites 110 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

instrumentation and running costs, the use of simpler, faster, and less expensive, but still sensitive, electrochemical techniques can be an interesting alternative. Stripping voltammetry

(Labat et al. 2002 ) ) 2000 et al. comprises a variety of electrochemical approaches, having a step of preconcentration onto the electrode surface prior to the voltammetric measurement. In adsorptive stripping voltammetry, the analyte is adsorbed on the working electrode by means of a non-electrolytic process prior to the voltammetric scan. The high sensitivity of adsorptive stripping voltammetric uids (blood, uids Biological ﬂ urine) drugBulk (Lucangioli methods makes it possible to work with very diluted samples with a corresponding decrease in possible interferences in the analysis. Adsorptive-stripping voltammetry (AdSV) is a technique mainly used for the analysis of organic compounds, which can be accumulated at, for example, the hanging mercury drop electrode surface and afterwards stripped-off by applying a potential scan. The introduction of high scan rate voltammetric techniques, for example, square-wave 0.995)

= voltammetry (SWV), increases the sensitivity of AdSV even further (Wang 1985, Barros et al. 1999 ). Nouws et al. (2008) presented two methods regarding the electrooxidative behavior of CIT with cyclic voltammetry and SWV at a glassy carbon electrode with a LOD value of 9.5 × 10 -6 mol/l in a phosphate buffer of pH 8.2 and a ﬂ ow- LOD and LOQ mg/ml 0.2 were and 0.7 mg/ml SER, for respectively LOD values and LOQ values ranged between and 20 and 10 between 20 and 30 ng/ml, all for the respectively, molecules Calibration curves established were for 30 2000 – ng/ml (r injection analysis system using amperometric detection with × -6

LOD of 1.9 10 mol/l. In addition, Nouws et al. (2007) m described an electroanalytical method based on square-

sodium wave adsorptive-stripping voltammetry and ﬂ ow-injection m analysis with square-wave adsorptive-stripping voltammetric detection for the determination of FL with the reduction of FL at a mercury drop electrode in a phosphate buffer of

-cyclodextrin and pH 12.0. Furthermore, Nouws et al. (2006a) developed elec- β -cyclodextrin

β troanalytical methods (square-wave adsorptive-stripping voltammetry) and ﬂ ow-injection analysis with square-wave

sulfated adsorptive-stripping voltammetry detection for the determi- m nation of PXT in a borate buffer of pH 8.8 at a mercury drop electrode with LOD and LOQ of 4.8× 10-7 and 1.6× 10 -6 mol/l, sodium borate with pH 8.55 20 of m hydroxypropyl- m respectively. Nouws et al. (2006b) also developed an elec- m sodium with borate 50 m (pH 9.0

m troanalytical method for the determination of CIT in phar- Uncoated fused silica capillary (600 mm, 75 mm with i.d.) migration buffer consisting of m20 Background electrolyte buffer consisting 20 of m mcholate) 15 5 m sodium dodecyl sulfate and 15 % isopropanol % sodium dodecyl sulfate and 15 at an operating 25 kV of voltage maceutical preparations with square-wave and square-wave adsorptive-stripping voltammetry using a mercury drop electrode at a potential of approximately -1.25 V vs. AgCl/ Ag in an aqueous electrolyte solution of pH 12. Linearity ed ed for the proposed method ranged between 1.0× 10 -7 and 2.0× 10 -6 mol/l with a LOD value of 5× 10-8 mol/l. Nouws et al. (2005) developed a ﬂ ow-injection square-wave cathodic- stripping voltammetric method for the determination of SER in a pharmaceutical preparation with LOD and LOQ values × -7 × -7

MEKC methods with DAD including extraction with diethyl ether ml) (5 at pH 9.6 Cyclodextrin-modiﬁ MEKC of 1.5 10 and 5.0 10 mol/l, respectively. Linearity was obtained in the range of 0.20× 10 -6 and 1.20 × 10-6 mol/l. The proposed method was applied to the determination of SER in a commercial product. Medyantseva et al. (2008) reported an amperometric biosensor based on a platinum screen-printed electrode and immobilized monoamine oxidase for the determination of antidepressants including petylyl, pyrazidol, and FL with determination limits of 8× 10 -9 , 8 × 10-7 , and 8 × 10-10 m , SER hydrochloride and synthesis-related substances (Table 5 (Table continued) AnalyteCIT, FLU, PXT, SER, FL, and other drugs Methods conditions Experimental respectively. Results Applications References Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 111

Lencastre et al. (2006) presented an assay with regard to the Orange II quantitatively extracted into dichloromethane sol- oxidative behavior of FL at a glassy carbon electrode in various vent was measured at 482 nm with linearity over concentration buffer systems and at different pH values using cyclic, differ- range of 0.2– 9.0 µ g/ml and a regression coeffi cient of 0.9995. ential pulse and SWV. SWV used a borate pH 9 buffer solution Darwish (2005) presented three spectrophotometric meth- as supporting electrolyte with a LOD value of 1.0 µ m . ods based on the reaction of the N -alkylvinylamine formed Fast Fourier transform continuous cyclic voltammetric from the interaction of the free secondary amino group in the techniques for the monitoring of ultra trace amounts of CIT in investigated drugs and acetaldehyde with each of three halo- a fl ow-injection system was reported by Norouzi et al. (2007) quinones to give colored vinyl amino-substituted quinones for with LOD and LOQ values of 2.3 and 7 pg/ml, respectively. determination of the hydrochloride salts of FL, SER, and PXT Nevado et al. (2000) described electroanalytical methods for in their pharmaceutical dosage forms. Absorption maxima for the determination of FLU in aqueous samples (pH 2.0 and 4.7) colored products obtained with chloranil, bromanil, and 2,3- and pharmaceutical formulations with square-wave techniques dichloronaphthoquinone appeared at 665, 655, and 580 nm, (the hanging mercury drop electrode at -0.76 V, using an accu- respectively. The LOD values for the assays ranged from 1.19 mulation potential of -0.50 V) with the determination of FLU to 2.98 mg/ml. between 2× 10-8 and 3 × 10-6 mol/l. Assay cyclic voltammetry, dif- Two methods with derivative spectrophotometry and ferential pulse voltammetry and Osteryoung SWV for the deter- HPLC with UV detector (265.0 nm for SER hydrochloride) mination of PXT hydrochloride were developed by Erk and for the determination of nefazodone hydrochloride and SER Biryol (2003) in the concentration range of 2× 10-5 to 8 × 10 -4 m in hydrochloride in pharmaceutical formulations were presented pure form and in human plasma. Erk and Biryol also developed by Erk (2003) . In the fi rst derivative spectrophotometry, sig- a comparative HPLC assay using DAD with linearity between nals for nefazodone hydrochloride and SER hydrochloride 2× 10-7 and 6× 10-5 m for PXT hydrochloride. were at 241.8 – /256.7 nm and 271.6 – /275.5 nm, respectively. Roque da Silva et al. (1999) investigated the electrochemi- In the HPLC-UV method, separation was performed on a cal reduction of FL with cyclic, linear sweep, differential Supercosil RP-18 column using mobile phases consisting of pulse and SWV by using a hanging mercury drop electrode in methanol:ACN:phosphate buffer at pH 5.5 (10:50:40, v/v/v) alkaline buffer solution in water and in a water:ACN mixed (nefazodone hydrochloride) and methanol:phosphate buffer solvent. at pH 4.5 (20:80, v/v) (SER hydrochloride). The LOD values Vela et al. (2001) presented an assay with electrochemi- for nefazodone and SER were 0.58 and 0.31 µ g/ml with the cal behavior of SER at a hanging mercury drop electrode spectrophotometric method. The LOD values for nefazodone with a LOD value of 1.98 × 10 -7 using different voltammet- and SER were 15.8 and 24.0 µ g/ml with the HPLC method. ric techniques such as cyclic, linear sweep, differential pulse The LOQ values for nefazodone and SER were 1.88 and and SWV. Britton-Robinson buffer solution was used as sup- 0.96 µ g/ml with the spectrophotometric method. The LOQ porting electrolyte at different pH values. Best results were values for nefazodone and SER were 26.4 and 45.3 µ g/ml found by SWV with electrodeposition at alkaline pH using a with the HPLC method. borate buffer of pH 8.2 containing 12 % (v/v) methanol. Table Methods for determining SSRI drugs have also been based 6 contains a summary of the applications of electroanalytical on nuclear magnetic resonance (NMR) detection which needs techniques to the analyses of SSRIs. no previous treatment steps nor derivatization. Trefi et al. (2008) developed a simple and selective 19 F NMR method for Spectrometric methods Spectrophotometric methods have the quantitation of FL and FLU in methanol solutions and in several advantages such as low interference level, good human plasma and urine with a good linearity in the range of analytical selectivity, easy-to-use and less expensive, and less 1.4 – 620 µg/ml with a LOD value of approximately 0.5 µ g/ml time-consuming compared with most of the other methods. and a LOQ value of approximately 2 µ g/ml (4.6× 10 -6 mol/l). Spectrophotometric methods are simple and rapid and thus Determining the enantiomeric purity of chiral therapeutic these methods can be successfully used for pharmaceutical agents is important in the development of active pharma- analysis, involving quality control of commercialized product ceutical ingredients. Shamsipur et al. (2007) described a 19 F and pharmacodynamic studies. These methods are mostly NMR spectroscopy method for the quantitative determination based on the formation of colored complexes between SSRI of FL enantiomers using different chiral recognition agents in drug and the reagent which can be determined by visible pharmaceutical formulations with LOD values of 5.9 and 7.5 spectrophotometry. UV spectrophotometric and derivative µg/ml for the pure solutions of ( R)- and ( S )-FL, respectively. spectrophotometric methods have also been used for SSRIs. Also, a NMR and chiral solvating agent (CSA, 1,1-bi-2- A spectrophotometric procedure for the determination of naphthyl) technique for the routine determination of enantio- SER and/or clidinium bromide in bulk sample and in dosage meric purity (SER, PRX, and fenfl uramine) was reported by forms were developed by Amin et al. (2009) . The procedure Salsbury and Isbester (2005) . Similar spectroscopic assays for is based on the formation of an ion pair complex by their reac- the separation of SSRIs have been studied by other authors tion with bromocresol green, bromophenol blue, and bromo- (Erk 2003 , Darwish 2005 , Salsbury and Isbester 2005 , Alarfaj thymol blue in buffered aqueous solution at pH 3. and Razeq 2006 , Onal et al. 2006 , Raza 2006 , Serebruany et Parham et al. (2008) developed an extraction-spectropho- al. 2007 , Shamsipur et al. 2007 , Darwish et al. 2008 , Parham tometric method for the determination of FL in pharmaceuti- et al. 2008 , Trefi et al. 2008 , Amin et al. 2009 , Darwish et al. cals with a LOD value of 0.17 µ g/ml. FL and an ion pair with 2009 ) (Table 7). 112 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs (Nevado et al. 2000) (Nouws et al.(Nouws 2005) (Nouws et al.(Nouws 2007) (Nouws et al.(Nouws 2006b) (Medyantseva et al. 2008) et (Lencastre al. 2006) 2008) 2006a) 2008) (Norouzi et al. al. et (Norouzi 2007) Aqueous samples (pH 2.0 and pharmaceu-and 4.7) tical formulations Pharmaceutical preparations Pharmaceutical products, human serum samples, and in drug dissolution studies Pharmaceutical preparations Pharmaceutical products et al. (Nouws Pharmaceutical products et al. (Nouws Pharmaceutical products et al. (Nouws

