The Postmortem Distribution of Vardenafil (Levitra

Journal of Analytical Toxicology, Vol. 31, July/August 2007

The Postmortem Distribution of Vardenafil (Levitra| in an Aviation Accident Victim with an Unusually High Blood Concentration*

Robert D. Johnson ~, Russell J. Lewis, and Mike K. Angler Downloaded from https://academic.oup.com/jat/article/31/6/328/682815 by guest on 27 September 2021 Civil Aerospace Medical Institute, Federal Aviation Administration, Analytical Toxicology and Accident Research Laboratory, AAM-610, CAMI Building, 6500 S. MacArthur Blvd., Oklahoma City, Oklahoma 73169-6901

I Abstract phodiesterase type 5 enzyme (PDE5) found predominantly in the penile corpus cavernosum (2-7). Vardenafil (tevitra) is one of the most widely prescribed Vardenafil undergoes hepatic metabolism, producing the treatments for erectile dysfunction. This report presents a active desethyl metabolite M1. M1 contributes to the ob- rapid and reliable method for the identification and quantification served pharmacological effects provided by vardenafil, as M1 of vardenafil in postmortem fluids and tissues, applies this method exhibits approximately 30% of the potency of the parent to a postmortem case, and describes the distribution of vardenafil in various fluids and tissues.This procedure utilizes sildenafil-d8, drug (1). Under steady-state conditions, the plasma concen- which is structurally closely related to vardenafil, as an tration of M1 is approximately 26% of that seen for vardenafil internal standard for more accurate and reliable quantitation. (1). After oral administration of vardenafil, peak plasma con- The method incorporates solid-phaseextraction and liquid centrations are obtained within 30-60 min (1). Vardenafil chromatography-tandem mass spectrometry (MS) and and its active metabolite have a terminal half-life of approx- MS-MS-MS utilizing an atmospheric pressure chemical imately 4-5 h (1). ionization ion trap MS in the positive chemical ionization mode. Vardenafi], though relatively safe, has certain side effects Solid-phase extraction proved to be exceptionally efficient that could create aviation safety-related hazards. Like its struc- providing recoveries that ranged from 94% to 97%. The limit tural relative sildenafil, vardenafil inhibits PDE5, and it also has of detection for vardenafil was determined to be 0.19 ng/mL. a high affinity for phosphodiesterase type 6 (PDE6), a retinal The linear dynamic range for this compound was 0.39-200 ng/mL. enzyme involved in phototransduction (1,8). The inhibition This method was successfullyapplied to postmortem fluid and tissue specimens obtained from an aviation accident victim. of PDE6 can result in a condition known as "blue tinge", the in- The distribution of vardenafil in various fluids and tissuesand ability to discriminate between blue and green colors (9). This the unusually high concentration of vardenafil in the victim's "blue tinge" impairment could hinder a pilot relying upon in- blood are examined. struments during adverse meteorological conditions and/or night flights (10). Additionally, vardenafil has been shown to potentiate the hypotensive effects of nitrates commonly employed in the treatment of certain heart conditions (11). Introduction Other work that utilizes liquid chromatography-electrospray ionization mass spectrometry (LC-ESI-MS) for the iden- Vardenafil (Levitra), prescribed as an oral treatment for erec- tification of vardenafil has been published (12,13). However, the tile dysfunction (ED), was introduced in the United States in authors did not find any atmospheric pressure chemical ion- 2003. Within a year of its introduction, vardenafil prescriptions ization (APCI) methods referenced or any other analytical work increased approximately 400%, and it is now one of the most done in postmortem specimens. Because of its increasing pop- widely prescribed treatments for ED (1). Vardenafil, 1-{[3-(1,4- ularity, the presence of vardenafil in aviation accident victims dihydro-5-methyl-4-oxo-7-propylimidazo[5,1-f][1,2,4]triazin- will become more common. This paper describes one such 2-yl)-4-ethoxyphenyl]sulfonyl/-4-ethyl-piperazine, shown in aviation accident victim. In this case, high concentrations of Figure 1, is a selective inhibitor of the cGMP-specific phos- vardenafil were found in various tissues and fluids obtained from the victim. Solid-phase extraction (SPE) and LC with

* This research is part of an Office of Aerospace Medicine Research Report. APCI ion trap (IT) MS were used to quantify vardenafil in each ~ Aulhor to whom correspondence should be addressed. E-mail: r specimen examined.

