www.elsevier.com/locate/forsciint

Available online at www.sciencedirect.com

al 178 (2008) 192–198

Forensic Science Internation

Off-line HPLC method combined to LC–MS for the determination

of sildenafil and its active metabolite in post-mortem human

blood according to confirmation criteria

Constantinos Pistos *, Ioannis Papoutsis, Artemis Dona, Maria Stefanidou,Sotiris Athanaselis, Constantinos Maravelias, Chara Spiliopoulou

University of Athens, Medical School, Laboratory of Forensic Medicine and Toxicology, 75M. Asias Str., Goudi 11527, Athens, Greece

Received 17 November 2007; received in revised form 2 March 2008; accepted 24 March 2008

Available online 16 May 2008

Abstract

A simple HPLC method has been validated for the determination of sildenafil and its active metabolite (N-desmethylsildenafil) in human blood,

using an octadecyl silica (ODS) hypersil column. The chromatographic run time is less than 25 min using a mobile phase of 35:65 (v/v)

acetonitrile–0.015 M disodium hydrogen phosphate (Na2HPO4), triethylamine 0.1%, pH 7.4 at 1 mL/min flow rate and UV–vis detection at

230 nm. The method is linear in the concentration range of 10–500 ng/mL (r > 0.999, n = 5) for each analyte, with relative standard deviation

(R.S.D.) less than 5.05%. Interday and intraday errors were found to be �11.94%. The limits of detection and quantitation for both analytes were

5.0 ng/mL (s/n > 3) and 10.0 ng/mL (s/n > 10), respectively. The method was applied in two post-mortem human blood samples, concerning two

fatal cases from sildenafil citrate use, reported for the first time in Greece, and the results were further confirmed with LC–MS. The method is

proposed as supplementary to LC–MS when inadequate mass fragmentation does not provide information appropriate to meet confirmation

criteria.

# 2008 Elsevier Ireland Ltd. All rights reserved.

Keywords: HPLC UV–vis; Confirmation criteria; Sildenafil; N-Desmethylsildenafil

1. Introduction

Sildenafil (S) (1-[4-ethoxy-3-(6,7-dihydro-1-methyl-7-oxo-

3-propyl-1H-pyrazolo-[4,3-d]pyrimidin-5-yl) phenylsulpho-

nyl]-4-methylpiperazine) (Fig. 1a) has been widely prescribed

for erectile dysfunction [1–4] and its bioavailabilty, metabo-

lism, elimination route and pharmacokinetics have been

extended reported [3,5]. Mean maximum sildenafil plasma

concentrations measured after a single oral dose of 100 mg to

healthy male volunteers is 450 ng/mL plasma. The lower

therapeutic concentrations in human plasma after a 25 mg

single oral dose are approximately 7 ng/mL [5].

Many analytical methods, using high performance liquid

chromatography (HPLC), have been published for quantification

of the parent drug sildenafil in plasma and not for its active

* Corresponding author. Tel.: +30 210 7462433; fax: +30 210 7716098.

E-mail address: cpistos@med.uoa.gr (C. Pistos).

0379-0738/$ – see front matter # 2008 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.forsciint.2008.03.018

metabolite, using ultraviolet visible (UV–vis) detector [6,7], or a

liquid chromatography system combined with a triple quadru-

pole mass spectrometry detector [8], as well as in oral fluids using

a liquid chromatography single mass spectrometry system [9].

Lewis and Johnson [10] reported the detection of both sildenafil

and N-desmethylsildenafil in post-mortem fluids and tissues,

while Weinmann et al. [11] reported their detection in urine and

tissue samples using liquid chromatography tandem mass

spectrometry (LC–MS/MS) system. Al-Ghazawi et al. [12]

developed a method for the determination of both analytes in

plasma using electrochemical detection, Cooper et al. [13] in

plasma using UV–vis detector and Saisho et al. [14] in human

hair by GC–MS. Although most of these methods are sensitive

and accurate, none of them describe how to overcome the

inadequate LC–MS fragmentation of both sildenafil and its active

metabolite in human blood. According to our experiments, the

proposed methods in the literature for the determination of both

compounds in plasma appear to be problematic when they are

applied in human blood. Considering that at a number of cases,

Fig. 1. Chemical structure of (a) sildenafil and (b) N-desmethylsildenafil.