-7 10) = 10 × and and -9 -6 10 10 × × mol/l mol/l, -6 -7 mol/l) and the -7 10 10 × × 10 × and 1.6 and 5.0 -7 -7 , and 9.0 -7 10 10 × × 10 × mol/l, with 3.54% square-wave (n mol/l linear in the range 5 -6 -6 mol/l with stripping square-wave 2.39% , 6.0 mol/l mol/l mol/l mol/l -7 mol/l mol/l mol/l 10 -7 10 -7 -6 -8 -8 -6 -6 -6 × × mol/l 10 -10 10 m -6 10 10 10 10 × 10 10 10 × µ × × × × × × mol/l 10 × 10 -6 × × and 3 mol/l stripping by mode. The standard relative 10 -8 -6 × 10) in the10) same day 10 10 = × × and 6.0 for Pharmaceutical products Recovery values between 98% and 102% Serum and samples recovery values between 81.1% 90.4% LOD and LOQ 4.8 of 1.20 deviations obtained concentration for FVX of levels were as as 1.5 low LOD: 1.0 LOD: 1.0 LOQ: 6.3 LOD 5 of Recovery 94–106% were assays LOD: 6.5 LOD: LOQ: 32 LOQ: 2.2 results obtained 98–102% Linearity proposed for method ranged between 1.0 Recovery 96–99% were assays respectively Linearity was obtained in the range 0.20 of and 2.0 1 Results Applications References (n Recovery performed were assays at three concentration (3.0 levels LOD and LOQ 1.5 of LOD and LOQ: 2.3 and 7 pg/ml The method was linear the concentration over range of pg/ml 7–116 LOD: 1.9 LOD: LOD: 9.5 LOD: ow-injectionsystem Amperometric biosensor LOD: 8 Square-wave adsorptive-Square-wave stripping voltammetry Cyclic, differentialCyclic, pulse and square-wave voltammetry Square-wave adsorptive-Square-wave stripping voltammetry Square-wave andSquare-wave square- wave adsorptive-stripping voltammetry separation wave cathodicwave stripping voltammetric method form continuous cyclic technique voltammetric ﬂ Square-wave techniquesSquare-wave 2 system using amperometric detection square-wave voltammetry at a glassy-carbon electrode Platinum screen-printed electrode and immobilized monoamine oxidase drop electrode drop buffer systems (borate pH 9 buffer electrolyte) supporting as solution electrode in a phosphate buffer pH 12.0 of tial of -1.25 V vs. AgCl/Ag, V vs. intial an -1.25 of aqueous electrolyte pH 12 of solution -0.76 V, using an accumulation V, poten- -0.76 tial -0.50 of V) Electroanalytical methods. Electroanalytical Table 6 Table Analyte Methods conditions Experimental Petylyl, pyrazidol, FL and PXT Borate buffer pH 8.8 at a mercury of SERFLCIT Glassy carbon electrode in various square Flow-injection Fourier Fast trans- FL Reduction FL of at a mercury drop CIT mercury A drop electrode at a poten- CIT analysis Flow-injection CIT Phosphate buffer pH 8.2 of voltammetry Cyclic and FLU Hanging mercury drop electrode at Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 113

Flow injection is a low-pressure, low-cost continuous technique, the versatility and availability of which enable the desired level of automation to be implemented in enological laboratories. (Erk and Biryol 2003) da (Roque et al. Silva 1999) (Vela et al. (Vela 2001) Alt ı okka and K ı rcal ı (2003) developed a ﬂ ow-injection analysis of PXT hydrochloride with 0.1 mol/dm3 acetate buffer at pH 3.07 and LOD and LOQ values of 3.2× 10 -7 and 9.5 × 10 -7 mol/dm, respectively. The analyte was detected at 293 nm and linearity ranged between 1.07× 10-6 and 5.35 × 10 -6 mol/dm 3 . Shah et al. (2008) developed a ﬂ ow-injection spectrophotometric method for the determination of FL in a pharmaceutical preparation based on base hydrolysis with sodium ethoxide of In pure form and in plasma human Pharmaceutical formulations the drug with LOD and LOQ values of 0.15± 0.01 mg/l and 0.29 ± 0.03 mg/l, respectively. with with m

-6 Conclusions 10 × m

-4

10 A review of the literature revealed that a variety of analyti- m × cal procedures for the determination of SSRIs have been and 3.15 -7 to 8 developed and described. The analytical methods used for -5 10 × 10 the determination of SSRIs are generally based on chromato- × graphic methods coupled to different detectors, electroanalytical methods, capillary zone electrophoretic methods, and spectrometric methods. Biological ﬂ uid sample preparation

-7 is usually performed by LLE or SPE. These methods usually

10 consume organic solvents, are laborious and time-consum- × ing. However, modern trends in analytical chemistry are near simplifi cation, in miniaturization of sample preparation, and minimization of organic solvent used, and sample volumes. Linearity ranged between 2.33 Linearity ranged between and 0.52 5.2 LOD 1.98 of recoveries to 100% close Results Applications References The concentration range 2 of SPME, stir bar SBSE, protein precipitation, direct injection of biological samples without sample preparation SLM, LSE, LPME, PLE have also been employed for the SSRI determination. On-line extraction and analysis of SSRIs and its metabolite by column-switching HPLC coupled to mass spectrometry has recently been published. A promising approach to HPLC with integrated fully automated sample extraction is high speed on-line SPE which enables direct injection of plasma samples without prior extraction by using large par- Cyclic, linearCyclic, sweep, differential pulse and square-wave voltammetry linearCyclic, sweep, differential pulse and square-wave voltammetry separation differential pulse voltammetry and Osteryoung square-wave voltammetry ticle size stationary phases with an extremely high linear fl ow velocity of the mobile phase. In HPLC, analyses have commonly used fl uorescence, ultraviolet (UV) and mass spectrometry (MS) detectors. The use of MS with HPLC provides high sensitivity and selectivity compared to traditional HPLC-UV and GC methods. In most HPLC-MS systems, ESI is employed as an interface between a HPLC instrument and an MS detector. APCI is also a useful technique coupling HPLC with MS. The column- switching technique has been found in the literature for SSRI analysis involving more expensive and complex instrumentation. Typical quantifi cation limits are in ng/ml, µ g/ml, and pg/ml range. alkaline buffer in water solution and mixedin solvent a water-ACN Britton-Robinson buffer was solution used as supporting electrolyte GC with nitrogen-phosphorus, electron-capture, mass spectrometric, fl ame ionization detection methods have been used to a lesser extent in SSRIs and analysis of their metabolites. GC-MS is a very good separation method for compounds that (Table 6 (Table continued) Analyte Methods conditions Experimental FLSER A hanging mercury drop electrode in Hanging mercury drop electrode PXT voltammetry, Cyclic are volatile. In recent years, the electrochemical techniques 114 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs et al. (Erk (Erk ) 2003 (Alarfaj and Razeq 2006 ) (Darwish (Darwish 2005 ) (Amin et al. 2009 ) al. 2008 ) al. 2009 ) (Shamsipur (Shamsipur ) 2007 et al. (Salsbury and Isbester 2005 ) (Onal et al. 2006 ) (Trefi 2008 ) Pharmaceutical formulations Pharmaceutical formulations Pharmaceutical formsdosage Tablet Tablet formulations (Raza 2006 ) Bulk sample and in formsdosage Pharmaceutical et (Parham Pharmaceutical formulations Pharmaceutical ingredients Pharmaceutical preparations In human plasma plasma human In and urine

-6 10 × g/ml µ g/ml and the g/ml with molar µ g/ml g/ml (4.6 µ µ g/ml (Darwish et g/ml. Linearity over g/ml. more For accu- 8 % 8 % µ µ µ < g/ml the for pure solu- µ )-FL, The linearity respectively. S 7.5 % . 103 103 l/mol/cm < 0.9995 cient of g/ml was used × )-FL: 0.10 – 1.35 mg/ml1.35 with recovery – )-FL: 0.10 µ S g/ml g/ml and LOQ: 2 µ )- and ( µ R )- and ( R LOD: 0.31 LOD: 0.31 LOQ: mg/ml 0.96 LOD of 0.015 mg/ml.LOD 0.015 of plasma Human concentration range of 0.2 – 9.0 9.0 – concentration range 0.2 of regression coefﬁ regression LOD to 2.98 values the for ranged assays from 1.19 mg/ml absorptivity 3.3 rate analysis, Ringbom optimum concentration 27 – range 2 of LOD and LOQ: and 0.2 0.6 Concentration range: 1 – 30 – Concentration range: 1 tions ( of for ( with rela- % to 110 % ranged from approximately 90 standardtive deviations of LOD: 0.5 96 % to 103 % , 93 % to 104 % in plasma, with stan- % to 104 mol/l).% The accuracy ranged from approximately, 93 % to 103 % 96 dard deviations F NMR method F NMR method 620 – Linearity in the range 1.4 of and LOD values: 7.5 5.9

and HPLC with UV detector 241.8 – /256.7 nm/256.7 and – and HPLC with UV detector 241.8 –271.6 /275.5 nm nm using 290 nm excitation for Extraction-spectrophotometric method nm) (482 LOD the of method 0.17 of Acetaldehyde with each three of haloquinones to colored vinylamino-substitutedgive quinones 590 nm590 250 – Concentration limit 10 of NMR method NMR Alkaline medium pH 9 with of an orange-colored product exhibiting maximum absorption peak at 470 nm