328 Reproduction(photocopying) of editorial contentof this journal is prohibitedwithout publisher'spermission. Journal of Analytical Toxicology, Vol. 31, July/August 2007

stored in 50:50, acetonitrile/H20. Initially, A B precursor ions were identified for both o 0 D D\ ID compounds. Following [M+H]§ ion identification, ionization conditions were op- N. II I \\ ~N~ /~--D timized by infusing each analyte directly into the mobile phase, which was then introduced into the MS at a flow rate of ) ) 1.00 mL/min. Tuning the MS for the de- sired ions was then accomplished using Vardenafil Sildenafil-d8 the autotune feature of the Xcalibur TM software. Each sample analysis consisted Figure 1. Chemical structures of vardenafil (A)and si Idenafil-d 8 (B). of one data collection segment. This segment collected data for both vardenafil

and the internal standard sildenafil-ds. Downloaded from https://academic.oup.com/jat/article/31/6/328/682815 by guest on 27 September 2021 Materials and Methods The operating conditions for the data collection segment were as follows: APCI capillary temperature, 175~ APCI va- Chemicals and solutions porizer temperature, 450~ source voltage, 4.00 kV; source All aqueous solutions were prepared using double deion- current, 4.00 pA; sheath gas flow (nitrogen), 80.0; auxiliary gas ized water (DDW), which was obtained using a Milli-QTplus flow (nitrogen), 20.0; capillary voltage, 23.0 V; tube lens offset, Ultra-Pure Reagent Water System (Millipore| Continental 35.0 V; octapole I offset, -1.75 V; octapole 2 offset, -6.00 V; in- Water Systems, El Paso, TX). All chemicals were purchased in teroctapole lens voltage,-18.00 V; trap DC offset voltage,-15 the highest possible purity and used without any further pu- V; multiplier voltage, 0.0 V; and 1 microscan having a max- rification. All solvents were of HPLC-grade and were obtained imum ion injection time of 100 ms. This segment was further from Fisher Scientific (Fischer Scientific, Fair Lawn, NJ). split into three separate scan events. Scan event 1 collected the Formic acid (97%) was purchased from ICN Biomedicais sildenafil-d8 and vardenafil precursor, [M+H]§ ions at m/z (Irvine, CA). Vardenafil was obtained from Bayer Pharmaceu- 483.3 and m/z 489.4, respectively. Scan event 2 collected the tical Corporation (West Haven, CT). Sildenafil-d8was synthe- vardenafil product ions at m/z 376.0 and 376.1 following col- sized by and obtained from SynFine Research (Richmond Hill, lision-induced dissociation (CID) of the precursor ion (m/z ON, Canada). 489.4) using a collision energy of 46%. Scan event 3 collected Stock standard solutions of vardenafil were prepared at a the vardenafil MS3 product ion at m/z 284.1 following CID of concentration of 1 mg/mL in methanol. A stock solution of the the MS-MS product ion (m/z 376.0) using a collision energy internal standard, sildenafil-ds, was prepared at 100 ~g/mL in of 42%. methanol. Formic acid (50raM) constituted the aqueous por- Ions having the highest abundance in the MS-MS mode tion of the high-performance liquid chromatography (HPLC) were used for quantitation of vardenafil. The MS-MS spectra mobile phase and was adjusted to pH 5.00 with concentrated of vardenafil provided two predominant ions. These ions, at ammonium hydroxide. The formic acid buffer was mixed with m/z 376.0 and m/z 377.1, were summed for the quantitation acetonitrile in a 98:2 (v/v) ratio to help prevent the growth of of this compound. The molecular ion of the internal stan- microbes. This mixture was filtered through a vacuum fil- dard, sildenafil-ds, at m/z 483.3 was utilized for quantitation. tering apparatus that incorporated a 0.45-~m GH polypro 47- The MS-MS-MS full spectrum was used for vardenafil confir- mm hydrophilic, polypropylene membrane filter obtained from mation. Pall Gelman Laboratory (East Hills, NY). The primary organic component of the mobile phase was HPLC-grade acetonitrile, Calibrator and control preparation which was filtered prior to use through a vacuum filter appa- Calibration curves were prepared by serial dilution utilizing ratus that incorporated the same type of membrane filter de- bovine whole blood as the diluent. Calibrators were prepared scribed. from one original stock standard solution of vardenafil. Con- trois were prepared in a similar manner as calibrators, using Instrumentation and LC-MS method the same bovine whole blood as diluent but employing a second The instrumentation and method development procedures original stock solution. Calibration curves were routinely pre- employed for this study were identical to those used in other pared at a concentration range of 0.39-200 ng/mL. A min- recently published work from our laboratory (14). Briefly, how- imum of seven calibrators was used to construct each calibra- ever, identification and quantitation were accomplished using tion curve. Controls used for the determination of accuracy, a Thermo Finnigan model LCQ APCI-IT-MS(Thermo Finnigan precision, and compound stability were prepared at 2, 20, and Corp., San Jose, CA). For all determinations, the HPLC was op- 200 ng/mL. Controls were prepared in pools large enough to erated in an isocratic mode with a flow rate of 1.00 mL/min. provide samples for the entire study. A sildenafil-d8 working The mobile phase ratio employed was 70:30 (acetonitrile/ standard was prepared at a final concentration of 50 ng/mL by buffer). The sample injection volume was held constant at 10 dilution with DDW from the stock solution. pL. The HPLC column was routinely equilibrated overnight Quantitation of vardenafil in biological specimens prior to use. Following use, the column was washed with and was achieved via an internal standard calibration procedure.