Fig. 2. Representative mass fragmentation obtained in positive mode and

selected ion monitoring from the analysis of a standard solution 250 ng/mL

sildenafil (tr = 13.3 min, m/z 475, 461).

C. Pistos et al. / Forensic Science International 178 (2008) 192–198 193

human blood is the only biological fluid available, and especially

in post-mortem cases, where we cannot isolate plasma or serum

or the body is putrefied, toxicologists are obliged to work with

whole blood.

Two sudden-death suspicious cases for sildenafil citrate use,

reported for the first time in Greece, led us to the development

of a sensitive method for the simultaneous detection of both

analytes in human blood. This paper describes a simple,

specific and accurate reverse-phase liquid chromatographic

method, which allows the concurrent quantification of sildenafil

and its active metabolite N-desmethylsildenafil (Fig. 1b), at

lower concentrations in human blood using only 0.5 mL of

sample. The ability of the method to detect sildenafil and its

active metabolite even when a single dose of a 25 mg

formulation is administered is an essential factor. The

quantitation of the metabolite would provide more data on

sildenafil metabolite and could help the estimation of the time

of death.

Initially, we investigated the use of gas chromatography

mass spectrometry (GC–MS) method in these samples

without success. GC–MS appears to be problematic for the

detection of sildenafil and particularly for its metabolite N-

desmethylsildenafil. According to a number of experiments in

our laboratory, this detection failure seems to present during

the derivatisation process probably due to the matrix effect.

On the other hand, the detection of these compounds without a

derivatisation step leads to a higher sensitivity limit. These

findings seem to be confirmed not only by the absence of GC–

MS methodology for the simultaneous detection of these

compounds in literature, but also by the contradictory results

on the parent drug [15,16]. The next step was to investigate

the mass fragmentation procedure of both compounds in a

liquid chromatography single mass spectrometer. The result

was a poor number of fragments and thus we decided to

develop a supplementary HPLC UV–vis method for the

determination of these compounds and use it in combination

with LC–MS in order to finally meet the confirmation criteria.

Fig. 2 represents the inadequate fragmentation of sildenafil in

LC–MS, after injection of standard solution at 250 ng/mL (m/

z 475! 461 SIM, positive mode).

2. Experimental

2.1. Reagents and materials

Acetonitrile and absolute methanol of HPLC grade (Lichrosolv1) were

obtained from Merck (Darmstadt, Germany). Water was deionised and further

purified by means of a Direct-Q water purification system from Millipore SA

(Molsheim, France). Disodium hydrogen phosphate dihydrate (Na2H-

PO4�2H2O), formic acid, triethylamine (TEA), phosphoric acid, isopropanol,

chloroform and n-pentane were of analytical grade and were obtained also from

Merck. Ammonium formate of analytical grade was obtained from Alfa Aesar

GmbH Co KG (Karlsruhe, Germany). Sildenafil citrate and N-desmethylsilde-

nafil were provided from Pfizer Hellas A. E. (Athens, Greece). Tadalafil (Cialis)

and vardenafil (Levitra) were provided from Pharmaserv-Lilly A.E.V.E. and

Bayer Hellas A.V.E. E., respectively. Drug free human blood was obtained from

blood units of the blood bank of Aghia Sophia Pediatric Hospital (Athens,

Greece). Octadecyl silica (ODS) Thermo Hypersil and symmetry analytical

columns were purchased from Thermo Fisher Scientific Inc. (Waltham, MA,

USA) and Waters (Milford, MA, USA), respectively.

2.2. Apparatus

Chromatography was performed with a HPLC system consisting of a Jasco

Model 880 PU pump fitted with a model 880-2 Ternary Gradient Unit and a

Model 875 UV variable wavelength UV–vis from Japan Spectroscopic Co. Ltd.

(Tokyo, Japan). The pump was used under isocratic conditions on manual mode

and detector was operated at 230 nm. The system fitted with a Model 7125

manual injector Rheodyne (Cotati, CA, USA) and a 50 mL sample loop.