Spectrophotometric procedure based on formation anof ion-pair their by complex reaction with bromocresol green (BCG), (BPB), bromophenol blue and bromothymol (BTB) blue in buffered aqueous at pH 3 solution 19 19 -alkylvinylamine formed Orange II extracted into dichloromethane solvent N from the interaction of the free secondary amino group 1,4-benzoquinone Chiral solvating agent 1,1-bi-2-naphthyl) (CSA, Ion-pair agents (bromo- bromocresolthymol blue, green, or bromophenol blue) FLU sulfonate reagent Different chiral recogni- chiral Different agents tion Bromocresol green, andbromophenol blue, blue bromothymol u- Spectrometric methods. FL and FLU SER methods with spectrophotometry derivative Two PXT Fluorescence intensity obtained in methanol at 340 FL FL and an ion-pair with SER and PXT CIT 2,3-Dichloro-5,6-dicyano- ramine SER, PRX, fenﬂ and SER, FL, and venlafaxine FLU 1,2-naphthoquinone-4- FL FL enantiomers Table 7 Table AnalyteSER and Reagentclidinium bromide Experimental conditions Results Applications References Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 115

have led to the advancement in the ﬁ eld of analysis because of their sensitivity, low cost, and relatively short analysis time when compared to other techniques such as chromatographic (Darwish et et (Darwish al. 2008 ) ) 2007 et al. (Shah et al. 2008 ) (Alt ı okka and K ı rcal ı ) 2003 techniques. Although SSRIs are being electrochemically active, there are limited assays with regard to determination of SSRIs with electrochemical techniques. However, owing to the importance of these compounds it is very likely to be widely studied and developed in the future with electrochemical techniques as well as HPLC with electrochemical detection methods for determination of SSRI drugs. Dosage forms and plasma plasmaHuman (Serebruany Pharmaceutical preparation In general, spectrophotometric methods have several advantages such as low interference level, good analytical selectiv-

, ity, easy-to-use and less expensive and less time-consuming 3 -6

10 compared with most of the other methods. However, the 0.03 ×

± spectrophotometric methods are less sensitive due to mea-

mol/dm surement in the ultraviolet region compared with most of the -7

10 other methods. and 5.35 × -6 Very few methods are reported in the literature for SSRIs 10 × with CE due to the low sensitivity of this technique which and 9.5

-7 limits its applicability to drug analysis in biological samples. 0.01 mg/l0.01 and 0.29

4 10 ± MEKC is one of the most popular techniques in CE and is × 10 uorescence intensity and PXT × capable of separating neutral compounds as well as charged

cient (0.9993) and LOD and LOQ solutes. However, the combination of CE and MS-MS and MEKC-MS methods such as partial ﬁ lling, high-molecular

mass surfactan, adsorption, sorption, ESI, APCI could gain 3 considerable importance in SSRI drugs analysis. In summary, different pretreatment and detection pro- LOD and LOQ 3.2 of Linearity: between fl 800 ng/ml with –concentration in the range 80 of coefficorrelation 25of and ng/ml, 77 respectively linearity ranged between 1.07 mol/dm mg/l, respectively Absorbance nm measured was with a at 510 linearity between 0.5 and 25 mg/l and a molar absorptivity 2.19 of cedures are usually applied for SSRI drugs analysis. Chromatographic methods potentially meet these require- ments but at present are focused on SSRI drugs analysis and often have insuffi ciently low detection limits. Therefore, the development of more sensitive methods that are still simple to perform and methods is desirable. uorescent

References

Addison, R. S.; Franklin, M. E.; Hooper, W. D. Sensitive, high- throughput gas chromatographic – mass spectrometric assay for fl uoxetine and norfl uoxetine in human plasma and its application to pharmacokinetic studies. J. Chromatogr. B 1998 , 716 , 153 – 160. uorimeter Alarfaj, N.; Razeq, S. A. Spectrofl uorimetric determination of paroxetine hydrochloride in its formulations and human plasma. Fluorimetry methods– measured in the range 270 nm450 with at 240 excitation nm on a FluoroMax 3 spectrofl buffer detected at pH 3.07 nm at 293 derivative 545 nm545 after at 490 excitation nm in an alkaline medium pH 8 to form of a highly fl Flow-injection spectrophotometricFlow-injection method LOD and LOQ 0.15 of Chem. Pharm. Bull. 2006 , 54 , 564 – 566. Alt ı okka, G.; K ı rcal ı , K. Simple method of paroxetine determination using a single channel FIA with no in-line reaction process. Anal. Sci. 2003 , 19 , 629 – 631. Alvarez, J. C.; Bothua, D.; Colignon, I.; Advenier, C.; Spreux- Varoquaux, O. Determination of fl uoxetine and its metabolite norfl uoxetine in serum and brain areas using high-performance liquid chromatography with ultraviolet detection. J. Chromatogr. B Biomed. Sci. Appl. 1998 , 707 , 175 – 180. Amin, A. S.; Dessouki, H. A.; Moustafa, M. M.; Ghoname, M. S. oxa-1,3-diazole with sodium ethoxide of of ethoxide sodium with drug the Spectrophotometric methods for sertraline hydrochloride and/ or clidinium bromide determination in bulk and pharmaceutical preparations. Chem. Papers 2009 , 63 , 716 – 722. Andersen, S.; Halvorsen, T. G.; Pedersen-Bjergaard, S.; Rasmussen, K. E.; Tanum, L.; Refsum, H. Stereospecifi c determination of CIT and escitalopram PXT mol/dm analysis Flow-injection with 0.1 acetate AnalytePXT Reagent 4-Chloro-7-nitrobenzo-2- Experimental conditions Results Applications References FL Based on base hydrolysis (Table 7 (Table continued) citalopram and desmethylcitalopram by capillary electrophoresis 116 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