329 Journal of Analytical Toxicology, Vol. 31, July/August 2007

Response factors for vardenafil were determined for each and subsequent fragmentation of these precursor ions. The sample. The response factor was calculated by dividing the vardenafil precursor ion used to create the MS-MS spectra area of the analyte peak by the area of the internal standard was found at m/z 489.4. MS3 spectra were then obtained by col- peak. Calibration curves were prepared by plotting a linear lection and subsequent fragmentation of the vardenafilproduct regression of the analyte/internal standard response factor ion at m/z 376.0. The full scan MS-MS and MS-MS--MSspectra versus the analyte concentration for the calibrators and were for vardenafil provided "fingerprints" used for analyte identifi- used to determine the concentrations of vardenafil in controls cation and confirmation. and specimens. Both vardenafil and sildenafil-d8 were eluted from the analytical column in less than 4 rain. Figure 2 shows a represen- Sample preparation and extraction procedure tative LC-MS chromatogram. LC retention times were used as The extraction procedure used for this study was previously one analyte acceptability criteria. Vardenafil retention times ob- developed for the analysis of sildenafil (14). Briefly,tissue spec- tained from postmortem specimens were required to be within imens were homogenized following a 2:1 dilution with 1.00% • 2.0% of the average calibrator retention time. Typical reten- sodium fluoride (2 g 1.00% NaF/1 g wet tissue). Three-milliliter tion times were 2.41 and 2.54 rain for vardenafil and sildenafil- Downloaded from https://academic.oup.com/jat/article/31/6/328/682815 by guest on 27 September 2021 aliquots of calibrators, controls, and postmortem fluids and 3- d8, respectively.Matrix interference was monitored by the use g aliquots of tissue homogenate (1 g tissue) were transferred to of whole-blood negative controls that were spiked only with individual 15-mL screw-top vials. To each sample, 50 ng of in- sildenafil-d8 prior to extraction. ternal standard was added (1.00 mL of the 50 ng/mL stock so- Quantitation was accomplished by collecting the product lution). Nine milliliters ice-cold acetonitrile was added to each ion with the highest abundance. Vardenafil yielded two pre- sample. Centrifugation at 820 • g for 5 min provided removal dominant ions with nearly equal abundance in the MS-MS of cellular debris and proteins. Following centrifugation, the mode, the abundance obtained for each of these ions was supernatant was evaporated under a stream of dry nitrogen to summed for quantitative purposes. The total area of the a volume less than 1 mL. Followingacetonitrile concentration, two predominant product ions for vardenafil divided by the 4.00 mL 0.10M phosphate buffer (pH 6.00) was added to each area of the precursor ion of the internal standard, sildenafil- sample. The extracts were transferred to SPE columns, which ds, resulted in a response factor that was used for quantitation. had been pre-conditioned with 2.00 mL methanol, followedby The linear dynamic range (LDR), limit of detection (LOD), 2.00 mL 0.10M phosphate buffer (pH 6.00). The Bond Elute and limit of quantitation (LOQ) were determined by analysis Certify| SPE columns, obtained from Varian (Harbor City, CA) of a calibration curve that contained calibrators ranging contained 130 mg of sorbent material and were 10 mL in in concentration from 0.10 to 6400 ng/mL. The LDR for volume. Once each sample had passed through its respective vardenafil was determined following this analysis and found column, the columns were washed with 1.00 mL of 1.0M acetic to be 0.39-200 ng/mL. The correlation coefficient for this acid and then 6.00 mL methanol with drying steps placed be- calibration curve exceeded 0.992 when a weighting factor of tween each wash cycle. The analytes were eluted from the 1/x was used. The LOD and LOQ determined for vardenafil columns with 3.00 mL of 2% ammonium hydroxidein ethyl ac- etate, evaporated to dryness, and reconstituted in 50.0 BL ace- 2.54 tonitrile for analysis. ~',, Sildenafil-ds MS 4. m z 483.3 Extraction efficiency 2o .... 1 i L The extraction efficiencyfor vardenafilwas determined using 2.41 a procedure commonly employed in our laboratory (15). u ~. Vardenafil MS Two control groups, X and Y, prepared using negative whole 4~ m z 4894 blood diluent were extracted in the same manner as described 2o in the previous section. Group X was spiked with a precisely 241 known amount of vardenafil prior to extraction, and group s,, Vardenafil MS-MS .~ m z 3760. 377.1 Y was spiked with the same precisely known amount of -" , i vardenafil following SPE. Upon analysis, the average response 241 factor obtained from group X was divided by the average so Vardenafil MS-MS-MS 6O response factor obtained from group Y to yield the percent 4,, m z 284 I recovery value. 2cP 00 o 5 Io I $ 20 2. 31~ ': 40 Time (rain) Figure 2. A representativechromatogram of sildenafil-d8 and vardenafil Results and Discussion obtained from an extracted whole blood calibrator. Chromatographic peaks represent ions monitored in SIM mode for each compound as follows: sildenafil-d8precursor ion at m/z483.3 and vardenafil precursor ion Method validation at m/z 489.4; MS-MS ion at m/z 376.0 + 377.1; and MS-MS-MS ion at By incorporating a "soft" ionization technique MS spectra m/z 284.1. Peaks were obtained from a 10-pL injection of an extracted were produced consisting predominantly of the protonated 6.25 ng/mL calibrator. [M+H]§ ion. Ion trap technology allowed for the collection