Samples were chromatographed on a 25 cm � 4.6 mm (i.d.), 5 mm ODS

hypersil reversed-phase column. A Hewlett Packard HP 3394A integrator from

Hewlett Packard Company (Avondale, PA, USA) was used to record chroma-

tograms, at peak area mode (chart speed 0.2 cm/min). LC–MS 2010 EV by

Shimadzu Corporation (Kyoto, Japan) was used to obtain the mass spectrum of

the samples using a 15 cm � 2.1 mm (i.d.), 5 mm C18 symmetry analytical

column. A Millipore filtration system from Millipore SA (Molsheim, France)

with type HV Millipore filters (pore size 0.45 mm) was used, for degassing the

mobile phase under vacuum. Vortex was supplied by Heidolph Instruments

GMBH & Co KG (Schwabach, Germany). A Fisher isotemp dry bath Model

145 from Thermo Fisher Scientific Inc. (Waltham, MA, USA) was used for the

evaporation of the extraction solvents.

C. Pistos et al. / Forensic Science International 178 (2008) 192–198194

2.3. Methods

2.3.1. Calibration curves

Two separate stock solutions containing 1.0 mg/mL of sildenafil and N-

desmethylsildenafil were prepared in absolute methanol and stored at �20 8C.

The stability of stock solutions was checked comparing the peak area of frozen

versus a fresh solution originated by a new weight of the reference standard. Stock

solutions were stable for up to 1 month. A combined working solution containing

both sildenafil and N-desmethylsildenafil was prepared in absolute methanol

containing 10 mg/mL from stock standard solutions of each compound. Addi-

tional dilutions were prepared containing 0.1, 0.2, 0.5, and 5.0 mg/mL of sildenafil

and N-desmethylsildenafil by appropriate dilutions of the reference solution with

absolute methanol. Blood standards for calibration curves were prepared by

spiking 0.9 mL aliquots of drug free human blood with 100 mL of the above-

mentioned working solutions, to make sildenafil and N-desmethylsildenafil blood

standards ranging from 10 to 500 ng/mL. Calibration graphs of the recovered

standards were prepared for each of the 5 days of analysis to establish linearity and

reproducibility of the HPLC system. Graphs were constructed of the peak area

against drug concentration according to European Union criteria [17].

2.3.2. Chromatographic conditions

Separation of the analytes was achieved on the ODS hypersil column with

the detector set at 230 nm and the column maintained at 30 � 1 8C temperature.

The mobile phase was a mixture of 35:65 (v/v) acetonitrile–0.015 M disodium

hydrogen phosphate buffer and 0.1% TEA. The pH of mobile phase was

adjusted at 7.4 using phosphoric acid 85% without prior dilution and the flow

rate was maintained at 1.0 mL/min, resulting in a pressure of about 70 bar.

Mobile phase was degassed by vacuum through filtration, after mixing. The

same ratios of solvents in mobile phase were used to apply the samples in LC–

MS. Thus, a mixture of 35:65 (v/v) acetonitrile–0.02 M ammonium formate

with the addition of 0.05% formic acid was prepared using a flow rate of

0.3 mL/min in SIM positive mode. The [M+H]+ ions used to create the MS

spectra were m/z 461 for N-desmethylsildenafil and m/z 475 for sildenafil,

respectively. Daughter ion for sildenafil was further obtained by the fragmenta-

tion of m/z 475 (m/z 475! 461).

2.3.3. Sample extraction

In 10 mL conical glass tube, 0.5 mL of human blood and 2.0 mL of

saturated solution of disodium hydrogen phosphate buffer (pH 9.0) were added

and mixed for 10 s on a vortex. Each sample was extracted with 4.0 mL of

chloroform: isopropanol: n-heptane (25:10:65, v/v/v) on vortex for 2 min at

speed 4. The sample was centrifuged for 10 min at 3000 rpm. The upper

(organic) layer was then transferred into a 10 mL conical glass tube and

evaporated to dryness at room temperature under a gentle stream of nitrogen.

The residue was reconstituted in 100 mL of mobile phase and an aliquot of

50 mL was injected onto the HPLC system.

2.3.4. Precision and accuracy

Within-run and between-run precision and accuracy were determined by

extracting blood supplemented with sildenafil and N-desmethylsildenafil to 35,

200 and 450 ng/mL (n = 3). Between-run assays were performed in three

different days.