and liquid-phase microextraction. J. Pharmaceut. Biomed. 2003 , Cruz-Vera, M.; Lucena, R.; Cá rdenas, S.; Valcá rcel, M. Combined 33 , 263 – 273. use of carbon nanotubes and ionic liquid to improve the deter- Asberg, M.; Thorne, P.; Transkman, L. Serotonin depression: a bio- mination of antidepressants in urine samples by liquid chroma- chemical subgroup within the affective disorders? Science 1976 , tography. Anal. Bioanal. Chem . 2008 , 391 , 1139 – 1145. 191 , 478 – 480. Darwish, I. A. Development and validation of spectrophotometric Bagli, M.; Rao, M. L.; Sobanski, T.; Laux, G. Determination of methods for determination of fl uoxetine, sertraline, and parox- fl uvoxamine and paroxetine in human serum with high-perfor- etine in pharmaceutical dosage forms. J. AOAC Int. 2005 , 88 , mance liquid chromatography and ultraviolet detection. J. Liq. 38 – 45. Chromatogr. R.T. 1997 , 20 , 283 – 295. Darwish, I. A.; Amer, S. M.; Abdine, H. H.; Al-Rayes, L. I. New Bahrami, A.; Mohammadi, B. Rapid and sensitive bioanalytical spectrofl uorimetric method with enhanced sensitivity for deter- method for measurement of fl uvoxamine in human serum using mination of paroxetine in dosage forms and plasma. Anal. Chem. 4-chloro-7-nitrobenzofurazan as pre-column derivatization agent: Insigh. 2008 , 3 , 145 – 155. application to a human pharmacokinetic study. J. Chromatogr. B Darwish, I. A.; Abdine, H. H.; Amer, S. M.; Al-Rayes, L. I. 2007 , 857 , 322 – 326. Spectrophotometric study for the reaction between fl uvoxam- Baltussen, E.; Sandra, P.; David, F.; Cramers, C. Stir bar sorptive ine and 1,2-naphthoquinone-4-sulphonate: kinetic, mechanism extraction (SBSE), a novel extraction technique for aqueous and use for determination of fl uvoxamine in its dosage forms. samples: theory and principles (Citations: 55). J. Microcolumn Spectroc. Acta Part A 2009 , 72 , 897 – 902. Sep. 1999 , 11 , 737 – 747. David, F.; Tienpont, B.; Sandra, P. Stir-bar sorptive extraction of Barlow, D. H.; Durand, V. M. Mood disorders and suicide. In: trace organic compounds from aqueous matrices. LCGC Europe Abnormal Psychology: An Integrative Approach. 5th edition. 2003, 17, 2–7. Wadsworth Cengage Learning: Belmont, CA, 2009, 239. ISBN de Castro, A.; Fernandez, M. D. M. R.; Laloup, M.; Samyn, N.; Boeck, 0495095567. OCLC 192055408. G. D.; Wood, M.; Maesd, V.; Rivadulla, M. L. High-throughput Barri, T.; Jö nsson, J. A. Supported liquid membrane work-up of on-line solid-phase extraction-liquid chromatography-tandem blood plasma samples coupled on-line to liquid chromatographic mass spectrometry method for the simultaneous analysis of 14 determination of basic antidepressant drugs. Chromatographia antidepressants and their metabolites in plasma. J. Chromatogr. 2004 , 59 , 161 – 165. A 2007 , 1160 , 3 – 12. Barros, A. A.; Rodrigues, J. A.; Almeida, P. J.; Rodrigues, P. G.; Fogg, de Castro, A.; Concheiro, M.; Quintela, O.; Cruz, A.; Rivadulla, M. L. A. G. Voltammetry of compounds confi ned at the hanging mercury LC – MS/MS method for the determination of nine antidepres- drop electrode surface. Anal. Chim. Acta 1999, 385 , 315 – 323. sants and some of their main metabolites in oral fl uid and plasma: Berzas, J. J.; Guiberteau, C.; Contento, A. M.; Rodrfguez, V. Sensitive study of correlation between venlafaxine concentrations in both and rapid high-performance liquid chromatographic method for matrices. J. Pharmaceut. Biomed. 2008 , 48 , 183 – 193. simultaneous determination of antidepressants in pharma ceutical Djordjevic, S.; Kovacevic, I.; Miljkovic, B.; Vuksanovic, J.; Pokrajac, formulations. Chromatographia 2002, 56 , 545 – 551. M. Liquid chromatographic-mass spectrometric method for the Bones, J.; Thomas, K.; Nesterenko, P. N.; Paull, B. On-line precon- determination of fl uoxetine and norfl uoxetine in human plasma: centration of pharmaceutical residues from large volume water application to clinical study. II Farmaco 2005 , 60 , 345 – 349. samples using short reversed-phase monolithic cartridges cou- Doherty, B.; O ’ Donnell, F.; Smyth, W. F.; Leslie, J. C.; Ramachandran, pled to LC-UV-ESI-MS. Talanta 2006 , 70 , 1117. V. N.; Boyd, N. S.; Hack, C. J.; O ’ Kane, E.; McClean, S. A study Castaing, N.; Titier, K.; Daure, M. R.; Deodic, M. L.; Le-bars, D.; of the analytical behaviour of selected psycho-active drugs using Moore, N.; Molimard, M. Quantifi cation of eight new antidepres- liquid chromatography, ion trap mass spectrometry, gas chroma- sants and fi ve of their active metabolites in whole blood by high- tography and polarography and the construction of an appropriate performance liquid chromatography-tandem mass spectrometry. database for drug characterisation. Talanta 2007 , 72 , 755 – 761. J. Anal. Toxicol. 2007 , 31 , 334 – 341. Duverneuil, C.; de la Grandmaison, G. L.; de Mazancourt, P.; Alvarez, Chaves, A. R.; Silva, S. M.; Queiroz, R. H. C.; Lan ç as, F. M.; Queiroz, J. C. A high-performance liquid chromatography method with M. E. C. Stir bar sorptive extraction and liquid chromatography photodiode-array UV detection for therapeutic drug monitoring with UV detection for determination of antidepressants in plasma of the nontricyclic antidepressant drugs. Ther. Drug Monit. 2003 , samples. J. Chromatogr. B 2007 , 850 , 295 – 302. 25 , 565 – 573. Chaves, A. R.; J ú nior, G. C.; Queiroz, M. E. C. Solid-phase microex- Eap, C. B.; Gaillard, N.; Powell, K.; Baumann, P. Simultaneous deter- traction using poly(pyrrole) fi lm and liquid chromatography with mination of plasma levels of fl uvoxamine and of the enantiomers UV detection for analysis of antidepressants in plasma samples. of fl uoxetine and norfl uoxetine by gas chromatography-mass J. Chromatogr. B 2009 , 877 , 587 – 593. spectrometry. J. Chromatogr. B 1996 , 682 , 265 – 272. Chen, D.; Chen, Y.; Hu, Y. Optimized separation of cis-trans isomers El-Gindy, A.; Emara, S.; Mesbah, M. K.; Hadad, G. M. Liquid and enantiomers of sertraline using cyclodextrin-modifi ed micel- chromatography determination of citalopram enantiomers using lar electrokinetic chromatography. Chromatographia 2004 , 60 , beta-cyclodextrin as a chiral mobile phase additive. J. AOAC Int. 469 – 473. 2006 , 89 , 65 – 70. Chen, X.; Duan, X.; Dai, X.; Zhong, D. Development and validation Erk, N. Rapid and simple methods for quantitative analysis of some of a liquid chromatographic/tandem mass spectrometric method antidepressant in pharmaceutical formulations by using fi rst for the determination of sertraline in human plasma. Rapid derivative spectrophotometry and HPLC. II Farmaco 2003 , 58 , Commun. Mass Spectrom. 2006 , 20 , 2483 – 2489. 1209 – 1216. Chu, S.; Metcalfe, C. D. Analysis of paroxetine, fl uoxetine and nor- Erk, N.; Biryol, I. Voltammetric and HPLC techniques for the deter- fl uoxetine in fi sh tissues using pressurized liquid extraction, mination of paroxetine hydrochloride. Pharmazie 2003 , 58 , mixed mode solid phase extraction cleanup and liquid chroma- 699 – 704. tography – tandem mass spectrometry. J. Chromatogr. A 2007 , Esrafi li, A.; Yamini, Y.; Shariati, S. Hollow fi ber-based liquid 1163 , 112 – 118. phase microextraction combined with high-performance liquid Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 117

chromatography for extraction and determination of some anti- the determination of fl uoxetine and norfl uoxetine enantiomers depressant drugs in biological fl uids. Anal. Chim. Acta 2007 , in human plasma by liquid chromatography. J. Chromatogr. B 604 , 127 – 133. Analyt. Technol. Biomed. Life Sci. 2003 , 784 , 375 – 383. FDA. 2 May, 2007. Available at: http://www.fda.gov/bbs/topics/ Gondova, T.; Halamova, D.; Spacayova, K. Simultaneous analysis NEWS/2007/NEW01624.html. Accessed 29 May, 2008. of new antidepressants by densitometric thin-layer chromatogra- Fernandes, C.; Jiayu, P.; Sandra, P.; Lanç as, F. M. Stir bar sorp- phy. J. Liq. Chromatogr. RT 2008 , 31 , 2429 – 2441. tive extraction-LC/MS for the analysis of fl uoxetine in plasma. Goodnough, D. B.; Baker, G. B.; Coutts, R. T. Simultaneous quan- Chromatographia 2006 , 64 , 517 – 521. tifi cation of fl uoxetine, norfl uoxetine, and desipramine using Fernandes, A.; Santos Neto, A. J.; Rodrigues, J. C.; Alves, C.; Mauro gas chromatography with nitrogen-phosphorus detection. J. Lanç as, F. Solid-phase microextraction – liquid chromatog- Pharmacol. Toxicol. Methods 1995, 34, 143–147. raphy (SPME– LC) determination of fl uoxetine and norfl uox- Green, R.; Houghton, R.; Scarth, J.; Gregory, C. Determination of etine in plasma using a heated liquid fl ow through interface. J. fl uoxetine and its major active metabolite norfl uoxetine in human Chromatogr. B 2007 , 847 , 217 – 223. plasma by liquid chromatography-tandem mass spectrometry. Fernandes, C.; Hoeck, E. V.; Sandra, P.; Lancas, F. M. Determination Chromatographia 2002 , 55 , 133 – 136. of fl uoxetine in plasma by gas chromatography – mass spectrom- Greiner, C.; Hiemke, C.; Bader, W.; Haen, E. Determination of citalo- etry using stir bar sorptive extraction. Anal. Chim. Acta 2008 , pram and escitalopram together with their active main metabolites 614 , 201 – 207. desmethyl(es-)citalopram in human serum by column-switching Ferretti, R.; Gallinella, B.; La Torre, F.; Turchetto, L. Validated chi- high performance liquid chromatography (HPLC) and spectro- ral high-performance liquid chromatographic method for the photometric detection. J. Chromatogr. B 2007, 848 , 391 – 39. determination of trans-(-)-paroxetine and its enantiomer in bulk Guo, X.; Fukushima, T.; Li, F.; Imai, K. Determination of fl uox- and pharmaceutical formulations. J. Chromatogr. B 1998 , 710 , etine and norfl uoxetine in rat plasma by HPLC with pre-column 157 – 164. derivatization and fl uorescence detection. Biomed. Chromatogr. Flores, J. R.; Nevado, J. J. B.; Salcedo, A. M. C.; Diaz, M. P. C. Non- 2003 , 17 , 1 – 5. aqueous capillary zone electrophoresis method for the analysis of Gutteck, U.; Rentsch, K. M. Therapeutic drug monitoring of 13 anti- paroxetine, tamoxifen, and their main metabolites in urine. Anal. depressant and fi ve neuroleptic drugs in serum with liquid chro- Chim. Acta 2004a , 512 , 287 – 295. matography-electrospray ionization mass spectrometry. Clin. Flores, J. R.; Nevado, J. J. B.; Salcedo, A. M. C.; Dí az, M. P. C. Chem. Lab. Med. 2003 , 41 , 1571 – 1579. Development and validation method for determination of Halvorsen, T. G.; Pedersen-Bjergaard, S.; Rasmussen, K. E. Liquid- Paroxetine and its metabolites by nonaqueous capillary electro- phase microextraction and capillary electrophoresis of cit- phoresis in human urine. Experimental design for evaluating the alopram, an antidepressant drug. J. Chromatogr. A 2001 , 909 , ruggedness of the method. Electrophoresis 2004b , 25 , 454 – 462. 87 – 93. Flores, J. R.; Salcedo, A. M. C.; Llerena, M. J. V.; Fernandez, L. M. Hattori, A.; Ito, K.; Iwai, M.; Masuda, K.; Mizutani, Y.; Senoo, H.; Micellar electrokinetic chromatographic method for the determi- Arinobu, T.; Kumazawa, T.; Ishii, A.; Suzuki, O. Rapid and nation of letrozole, citalopram and their metabolites in human simple determination of selective serotonin reuptake inhibitor urine. J. Chromatogr. A 2008 , 1185 , 281 – 290. (SSRI) in human serum or plasma by LC/MS and LC/MS/MS Foglia, J. P.; Sorisio, D.; Kirshner, M.; Pollock, B. G. Quantitative Jpn. J. Forens. Toxicol. 2005 , 23 , 162 – 163. determination of paroxetine in plasma by high-performance liq- Hattori, H.; Ito, K.; Iwai, M.; Arinobu, T.; Mizutani, Y.; Kumazawa, uid chromatography and ultraviolet detection. J. Chromatogr. B T.; Ishii, A.; Suzuki, O.; Seno, H. Rapid analysis of sertraline, 1997 , 693 , 147 – 151. fl uvoxamine, and paroxetine in serum specimens by LC-MS- Fontanille, R.; Jourdil, N.; Villier, C.; Bessard, G. Direct analysis of MS using a new polymer column. Forens. Toxicol. 2007 , 25 , fl uoxetine and norfl uoxetine in plasma by gas chromatography 100 – 103. with nitrogen-phosphorus detectio. J. Chromatogr. B 1997 , 692 , He, L.; Feng, F.; Wu, J. Determination of sertraline in human plasma 337 – 343. by high-performance liquid chromatography – electrospray ion- Frahnert, A.; Rao, M. L.; Grasmader, K. Analysis of eighteen anti- ization mass spectrometry and method validation. J. Chromatogr. depressants, four atypical antipsychotics and active metabolites Sci. 2005 , 43 , 532 – 535. in serum by liquid chromatography: a simple tool for therapeutic Higashi, Y.; Hiroki, M.; Youichi, F. Determination of fl uvoxamine drug monitoring. J. Chromatogr. B 2003 , 794 , 35 – 47. in rat plasma by HPLC with pre-column derivatization and fl uo- Franceschi, L.; Faggiani, A.; Furlanut, M. A simple method to moni- rescence detection using 4-fl uoro-7-nitro-2,1,3-benzoxadiazole. tor serum concentrations of fl uoxetine and its major metabolite Biomed. Chromatogr. 2005 , 19 , 771 – 777. for pharmacokinetic studies. J. Pharmaceut. Biomed. 2009 , 49 , Holladay, J. W.; Dewey, M. J.; Yoo, S. D. Quantifi cation of fl uox- 554 – 557. etine and norfl uoxetine serum levels by reversed-phase high- Frison, D. F.; Favretto, D.; Vogliardi, S.; Terranova, C.; Ferrara, S. performance liquid chromatography with ultraviolet detection. J. D. Quantifi cation of citalopram or escitalopram and their dem- Chromatogr. B Biomed. Sci. Appl . 1997 , 704 , 259 – 263. ethylated metabolites in neonatal hair samples by liquid chro- Jain, D. S.; Sanyal, M.; Subbaiah, G.; Pander, U. C.; Shrivastav, P. matography-tandem mass spectrometry. Ther. Drug Monit. 2008 , Rapid and sensitive method for the determination of sertraline in 30 , 467 – 473. human plasma using liquid chromatography – tandem mass spec- Fukushima, T.; Nakaaki, E.; Guo, X.; Li, F.; Vankeirsbilck, T.; trometry (LC – MS/MS). J. Chromatogr. B 2005 , 829 , 69 – 74. Baeyens, W. R. G.; Imai, K.; Toyooka, T. Determination of fl uox- Jawecki, A. N. Evaluation of the in vitro biotransformation of fl uoxetine and norfl uoxetine in rat brain microdialysis samples follow- etine with HPLC, mass spectrometry and ecotoxicological tests. ing intraperitoneal fl uoxetine administration. Anal. Chim. Acta Chemosphere 2007 , 70 , 29 – 35. 2004 , 522 , 99 – 104. Jia, F.; Gao, S. Y.; Li, Y.; Zeng, S. Determination of sertraline hydro- Gatti, G.; Bonomi, I.; Marchiselli, R.; Fattore, C.; Spina, E.; Scordo, chloride in human plasma by LC-MS/MS. Chin. Pharm. J. 2007 , G.; Pacifi ci, R.; Perucca, E. Improved enantioselective assay for 42 , 1023 – 1025. 118 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