330 Journal of Analytical Toxicology, Vol. 31, July/August 2007

Table I. LDR, LOD, LOQ, and Recovery Data for Vardenafil

Extraction Efficiency (%) • SD* LDR LOD LOQ Compound (ng/mL) (ng/mt) (ng/mt) 2 ng/mL 20 ng/mt 200 ng/mL

Vardenafil 0.39-200 0.19 0.39 95+7 98+6 94+4

* n = 5 at each concentration.

Table II. Intra- and Interday Accuracy and Precision for Intraday (within-day) and interday (between*day)accuracy Repeated Determinations over Seven Days* and precision were examined for this extraction as our laboratory policy dictates. Accuracy was measured as the percent

Vardenafil relative error between the experimentally determined and pre- Downloaded from https://academic.oup.com/jat/article/31/6/328/682815 by guest on 27 September 2021 pared concentrations of a sample. Precision was measured as Target Conc. Mean • SD Relative the relative standard deviation (RSD) for the experimentallyde- (ng/mL) (ng/mL) Error R.S.D. termined concentrations. Accuracy and precision studies were Day I performed using whole blood controls at concentrations of 2, 2 2.05 _+ 0.02 +3% I% 20, and 200 ng/mL. These values were chosen because they are 20 20.4 _+ 0.2 +I % 2% distributed throughout the extensive LDR of this compound. 200 199 • 2 -I % 1% These controls were prepared in 500-mL quantities on Day 1 of the experiment and stored at 4~ until extracted. Day2 For intraday analyses, a calibration curve was extracted 2 1.97 + 0.05 -2% 3% along with five replicates of each control concentration on 20 20.2 • 0.7 +1% 4% Day I of the experiment. Intraday relative errors in the 2, 20, 200 194 • 5 -3% 2% and 200 ng/mL control groups were ,~ 3% for this analyte. Day 4 Furthermore, the intraday RSD was ~ 2% at each vardenafil 2 2,03 • 0.03 +2% 2% control concentration. Intraday results are summarized in 20 20.8 + 0.3 +4% 1% Table II. 200 185 • 7 -8% 4% Interday accuracy and precision were evaluated by ex- Day7 tracting five replicates of each of the three control concen- 2 2.02 • 0.04 +1% 2% trations on Days 2, 4, and 7 and generating quantitative 20 19.7 • 0.2 -2% I% values by utilizing the calibration curve originally prepared 200 183 _+ 7 -9% 4% on Day 1. The interday relative errors for this analyte at each control concentration did not exceed 9%. The RSD for the 2 * n = 5 at each concentration for each day; controls were run on Days I, 2, 4, and 7. ng/mL vardenafil control was ,~ 3% over Days 2, 4, and 7. Both the 20 and 200 ng/mL controls had RSD values ,~ 4% over the same time period. These interday results show that are listed in Table I. Our laboratory defines LOD as the lowest this method is both accurate and precise over a seven-day pe- concentration of analyte having a minimum signal-to-noise riod (Table II). ratio (S/N) of 5, in addition to meeting the MS-MS and The stability of vardenafil in whole blood was evaluated by MS-MS-MS spectral "fingerprint" identification and reten- evaluating the control concentrations obtained on Day 7 of the tion time criteria. The LOQ was defined as meeting all LOD interday experiment (Table II). Vardenafil showed no apparent criteria, plus having a S/N of 10, and an experimentally decrease in concentration after one week of storage at 4~ at determined value within + 20% of its prepared concentra- concentrations of 2 and 20 ng/mL. The 200 ng/mL control tion. The LOD and LOQ for vardenafil when extracted from showed a slight decrease in concentration over this time pe- whole blood were determined to be 0.19 ng/mL and 0.39 riod. These results demonstrate that whole blood specimens ng/mL, respectively. may be stable for seven days when stored at 4~ However, as