2.3.5. Recovery and stability

Extraction recovery and stability were calculated at two levels of spiked

blood samples (35 and 450 ng/mL). Extraction recovery was determined by

comparing the peak areas from extracted samples with those obtained from a

direct injection of the corresponding unextracted standards dissolved in mobile

phase (n = 3 in three different days). Stability was investigated also for stock

and working solution at 35 and 450 ng/mL.

2.3.6. Selectivity

Selectivity of the method was examined for 10 different blood sources and

eight other drugs (Table 4), including two more selective inhibitors of (cGMP)-

specific phosphodiesterase type V, tadalafil and vardenafil. All compounds were

assayed independently by injecting standard solutions at 250 ng/mL in the HPLC

system using the same chromatographic conditions as the proposed method.

2.3.7. Application in post-mortem blood samples

The method was further applied in two sudden-death cases of two middle-

aged married men who were found dead after an extramarital sexual intercourse,

and they were suspected due to some investigation findings of taking sildenafil

citrate tablets.

3. Results and discussion

3.1. Chromatographic conditions optimization

During the preliminary experiments, several combinations

of mobile phase composition and mobile phase pH were

investigated, in order to obtain the optimum separation of

sildenafil and N-desmethylsildenafil. Thus, the pH of mobile

phase at 3, 4, 4.5, 5, and 7.4, the acetonitrile percentage at 15,

25, 35, 45 and 55% and the buffer concentrations at 0.015,

0.05, 0.10, 0.15 and 0.2 M disodium hydrogen phosphate

buffer were examined, respectively. It was observed that pH

affected the retention times of the analytes and their

separation. Decreasing the pH value resulted in an over-

lapping between the analyte peaks (sildenafil, N-desmethyl-

sildenafil) and matrix interferences, which consequently led to

a separation failure. Although low pH values were sufficient

for the separation of sildenafil from endogenous compounds,

it seemed not to be for N-desmethylsildenafil. Therefore, pH

7.4 was chosen as the optimum value. In addition, as the

percentage of acetonitrile decreased, the retention times for

the drugs were longer and the separation of the analytes

were better until 35%. At higher values, retention time of

sildenafil was too short and metabolite’s peak overlapped

with endogenous compounds. Thus, 35% of acetonitrile was

chosen as the optimum percentage of organic modifier.

Disodium hydrogen phosphate buffer concentration in the

mobile phase also affected retention time of the analytes. It

was found that a 0.015 M buffer concentration gave the best

separation in the shortest time since at higher concentrations

of the buffer; the endogenous blood peaks, particularly for the

metabolite, affected the analyte peaks. Consequently, the best

separation of sildenafil and N-desmethylsildenafil on the ODS

column was achieved using a mobile phase of 35:65 (v/v)

acetonitrile–0.015 M Na2HPO4�2H2O, TEA 0.1% with pH

7.4, to obtain retention times of 19.23 and 10.85 min,

respectively (Fig. 3b).

3.2. Sample extraction optimization

For the extraction of the analytes, liquid–liquid extraction

was selected comparing to solid phase extraction in order to

allow the optimization of a simple and rapid method with less

cost that would be applicable in routine analysis. Thus, an

alkaline liquid–liquid extraction method by Sheu et al. [18]

was applied with no success. However the replacement of

sodium hydroxide with disodium hydrogen phosphate buffer

appeared to facilitate the separation of the drug from the

protein complex providing a significant response. The

extraction efficiency of four different solvents or solvent

mixtures [diethylether, ethylacetate, ether/dichloromethane

(3:2, v/v), chloroform/isopropanol/n-heptane (25:10:65, v/v/

Fig. 3. Representative HPLC–UV–vis chromatograms obtained from the analysis of (a) blank blood sample, (b) spiked sample at 50 ng/mL with sildenafil

(tr = 19.23 min) and N-desmethylsildenafil (tr = 10.85 min) and (c) post-mortem blood sample.

C. Pistos et al. / Forensic Science International 178 (2008) 192–198 195

v)] were further investigated. The first three solvents provided

high matrix noise that overlapped with analytes’s peaks. The

optimum solvent mixture was chloroform/isopropanol/n-

heptane (25:10:65, v/v/v) that yielded a satisfactory baseline

and higher sensitivity.