Juan, H.; Zhiling, Z.; Huande, L. Simultaneous determination of fl u- chromatography – mass spectrometry. J. Chromatogr. B 2002 , oxetine, citalopram, paroxetine, venlafaxine in plasma by high 779 , 353 – 357. performance liquid chromatography – electrospray ionization Lencastre, R. P.; Matos, C. D.; Garrido, J.; Borges, F.; Garrido, E. M. mass spectrometry (HPLC– MS/ESI). J. Chromatogr. B 2005 , Voltammetric quantifi cation of fl uoxetine: application to qual- 820 , 33 – 39. ity control and quality assurance processes. J. Food Drug Anal. Khawam, E. A.; Laurancic, G.; Malone, D. A. Side effects of antide- 2006 , 14 , 242 – 246. pressants: an overview. Clev. Clin. J. Med. 2006 , 73 , 351. Levenson, M.; Holland, C. Overview for December 13 Meeting of Kim, K. M.; Jung, B. H.; Ho Choi, M.; Woo, J. S.; Paeng, K.-J.; Psychopharmacologic Drugs Advisory Committee (PDAC). Chung, B. C. Rapid and sensitive determination of sertraline in FDA (17 November, 2006); 75– 140. Available at: http://www. human plasma using gas chromatography-mass spectrometry. J. fda.gov/ohrms/dockets/ac/06/briefing/2006-4272b1-01-FDA. Chromatogr. B 2002 , 769 , 333 – 339. pdf. Accessed 22 September, 2007. Kirchherr, H.; Kuhn-Velten, W. N. Quantitative determination of for- Li, C.; Ji, Z.; Nan, F.; Shao, Q.; Liu, P.; Dai, J.; Zhen, J.; Yuan, H.; ty-eight antidepressants and antipsychotics in human serum by Xu, F.; Cui, J.; Huang, B.; Zhang, M.; Yu, C. Rapid Commun. HPLC tandem mass spectrometry: a multi-level, single-sample Mass Spectrom. 2002 , 16 , 1844 – 1850. approach. J. Chromatogr. B 2006 , 843 , 100 – 113. Li, K. M.; Murray, R. T.; Iain, S. M. Rapid quantitation of fl uoxetine Knoeller, J.; Vogt-Schenkel, R.; Bret ı , M. A. Simple and robust HPLC and norfl uoxetine in serum by micro-disc solid-phase extrac- method for the determination of paroxetine in human plasma. J. tion with high-performance liquid chromatography – ultraviolet Pharm. Biomed. Anal. 1995 , 13 , 635 – 638. absorbance detection. J. Chromatogr. B 2004a , 804 , 319 – 326. Kollroser, M.; Schober, C. An on-line solid phase extraction-liquid Li, K. M.; Thompson, M. R.; McGregor, I. S. Rapid quantitation chromatography-tandem mass spectrometry method for the of fl uoxetine and norfl uoxetine in serum by micro-disc solid- analysis of citalopram, fl uvoxamine, and paroxetine in human phase extraction with high-performance liquid chromatography plasma. Chromatographia 2003 , 57 , 133 – 138. – ultraviolet absorbance detection. J. Chromatogr. B 2004b , 804 , Korytar, P.; Janssen, H.- G.; Matisova, E.; Brinkman, U. A. Th. 319 – 326. Practical fast gas chromatography: methods, instrumentation and Liu, S.; Shinkai, N.; Kakubari, I.; Saitoh, H.; Noguchi, K.; Saitoh, T.; applications. TRAC Trends Anal. Chem. 2002 , 21 , 558 – 572. Yamauchi, H. Automated analysis of fl uvoxamine in rat plasma Kosel, M.; Eap, C. B.; Amey, M.; Baumann, P. Analysis of the using a column-switching system and ion-pair high-perfor- enantiomers of citalopram and its demethylated metabolites mance liquid chromatography. Biomed. Chromatogr. 2008 , 22 , using chiral liquid chromatogra. J. Chromatogr. B 1998 , 719 , 1442 – 1449. 234 – 238. Llerena, A.; Dorado, P.; Berecz, R.; Gonzalez, A.; Norberto, M. J.; Kovacevic, I.; Pokrajac, M.; Miljkovic, B.; Jovanovic, D.; Prostran, de la Rubia, A.; Caceres, M. Determination of fl uoxetine and M. Comparison of liquid chromatography with fl uorescence norfl uoxetine in human plasma by high-performance liquid chro- detection to liquid chromatography-mass spectrometry for the matography with ultraviolet detection in psychiatric patients. determination of fl uoxetine and norfl uoxetine in human plasma. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2003 , 783 , J. Chromatogr. B 2006 , 830 , 372 – 376. 25 – 31. Kristoffersen, L.; Bugge, A.; Lundanes, E.; Slordal, L. Simultaneous Lopez-Calull, C.; Dominguez, N. Determination of paroxetine determination of citalopram, fl uoxetine, paroxetine and their in plasma by high-performance liquid chromatography for metabolites in plasma and whole blood by high-performance liq- bioequivalence studies. J. Chromatogr. B 1999 , 724 , 393 – 398. uid chromatography with ultraviolet and fl uorescence detection. Lucangioli, S. E.; Hermida, L. G.; Tripodi, V. P.; Rodriguez, V. G.; J. Chromatogr. B 1999 , 734 , 229 – 246. Lopez, E. E.; Rouge, P. D.; Carducci, C. N. Analysis of cis - trans Labat, L.; Deveaux, M.; Dallet, P.; Dubost, J. P. Separation of new isomers and enantiomers of sertraline by cyclodextrin-modifi ed antidepressants and their metabolites by micellar electroki- micellar electrokinetic chromatography. J. Chromatogr. B 2000 , netic capillary chromatography. J. Chromatogr. B 2002 , 773 , 871 , 207 – 215. 17 – 23. Lucca, A.; Gentilini, G.; Lopez-Silva, S.; Soldarini, A. Simultaneous Lacassie, E.; Gaulier, J. M.; Marquet, P.; Rabatel, J. F.; Lachatre, determination of human plasma levels of four selective serotonin G. Methods for the determination of seven selective serotonin reuptake inhibitors by high-performance liquid chromatography. reuptake inhibitors and three active metabolites in human serum Ther. Drug Monit. 2000 , 22 , 271 – 276. using high-performance liquid chromatography and gas chroma- Macek, J.; Ptacek, P.; Klima, J. Rapid determination of citalopram in tography. J. Chromatogr. B 2000 , 742 , 229 – 238. human plasma by high-performance liquid chromatography. J. Lai, C.- T.; Gordon, E. S.; Kennedy, S. H.; Bateson, A. N.; Coutts, Chromatogr. B Biomed. Sci. Appl. 2001 , 755 , 279 – 285. R. T.; Baker, G. B. Determination of paroxetine levels in human MacLeod, S. L.; Sudhir, P.; Wong, C. S. Stereoisomer analysis of plasma using gas chromatography with electron-capture detec- wastewater-derived β -blockers, selective serotonin re-uptake tion. J. Chromatogr. B 2000 , 749 , 275 – 279. inhibitors, and salbutamol by high-performance liquid chroma- Lamas, J. P.; Salgado-Petinal, C.; Garcia-Jares, C.; Llompart, M.; tography – tandem mass spectrometry. J. Chromatogr. A 2007 , Cela, R.; Gomez, M. Solid-phase microextraction– gas chroma- 1170 , 23 – 33. tography – mass spectrometry for the analysis of selective sero- Malfara, W. R.; Bertucci, C.; Queiroz, M. E. C.; Carvalho, S. A. D.; tonin reuptake inhibitors in environmental water. J. Chromatogr. Bianchi, M. L. P.; Cesarino, E. J.; Crippa, J. A.; Queiroz, R. H. C. A 2004 , 1046 , 241 – 247. Reliable HPLC method for therapeutic drug monitoring of fre- Lefebvre, M.; Marchand, M.; Horowitz, J. M.; Torres, G. Detection quently prescribed tricyclic and nontricyclic antidepressants. J. of fl uoxetine in brain, blood, liver and hair of rats using gas chro- Pharmaceut. Biomed. 2007 , 44 , 955 – 962. matography-mass spectrometry. Life Sci. 1999 , 64 , 805 – 811. Mandrioli, R.; Saracino, M. A.; Ferrari, S.; Berardi, D.; Kenndler, Leis, H. J.; Windischhofer, W.; Fauler, G. Improved sample prepara- E.; Raggi, M. A. HPLC analysis of the second-generation antide- tion for the quantitative analysis of paroxetine in human plasma pressant sertraline and its main metabolite N -desmethylsertraline by stable isotope dilution negative ion chemical ionisation gas in human plasma. J. Chromatogr. B 2006 , 836 , 116 – 119. Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 119

Mandrioli, R.; Mercolini, L.; Ferranti, A.; Furlanetto, S.; Boncom- Nevado, J. J. B.; Flores, J. R.; Penalvo, G. C. Voltammetric behav- pagni, G.; Raggi, M. A. Determination of the antidepressant par- ior of fl uvoxamine using square-wave and adsorptive strip- oxetine and its three main metabolites in human plasma by liquid ping square-wave techniques. Determination in pharmaceutical chromatography with fl uorescence detection. Anal. Chim. Acta products. Electroanalysis 2000, 12 , 1059 – 1063. 2007 , 591 , 141 – 147. Nevado, J. J. B.; Llerena, M. J. V.; Salcedo, A. M. C.; Nuevo, E. A. Marriott, P. J. Encyclopedia of Analytical Science. In: Worsfold, J. Assay validation for three antidepressants in pharmaceutical P.; Townshend, A.; Poole, C., Eds. Elsevier: Amsterdam, 2005, formulations: practical approach using capillary gas chromatog- pp. 7 – 18. raphy. Pharmaceut. Biomed . 2005 , 38 , 52 – 59. Massaroti, P.; Cassiano, N. M.; Duarte, L. F.; Campos, D. R.; Nevado, J. J. B.; Llerena, M. J. V.; Cabanillas, C. G.; Robledo, V. R. Marchioretto, M. A. M.; Bernasconi, G.; Calafatti, S.; Barros, Screening of citalopram, fl uoxetine and their metabolites in F. A. P.; Meurer, E. C.; Pedrazzoli, J. Validation of a selective human urine samples by gas chromatography – mass spectrom- method for determination of paroxetine in human plasma by etry: a global robustness/ruggedness study. J. Chromatogr. A LC-MS/MS. J. Pharm. Pharmaceut. Sci. 2005 , 8 , 340 – 347. 2006a , 1123 , 130 – 133. Matsui, A.; Hoshino, M.; Matsui, A.; Okahira, A. Simultaneous deter- Nevado, J. J. B.; Llerena, M. J. V.; Cabanillas, C. G.; Robledo, V. R.; mination of citalopram and its metabolites by high-performance Buitrago, S. Sensitive capillary GC-MS-SIM determination of liquid chromatography with column switching and fl uorescence selective serotonin reuptake inhibitors: reliability evaluation by detection by direct plasma injection. J. Chromatogr. B 1995 , 668 , validation and robustness study. J. High Resolut. Chrom. 2006b , 299 – 307. 29 , 103 – 113. Maya, M. T.; Domingos, C. R.; Guerreiro, M. T.; Morais, J. A. Norouzi, P.; Daneshgar, P.; Ganjali, M. R.; Moosavi-Movahedi, A. Determination of the antidepressant fl uoxetine in human plasma Voltammetric behavior of fl uvoxamine using square-wave and by LC with UV detection. J. Pharmaceut. Biomed. 2000 , 23 , adsorptive stripping square-wave techniques. Determination in 989 – 996. pharmaceutical products. J. Brazil. Chem. Soc. 2007 , 18 , 231 – Medyantseva, E. P.; Varlamova, R. M.; Gimaletdinova, D. A.; 238. Fattakhova, A. N.; Budnikov, G. K. An amperometric mono- Nouws, H. P. A.; Delerue-Matos, C.; Barros, A. A.; Rodrigues, J. A. amine oxidase biosensor for determining some antidepressants. Electroanalytical study of the antidepressant sertraline. J. J. Anal. Chem. 2008 , 63 , 275 – 279. Pharmaceut. Biomed. 2005 , 39 , 290 – 293. Meineke, I.; Schreeb, K.; Kress, I.; Gundert-Remy, U. Routine mea- Nouws, H. P. A.; Delerue-Matos, C.; Barros, A. A.; Rodr ı gues, J. A. surement of fl uoxetine and norfl uoxetine by high-performance Electroanalytical determination of paroxetine in pharmaceuti- liquid chromatography with ultraviolet detection in patients cals. J. Pharmaceut. Biomed. 2006a , 42 , 341 – 346. under concomitant treatment with tricyclic antidepressants. Ther. Nouws, H. P. A.; Delerue-Matos, C.; Barros, A. A. Electrochemical Drug Monit. 1998 , 20 , 14 – 19. determination of citalopram by adsorptive stripping voltammetry Melo, L. P.; Nogueira, A. M.; Lan, F. M.; Queiroz, M. E. C. Poly- – determination in pharmaceutical products. Anal. Lett. 2006b , dimethylsiloxane/polypyrrole stir bar sorptive extraction and 39 , 1907 – 1915. liquid chromatography (SBSE/LC-UV) analysis of antidepres- Nouws, H. P. A.; Delerue-Matos, C.; Barros, A. A.; Rodrigues, J. A.; sants in plasma samples. Anal. Chim. Acta 2009 , 633 , 57 – 64. Santos-Silva, A.; Borges, F. Square-wave adsorptive-stripping Meng, Q. H.; Gauthier, D. Simultaneous analysis of citalopram voltammetric detection in the quality control of fl uoxetine ana- and desmethylcitalopram by liquid chromatography with fl uo- lytical letters. Anal. Lett. 2007 , 40 , 1131 – 1146. rescence detection after solid-phase extraction. Clin. Biochem. Nouws, H. P. A.; Delerue-Matos, C.; Barros, A. A.; Maesen, E.; 2005 , 38 , 282 – 285. Moreira, S. C. P. A.; Neves, M. M. P. S. Static and hydrodynamic Millan, S.; Goicolea, M. A.; Sanchez, A.; Gomez-Caballero, A.; monitoring of citalopram based on its electro-oxidation behavior Sampedro, M. C.; Unceta, N.; Barrio, R. J. Stereoselective deter- at a glassy-carbon surface. Anal. Lett. 2008 , 41 , 2171 – 2185. mination of demethyl- and didemethyl-citalopram in rat plasma Ohkubo, T.; Sh ı moyama, R.; Otan ı , K.; Yosh ı da, K.; H ı guch ı , H.; and brain tissue by liquid chromatography with fl uorescence Sh ı m ı zu, T. High-performance liquid chromatographic determi- detection using precolumn derivatization. Biomed. Chromatogr. nation of fl uvoxamine and fl uvoxamino acid in human plasma. 2008 , 22 , 265 – 271. Anal. Sci. 2003 , 19 , 859 – 864. Molander, P.; Thomassen, A.; Kristoffersen, L.; Greibrokk, T.; Ö nal, A.; Ö ztun ç , A. Determination of paroxetine in human plasma Lundanes, E. Simultaneous determination of citalopram, fl uox- by high-performance liquid chromatography using 7,7,8,8-tetra- etine, paroxetine and their metabolites in plasma by tempera- cyanoquinodimethane as the derivatization reagent. Ther. Drug ture-programmed packed capillary liquid chromatography with Monit. 2006 , 28 , 180 – 184. on-column focusing of large injection volumes. J. Chromatogr. Onal, A.; Kepek ç i, S¸. E.; Ç etin, S. M.; Ertü rk, S. “ Spectrophotometric B 2001 , 766 , 77 – 87. determination of certain antidepressants in pharmaceutical prep- Moraes, M. O.; Lerner, F. E.; Corso, G.; Bezzerra, F. A. F.; Moraes, M. rations. ” J. AOAC Int. 2006 , 89 , 966 – 971. E. A.; De Nucci, G. Fluoxetine bioequivalence study: quantifi cation Parham, H.; Pourreza, N.; Shafi ekhani, H. Solvent extraction- of fl uoxetine and norfl uoxetine by liquid chromatography coupled spectrophotometric determination of trace amounts of fl uox- to mass spectrometry. J. Clin. Pharmacol. 1999 , 39 , 1053 – 1061. etine in pharmaceutical formulations. J. Anal. Chem. 2008 , 63 , Musenga, A.; Kenndler, E.; Mercolini, L.; Amore, M.; Fanali, S.; 626 – 628. Raggi, M. A. Determination of sertraline and N-desmethylsertra- Patel, B. N.; Sharma, N.; Sanyal, M.; Shrivastav, P. S. Analysis of line in human plasma by CE with LIF detection. Electrophoresis second-generation antidepressant drug, sertraline and its active 2007 , 28 , 1823 – 1831. metabolite, N-desmethyl sertraline in human plasma by a sensi- Nevado, J. J. B.; Salcedo, A. M. C. Method development and vali- tive and selective liquid chromatography-tandem mass spectrom- dation for the simultaneous determination of fi ve selective etry method. J. Chromatogr. B 2009 , 877 , 221 – 229. serotonin reuptake inhibitors by capillary zone electrophoresis. Pedersen-Bjergaard, S.; Rasmussen, K. E.; Brekke, A.; Ho, T. S.; Chromatographia 2002 , 55 , 369 – 373. Halvorsen, T. G. Liquid-phase microextraction of basic drugs 120 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