Retention times of N-desmethylsildenafil and sildenafil, were

10.85 and 19.23 min, respectively. Fig. 3a shows a chromato-

gram obtained from an extracted drug free blood, while Fig. 3b

shows a chromatogram of extracted drug free plasma supple-

mented with N-desmethylsildenafil and sildenafil at 50.0 ng/mL.

Table 2

Recovery data for sildenafil and N-desmethylsildenafil in human blood (n = 9)

Concentration added (ng/mL) % Recovery (�S.D.) (ng/mL) %R.S.D.

S

35 92.7 (�10.8) 11.6

450 92.6 (�1.8) 1.9

DS

35 86.4 (�2.6) 3.0

450 88.2 (�0.6) 0.6

Table 1

Intraday and interday accuracy and precision data for sildenafil (S) and N-desmethylsildenafil (DS) in spiked blood samples (3 days, n = 3)

Concentration added (ng/mL) S DS

Concentration found (ng/mL) %R.S.D. %Er Concentration found (ng/mL) %R.S.D. %Er

Intraday

35 35.0 � 1.7 4.8 0.1 31.9 � 0.4 1.1 �8.9

200 194.3 � 2.3 1.2 �2.9 178.6 � 2.6 1.5 �10.7

450 410.2 � 4.2 1.0 �8.8 502.8 � 5.0 1.0 11.7

Interday

35 35.2 � 1.8 5.1 0.5 30.9 � 0.4 1.1 �11.6

200 193.4 � 2.1 1.1 �3.3 177.9 � 2.6 1.5 �11.1

450 409.3 � 4.1 1.0 �9.1 503.8 � 5.0 1.0 11.9

C. Pistos et al. / Forensic Science International 178 (2008) 192–198196

3.3. Linearity

The peak areas were linearly related to blood concentrations

of sildenafil and N-desmethylsildenafil, respectively, from 10 to

at least 500 ng/mL. The average regression equation and slope of

five calibration curves of sildenafil in blood, prepared in five

different days over a period of 2 months is Ss = 0.0076

Cs + 0.0028, standard error (Sr) < 0.11, n = 5, r = 0.997. Ss,

corresponds to peak area of sildenafil while Cs, corresponds to

blood concentration of sildenafil (ng/mL). The correlation

coefficients square for each standard curve constructed

invariably exceeded 0.9991 and intercepts were all close to

zero and not statistically significant. %R.S.D. of the slopes

between the different calibration curves for sildenafil was 4.3.

The average regression equation and slope of five calibration

curves of N-desmethylsildenafil in blood, prepared in five

different days over a period of 2 months is SDS = 0.0081 �

Table 3

Stability data for sildenafil and N-desmethylsildenafil in human blood under vario

Storage conditions Time (%)Dev.a (35 ng/mL) R.S.D

S

Ambient temperature 6 h �7.3 6.1

6 8C 30 days �12.3 6.4

�20 8C 30 days �10.7 5.9

Freeze–thaw cycles; �20 8C 3 cycles �2.9 7.1

DS

Ambient temperature 6 h �3.7 6.4

6 8C 30 days �13.5 7.9

�20 8C 30 days �13.3 8.1

Freeze–thaw cycles; �20 8C 3 cycles �2.5 6.6

a Percentage of mean deviation from t = 0.b Percentage relative standard deviation, n = 3.

CDS � 0.0075, standard error (Sr) < 0.14, n = 5, r = 0.999. SDS,

corresponds to peak area of N-desmethylsildenafil and CDS,

corresponds to blood concentration of N-desmethylsildenafil

(ng/mL). The correlation coefficients for each standard curve

constructed invariably exceeded 0.9998 and intercepts were all

close to zero and not statistically significant. The lower limit of

detection was 5.0 ng/mL for sildenafil and N-desmethylsidenafil,

while the lower limit of quantification was 10.0 ng/mL for both

compounds based on S/N > 3 and S/N > 10, respectively.

%R.S.D. of the slopes between the different calibration curves

for N-desmethylsildenafil was 4.8.

3.4. Precision—accuracy

Interday precision and error in three different days were

found to be less than 5.1% and 9.1% for sildenafil, 1.5% and

11.9% for N-desmethylsildenafil, respectively (Table 1).