– selection of extraction mode based on computer calculated Roque da Silva, A. M. S.; Lima, J. C.; Oliva Teles, M. T.; Brett, A. solubility data. J. Sep. Sci . 2005 , 28 , 1195 – 1203. M. O. Electrochemical studies and square wave adsorptive strip- Peyton, A. L.; Carpenter, R.; Rutkovvski, K. The stereospecifi c ping voltammetry of the antidepressant fl uoxetine. Talanta 1999 , determination of fl uoxetine and norfl uoxetine enantiomers in 49 , 611 – 617. human plasma by high-pressure liquid chromatography (HPLC) Sabbioni, C.; Bugamelli, F.; Varani, G.; Mercolini, L.; Musengaa, with fl uorescence detection. Pharm. Res. 1991 , 8 , 1528 – A.; Saracino, M. A.; Fanali, S.; Raggi, M. A. A rapid HPLC- 1532. DAD method for the analysis of fl uoxetine and norfl uoxetine in Pistos, C.; Panderi, I.; Atta-Politou, J. Liquid chromatography – posi- plasma from overdose patients. J. Pharmaceut. Biomed. 2004 , tive ion electrospray mass spectrometry method for the quantifi - 36 , 351 – 356. cation of citalopram in human plasma. J. Chromatogr. B 2004 , Saber, A. L. On-line solid phase extraction coupled to capillary 810 , 235 – 244. LC-ESI-MS for determination of fl uoxetine in human blood Psillakis, E.; Kalogerakis, N. Hollow-fi bre liquid-phase microextrac- plasma. Talanta 2009 , 78 , 295 – 299. tion of phthalate esters from water. Trends Anal. Chem. 2003 , Salsbury, J. S.; Isbester, P. K. Quantitative 1 H NMR method for the 22 , 565. routine spectroscopic determination of enantiomeric purity of Pucci, V.; Fanali, S.; Sabbioni, C.; Raggi, M. A. Separation of fi ve active pharmaceutical ingredients fenfl uramine, sertraline, and recently commercialized selective serotonin reuptake inhibitor paroxetine. Magn. Reson. Chem. 2005 , 43 , 910 – 917. antidepressants by capillary electrophoresis. J. High Resolut. Santos, F. J.; Galceran, M. T. The application of gas chromatography Chrom. 2002 , 25, 1096 – 1100. to environmental analysis. TRAC Trends Anal. Chem. 2002 , 21 , Pujadas, M.; Pichini, S.; Civit, E.; Santamarina, E.; Perez, K.; de la 672 – 685. Torre, R. A simple and reliable procedure for the determination Santos-Neto, A. J.; Bergquist, J.; Lanc, F. M.; Sjoberg, P. J. R. of psychoactive drugs in oral fl uid by gas chromatography – mass Simultaneous analysis of fi ve antidepressant drugs using direct spectrometry. J. Pharmaceut. Biomed. 2007 , 44 , 594 – 601. injection of biofl uids in a capillary restricted-access media- Queiroz, M. E. C.; Oliveira, E. B.; Breton, F.; Pawliszyn, J. liquid chromatography-tandem mass spectrometry system. J. Immunoaffi nity in-tube solid phase microextraction coupled Chromatogr. A 2008 , 1189 , 514 – 522. with liquid chromatography – mass spectrometry for analysis Saracino, M. A.; Mercolini, L.; Flotta, G.; Albers, L. J.; Merli, R.; of fl uoxetine in serum samples. J. Chromatogr. A 2007 , 1174 , Raggi, M. A. Simultaneous determination of fl uvoxamine isomers 72 – 77. and quetiapine in human plasma by means of high-performance Raggi, M. A.; Mandrioli, R.; Casamenti, G.; Volterra, V.; Desiderio, liquid chromatography. J. Chromatogr. B 2006 , 843 , 227 – C.; Fanali, S. Determination of recent antidepressant citalopram 233. in human plasma by liquid chromatography-fl uorescence detec- Sauvage, F. L.; Gaulier, J. M.; Lach â tre, G.; Marquet, P. A fully tion. Chromatographia 1999 , 50 , 423 – 427. automated turbulent-fl ow liquid chromatography-tandem mass Raggi, M. A.; Pucci, V.; Mandrioli, R.; Sabbioni, C.; Fanali, S. Deter- spectrometry technique for monitoring antidepressants in human mination of recent antidepressant citalopram in human plasma by serum. Ther. Drug Monit. 2006 , 28 , 123 – 130. liquid chromatography-fl uorescence detection. Chromatographia Schaller, D.; Hilder, E. F.; Haddad, P. R. ‘ Separation of antidepres- 2003 , 57 , 273 – 278. sants by capillary electrophoresis with in-line solid-phase extrac- Ratola, N.; Alves, A.; Kalogerakis, N.; Psillakis, E. Hollow-fi bre tion using a novel monolithic adsorbent.’ Anal. Chim. Acta 2006 , liquid-phase microextraction: a simple and fast cleanup step used 556 , 104 – 111. for PAHs determination in pine needles. Anal. Chim. Acta 2008 , Schatz, D. S.; Saria, A. Simultaneous determination of paroxetine, 6 , 70 – 78. risperidone and 9-hydroxyrisperidone in human plasma by high- Raza, A. Development and application of spectrophotometric meth- performance liquid chromatography with coulometric detection. ods for the determination of citalopram hydrobromide in dosage Pharmacology 2000 , 60 , 51 – 56. forms. Chem. Pharm. Bull. 2006 , 54 , 432 – 434. Segura, M.; Ortuno, J.; Farre, M.; Pacifi ci, R.; Pichini, S.; Joglar, J.; Reddy, B. V.; Reddy, K. V. N. S.; Sreeramulu, J.; Kanumula, G. V. Segura, J.; de la Torre, R. Quantitative determination of parox- Simultaneous determination of olanzapine and fl uoxetine by etine and its 4-hydroxy-3-methoxy metabolite in plasma by high- HPLC. Chromatographia 2007 , 66 , 111 – 114. performance liquid chromatography/electrospray ion trap mass Reubsaet, J. L. E.; Bjergaard, S. P. Screening for central nervous spectrometry: application to pharmacokinetic studies. Rapid system-stimulating drugs in human plasma by liquid chromatog- Commun. Mass Spectrom . 2003 , 17 , 1455 – 1461. raphy with mass spectrometric detection. J. Chromatogr. A 2004 , Serebruany, V.; Malinin, A.; Dragan, V.; Atar, D.; Zyl, L. V.; Dragan, 1031 , 203 – 211. A. Fluorimetric quantitation of citalopram and escitalopram in Reymond, M.; Amey, A.; Souche, S.; Lambert, H.; Konrat, C.; Eap, plasma: developing an express method to monitor compliance in B.; Baumann, P. Determination of plasma levels of citalopram clinical trials. Clin. Chem. Lab. Med. 2007 , 45 , 513 – 520. and its demethylated and deaminated metabolites by gas chro- Shah, A. R.; Shah, N. J.; Suhagia, B. N.; Patel, N. M. Simultaneous matography and gas chromatography – mass spectrometry. J. assay of olanzapine and fl uoxetine in tablets by column high- Chromatogr. B 1993 , 616 , 221 – 228. performance liquid chromatography and high-performance thin- Rodriguez-Mozas, S.; de Alda, M. J. L.; Barcelo, D. Advantages and layer chromatography. J. AOAC Int. 2007 , 90 , 1573 – 1578. limitations of on-line solid phase extraction coupled to liquid Shah, J.; Jan, M. R.; Rehman, F. Flow injection spectrophotometric chromatography – mass spectrometry technologies versus bio- determination of fl uoxetine in bulk and in pharmaceutical prepa- sensors for monitoring of emerging contaminants in water. J. rations. J. Chilean Chem. Soc. 2008 , 53 , 1605 – 1608. Chromatogr. A 2007 , 1152 , 97 – 115. Shamsipur, M.; Dastjerdi, L. S.; Haghgoo, S.; Armspach, D.; Matt, Rompa, M.; Kremer, E.; Zygmunt, B. Derivatisation in gas chro- D.; Aboul-Enein, H. Y. Chiral selectors for enantioresolution and matographic determination of acidic herbicides in aqueous quantitation of the antidepressant drug fl uoxetine in pharmaceuti- environmental samples. Anal. Bioanal. Chem. 2003 , 377 , 590 – cal formulations by 19 F NMR spectroscopic method. Anal. Chim. 599. Acta 2007 , 601 , 130 – 138. Z. S¸ ent ürk et al.: Analytical methods for determination of SSRIs 121