3.5. Recovery—stability

Extraction recovery and stability data, from blood samples

spiked with both analytes to 35 and 450 ng/mL are presented in

Tables 2 and 3, respectively.

3.6. Selectivity

Blank samples from 10 different sources and all eight drugs

did not yield any interference at the elution time for the

us storage conditions

. (%)b (35 ng/mL) (%)Dev.a (450 ng/mL) R.S.D. (%)b (450 ng/mL)

�3.9 8.4

�11.3 7.6

�9.4 5.3

�8.9 6.0

�4.9 5.4

�8.6 7.6

�7.2 7.5

�5.6 8.3

Table 4

Compounds studied for interferences (C = 250 ng/mL)

Compound Retention time (min)

Tadalafil 13.2

Vardenafil 22.1

Lorazepam 28.4

Temazepam 25.8

Diazepam 25.6

Caffeine 4.5

Barbital 8.4

Butobarbital 14.3

Fig. 4. Total ion current mass chromatogram obtained positive mode and

selected ion monitoring from the analysis of a post-mortem blood sample

(sildenafil, tr = 13.3 min, m/z 475, 461 and N-desmethylsildenafil, tr = 4.2 min,

m/z 461.

C. Pistos et al. / Forensic Science International 178 (2008) 192–198 197

detection of sildenafil and N-desmethylsildenafil. Table 4

presents the retention times of the studied drugs.

3.7. Application in post-mortem blood samples

No interfering peaks were observed in the two case samples

investigated, concerning sudden death of two middle-aged men.

Sildenafil and N-desmethylsildenafil blood concentrations were

50.1 and 72.0 ng/mL in the first case and 52.46 and 38.66 ng/

mL in the second case, respectively. Fig. 3c presents the

chromatogram of a post-mortem blood sample, concerning a

middle-aged man, analyzed for sildenafil and N-desmethylsil-

denafil. The presence of both analytes in these samples was

further confirmed using the same extraction method in LC–MS.

Fig. 4, shows a TIC chromatogram of the autopsy sample in

which N-desmethylsildenafil and sildenafil are eluted at 4.2 and

13.3 min, respectively. The comparison of the UV–vis and MS

chromatograms from the same autospy sample, illustrates the

reliability of UV–vis to detect both analytes but also its

disadvantage of low selectivity, a problem that mass spectro-

meter overcomes. MS chromatogram presents the absence of

any interference in the background of the baseline.

4. Conclusion

Most laboratories analyze sildenafil and its active metabolite

in human plasma and not in whole blood. The fact that the

identification of sildenafil and more particularly N-desmethyl-

sildenafil in blood using GC–MS is problematic has led the

analysts to the application of LC–MS or LC–MS/MS

instrumentation. However, the need to combine a single LC–

MS methodology with a supplementary technique is often. In

this case and when human blood is the only biological sample

available, or plasma and serum cannot be separated as in most

post-mortem cases, HPLC–UV–vis method is considered the

method of choice for the simultaneous determination of both

sildenafil and N-desmethylsildenafil. Therefore, a simple

method for the quantification of both compounds in human

blood is proposed that allows quantitating small amounts of the

analytes even after the intake of the lowest dose of 25 mg tablet

of sildenafil citrate using a high performance liquid chromato-

graphy UV–vis system.

Acknowledgment

We would like to thank Dr. Leandros Arvanitakis from Pfizer

Hellas A. E. for providing us the reference standards.

References

[1] I. Goldstein, T.F. Lue, H. Padma-Nathan, R.C. Rosen, W.D. Steers, P.A.

Wicker, Oral sildenafil in the treatment of erectile dysfunction. Sildenafil

Study Group, N. Engl. J. Med. 338 (1998) 1397–1404.

[2] R.B. Moreland, I. Goldstein, A. Traish, Sildenafil, a novel inhibitor of

phosphodiesterase type 5 in human corpus cavernosum smooth muscle

cells, Life Sci. 62 (1998) PL309–PL318.

[3] M. Boolell, M.J. Allen, S.A. Ballard, S. Gepi-Attee, G.J. Muirhead, A.M.

Naylor, I.H. Osterloh, C. Gingell, Sildenafil: an orally active type 5 cyclic

GMP-specific phosphodiesterase inhibitor for the treatment of penile

erectile dysfunction, Int. J. Impot. Res. 8 (1996) 47–52.

[4] C.G. Stief, S. Uckert, A.J. Becker, M.C. Truss, U. Jonas, The effect of the

specific phosphodiesterase (PDE) inhibitors on human and rabbit caver-

nous tissue in vitro and in vivo, J. Urol. 159 (1998) 1390–1393.

[5] Viagra/Clinical Pharmacology/Pharmacokinetics and Metabolism, http://

www.rxlist.com/cgi/generic/viagra.htm, May 2007.

[6] M. Lee, D. Min, Determination of sildenafil citrate in plasma by high-

performance liquid chromatography and a case for the potential interac-

tion of grapefruit juice with sildenafil citrate, Ther. Drug Monit. 23 (1)

(2001) 21–26.

[7] M. Thau Sheu, A.B. Wu, Y.G. Cheng, A. Hsia, H. Ho, Development of a

liquid chromatographic method for bioanalytical applications with silde-

nafil, J. Chromatogr. B 791 (2003) 255–262.

[8] Y. Wang, J. Wang, Y. Cui, J.P. Fawcett, J. Gu, Liquid chromatographic-

tandem mass spectrometric method for the quantitation of sildenafil in

human plasma, J. Chromatogr. B 828 (2005) 118–121.

[9] A. Tracqui, B. Ludes, HPLC-MS for the determination of sildenafil citrate

(Viagra) in biological fluids. Application to the salivary excretion of

sildenafil after oral intake, J. Anal. Toxicol. 27 (2003) 88–94.

[10] J.R. Lewis, R.D. Johnson, L.C. Blank, Quantitative determination of silde-

nafil (Viagra) and its metabolite (UK-103,320) in fluid and tissue specimens

obtained from six aviation fatalities, J. Anal. Tox. 30 (2006) 14–20.

[11] W. Weinmann, M. Bohnert, A. Wiedemann, M. Renz, N. Lehmann, S.

Pollak, Post-mortem detection and identification of sildenafil (Viagra) and

its metabolites by LC/MS and LC/MS/MS, Int. J. Legal Med. 114 (2001)

252–258.

[12] M. Al-Ghazawi, M. Tutunji, S. Aburuz, Simultaneous determination of

sildenafil and N-desmethyl sildenafil in human plasma by high-perfor-

mance liquid chromatography method using electrochemical detection

with application to a pharmacokinetic study, J. Pharm. Biomed. Anal. 43

(2007) 613–618.

[13] J.D. Cooper, D.C. Muirhead, J.E. Taylor, P.R. Baker, Development of an

assay for the simultaneous determination of sildenafil (Viagra) and its

metabolite (UK-103,320) using automated sequential trace enrichment of

dialysates and high-performance liquid chromatography, J. Chromatogr. B

701 (1997) 87–95.

C. Pistos et al. / Forensic Science International 178 (2008) 192–198198

[14] K. Saisho, K.S. Scott, S. Morimoto, Y. Nakahara, Hair analysis for

pharmaceutical drugs. II. Effective extraction and determination of silde-

nafil (Viagra) and its N-desmethyl metabolite in rat and human hair by

GC–MS, Biol. Pharm. Bull. 24 (12) (2001) 1384–1388.

[15] S. Pagani, D. Mirtella, R. Mencarelli, D. Rodriquez, M. Cingolani,

Postmortem distribution of sildenafil in histological material, J. Anal.

Toxicol. 29 (2005) 254–257.

[16] P. Van Hee, H. Neels, W. Lambert, V. Coucke, M. Dedoncker, Comment

on: J. Anal. Toxicol. 29 (2005) 254–257: Postmortem distribution of

sildenafil in histological material, J. Anal. Toxicol. 30 (2006) 403–404;

author reply 404–405.

[17] Official Journal of the European Communities L221, 8-36. Commission

decision (2002/657/EC) of 12 August 2002 concerning the performance of

analytical methods and the interpretation of results. Brussels, Belgium,

2002.

[18] M.T. Sheu, A.B. Wu, G.C. Yeh, A. Hsia, H. Ho, Development of a liquid

chromatographic method for bioanalytical applications with sildenafil, J.

Chromatogr. B 791 (2003) 255–262.