Sharma, R. N.; Bagul, M. S.; Chaturvedi, S. C.; Vasu, K. K.; Rajani, Trefi , S.; Gilard, V.; Balayssac, S.; Malet-Martino, M.; Martino, R. M. Validated TLC densitometric method for the quantifi cation of Quality assessment of fl uoxetine and fl uvoxamine pharmaceu- paroxetine hydrochloride in solid dosage form. Sci. Publ. Indian tical formulations purchased in different countries or via the Pharm. Assoc. 2007 , 69 , 436 – 438. Internet by 19 F and 2D DOSY 1 H NMR. J. Pharmaceut. Biomed. Shen, Z.; Wang, S.; Bakhtiar, R. Enantiomeric separation and quan- 2008 , 46 , 707 – 722. tifi cation of fl uoxetine (Prozac ® ) in human plasma by liquid Ulrich, S. Direct stereoselective assay of fl uoxetine and norfl uoxetine chromatography/tandem mass spectrometry using liquid-liq- enantiomers in human plasma or serum by two-dimensional gas uid extraction in 96-well plate format. Rapid Commun. Mass – liquid chromatography with nitrogen – phosphorus selective Spectrom. 2002 , 16 , 332 – 338. detection. J. Chromatogr. B 2003 , 783 , 481 – 490. Shin, J. G.; Kim, K. A.; Yoon, Y. R.; Cha, I. J.; Kim, Y. H.; Shin, S. G. Ulu, S. T. HPLC method for the determination of fl uvoxamine in Rapid simple high-performance liquid chromatographic determi- human plasma and urine for application to pharmacokinetic stud- nation of paroxetine in human plasma. J. Chromatogr. B 1998 , ies. J. Pharmaceut. Biomed. 2007 , 43 , 1444 – 1451. 713 , 452 – 456. Unceta, N.; Barrondo, S.; de Azua, I. R.; Gomez-Caballero, A.; Shinozuka, T.; Terada, M.; Tanaka, E. Solid-phase extraction and Goicolea, M. A.; Salles, J.; Barrio, R. J. Determination of fl uox- analysis of 20 antidepressant drugs in human plasma by LC/MS etine, norfl uoxetine and their enantiomers in rat plasma and brain with SSI method. Forens. Sci. Int . 2006 , 162 , 108 – 112. samples by liquid chromatography with fl uorescence detection. Silva, B. J. G.; Lanç as, F. M.; Queiroz, M. E. C. In-tube solid-phase J. Chromatogr. B 2007 , 852 , 519 – 528. microextraction coupled to liquid chromatography (in-tube Unceta, N.; Gomez-Caballero, A.; Sanchez, A.; Millan, S.; Sampedro, SPME/LC) analysis of nontricyclic antidepressants in human M. C.; Goicolea, M. A.; Salles, J.; Barrio, R. J. Simultaneous plasma. J. Chromatogr. B 2008 , 862 , 181 – 188. determination of citalopram, fl uoxetine and their main metabo- Singh, S. S.; Shah, H.; Gupta, S.; Jain, M.; Sharma, K.; Thakkar, lites in human urine samples by solid-phase microextraction P.; Shah, R. Liquid chromatography – electrospray ionisation coupled with high-performance liquid chromatography. J. mass spectrometry method for the determination of escitalopram Pharmaceut. Biomed. 2008 , 46 , 763 – 770. in human plasma and its application in bioequivalence study. Vela, M. H.; Garcia, M. B. Q.; Montenegro, M. C. B. S. M. J. Chromatogr. B 2004 811 , , 209 – 215. Electrochemical behaviour of sertraline at a hanging mercury Skibinski, R.; Misztal, G.; Olajossy, M. High performance liquid drop electrode and its determination in pharmaceutical products. chromatographic determination of fl uvoxamine and paroxetine Fres. J. Anal. Chem. 2001 , 369 , 563 – 566. in plasma. Chem. Analityczna 2000 , 45 , 815 – 823. Vergi-Athanasiou, N.; Atta-Politou, J.; Koupparis, M.; Spyropoulos, Smyth, W. F.; Leslie, J. C.; McClean, S.; Hannigan, B.; McKenna, J. Development and validation of an HPLC method, with fl uo- H. P.; Doherty, B.; Joyce, C.; O ’ Kane, E. The characterisation rescence detection, for simultaneous determination of paroxetine of selected antidepressant drugs using electrospray ionisation and its metabolites in plasma. J. Liq. Chromatogr. RT 2007 , 30 , with ion trap mass spectrometry and with quadrupole time-of- 1641 – 1655. fl ight mass spectrometry and their determination by high-per- Vivekanand, V. V.; Kumar, V. R.; Mohakhud, P. K.; Reddy, G. O. Rf formance liquid chromatography/electrospray ionisation tandem is the relative fl uorescence intensity and C is the concentration mass spectrometry. Rapid Commun. Mass Spectrom. 2006 , 20 , of paroxetine – HCl. J. Pharm. Biomed. Anal. 2003 , 33 , 803 – 1637 – 1642. 809. Somenath, M. Sample Preparation Techniques in Analytical Vlase, L.; Imre, S.; Leucuta, S. Determination of fl uoxetine and its Chemistry. Wiley-Interscience; 2003, p. 113. N Souverain, S.; Mottaz, M.; Cherkaoui, S.; Veuthey, J. L. Rapid analy- -desmethyl metabolite in human plasma by high-performance sis of fl uoxetine and its metabolite in plasma by LC-MS with liquid chromatography. Talanta 2005 , 66 , 659 – 663. column-switching approach. Anal. Bioanal. Chem. 2003 , 377 , Wang, J. Stripping Analysis: Principles, Instrumentation, and 880 – 885. Applications. VCH Publishers, Inc.: Deerfi eld Beach, FL, 1985. Stone, M. B.; Jones, M. L. Overview for December 13 Meeting of Wang, H.; Liu, W.; Guan, Y. In-tube solid-phase microextraction and Psychopharmacologic Drugs Advisory Committee (PDAC). on-line coupling with high-resolution GC. LC-GC Europe 2004 , FDA; pp. 11 – 74. Available at: http://www.fda.gov/ohrms/ 17 , 144 – 151. dockets/ac/06/briefi ng/2006-4272b1-01-FDA.pdf. Accessed 22 Waschgler, R.; Hubmann, M. R.; Conca, A.; Moll, W.; Koenig, P. September, 2007. Simultaneous quantifi cation of citalopram, clozapine, fl uoxetine, Su, H. L.; Hsieh, Y. Z. Using cation-selective exhaustive injection norfl uoxetine, maprotiline, desmethylmaprotiline and trazodone and sweeping micellar electrokinetic chromatography to deter- in human serum by HPLC analysis. Int. J. Clin. Pharm. Ther. mine selective serotonin reuptake inhibitors. J. Chromatogr. A 2002 , 40 , 554 – 559. 2008 , 1209 , 253 – 259. Weng, N.; Eerkes, A. Development and validation of a hydrophilic Sutherland, F. C. W.; Badenhorst, D.; de Jager, A. D.; Scanes, T.; interaction liquid chromatography-tandem mass spectrometric Hundt, H. K. L.; Swart, K. J.; Hundt, A. F. Sensitive liquid chro- method for the analysis of paroxetine in human plasma. Biomed. matographic-tandem mass spectrometric method for the determi- Chromatogr. 2004, 18, 28–36 nation of fl uoxetine and its primary active metabolite norfl uoxetine Wille, S. M. R.; Maudens, K. E.; Van Peteghem, C. H.; Lambert, in human plasma. J. Chromatogr. A 2001, 914 , 45 – 51. W. E. E. Development of a solid phase extraction for 13 ‘new’ Tienpont, B.; David, F.; Desmet, K.; Sandra, P. Stir bar sorptive generation antidepressants and their active metabolites for gas extraction-thermal desorption-capillary GC-MS applied to bio- chromatographic-mass spectrometric analysis. J. Chromatogr. B logical fl uids. Anal. Bioanal. Chem. 2002 , 373 , 46 – 55. 2005 , 1098 , 19 – 29. Tournel, G.; Houdret, N.; Hedouin, V.; Deveaux, M.; Gosset, D.; Wille, S. M. R.; Heeb, P. V.; Neels, H. M.; Peteghem, C. H. V.; Lhermitte, M. High-performance liquid chromatographic method Lambert, W. E. Comparison of electron and chemical ioniza- to screen and quantitate seven selective serotonin reuptake inhib- tion modes by validation of a quantitative gas chromatographic itors in human serum. J. Chromatogr. B 2001 , 761 , 147 – 158. – mass spectrometric assay of new generation antidepressants 122 Z. S ¸ ent ürk et al.: Analytical methods for determination of SSRIs

and their active metabolites in plasma. J. Chromatogr. A 2007 , sourced ﬂ uoxetine hydrochloride drug substances and formula- 1176 , 236 – 245. tions. Chromatographia 1997 , 46 , 511 – 523. Wille, S. M. R.; Cooreman, S. G.; Neels, H. M.; Lambert, W. E. E. Zainaghi, A.; Lanchote, V. L.; Queiroz, R. H. C. Determination of Relevant issues in the monitoring and the toxicology of old and paroxetine in geriatric depression by high-performance liquid new antidepressants. Crit. Rev. Clin. Lab. Sci. 2008 , 45 , 1 – 66. chromatography. Pharmacol. Res. 2003 , 48 , 217 – 221. Wille, S. M. R.; De Letter, E. A.; Piette, M. H. A.; Overschelde, L. Zhu, Z.; Neirinck, L. High-performance liquid chromatography – K. V.; Peteghem, C. H. V.; Lambert, W. E. Determination of anti- mass spectrometry method for the determination of paroxetine depressants in human postmortem blood, brain tissue, and hair in human plasma. J. Chromatogr. B 2002 , 780 , 295 – 300. using gas chromatography – mass spectrometry. Int. J. Legal Zhu, L.; Zhu, L.; Lee, H. K. Liquid-liquid-liquid microextraction Med. 2009 , 123 , 451 – 458. of nitrophenols with a hollow ﬁ ber membrane prior to capil- Wirth, D. D.; Olsen, B. A.; Hallenbeck, D. K.; Lake, M. E.; Gregg, lary liquid chromatography. J. Chromatogr. A 2001 , 924 , 407 – S. M.; Perry, E. M. Screening methods for impurities in multi- 414.