Instrumental carryover from one sample to the next was good laboratory practice and to ensure the highest quality an- not an issue. An acetonitrile blank, injected following the alytical data, we recommend that biological specimens always highest calibrator, showed no carryover contamination. Sub- be analyzed promptly after thawing. sequently, blanks were used following each postmortem specimen in the sample sequence to verify that no sample-to- Postmortem specimen analysis sample contamination occurred. In fatal aviation accidents, specimens from accident vic- The extraction efficiency for vardenafil at various concen- tims are sent to the Federal Aviation Administration's Civil trations obtained from this SPE procedure was exceptional.As Aerospace Medical Institute for toxicological analysis. Post- can be seen in Table I, the average recoveryof vardenafil ranged mortem fluid and tissue samples obtained from one such from 94 to 98% over a wide range of concentrations. victim were examined for vardenafil utilizing the described

331 Journal of Analytical Toxicology, Vol. 31, July/August 2007 method. This case was specifically selected because evi- Conclusions dence obtained from the scene of the accident suggested that the accident victim had been prescribed Levitra. All The use of vardenafil for the treatment of erectile dysfunc- available specimens were examined, including heart blood, tion has increased dramatically over the past two years. Thus, bile, liver, kidney, heart muscle, lung, and skeletal muscle. the possible occurrence of undesirable side effects is of con- The quantitative results of this analysis are presented in cern in the aviation safety community. With this in mind, a Table III. method for the identification and quantitation of vardenafil As previously stated, sildenafil-d8 was used as the internal has been developed that is rapid, reliable, and sensitive. By standard in this study as an alternative to deuterated varde- utilizing SPE, a clean extract was achieved with minimal sol- nafil which is not commercially available. These compounds vent use. Additionally, the method provided excellent extrac- are closely related structurally; however, the interpretation tion efficiency. The postmortem distribution case described of quantitative data for vardenafil obtained from specimen herein involves a seemingly high concentration of vardenafil types other than blood should be closely scrutinized because in the victim's blood. However, the lack of published data on of possible variations in extraction efficiency between spec- the distribution of vardenafil in postmortem cases provides no Downloaded from https://academic.oup.com/jat/article/31/6/328/682815 by guest on 27 September 2021 imen types. insight into whether this concentration could be considered Each specimen type analyzed in this case was found posi- within a normal range. tive for vardenafil. The quantitative values obtained for each specimen type can be seen in Table III. The highest concentration of vardenafil was found in bile. This was an expected result, as the major excretion route for this analyte is in the References feces (1). Following bile, the highest concentrations of vardenafil were found in the blood, lung, liver, heart, kidney, and 1. U.S. prescribing information. Levitra: compound data sheet, skeletal muscle. Although little to no data exist for tissue Bayer, 2002. concentrations of vardenafil, antemortem plasma values have 2. E.P. Krenzelok. Sildenafil: clinical toxicology profile. J. Toxicol. been reported. Peak plasma concentrations of vardenafil Clin. Toxicol. 38" 645-651 (2000). 3. R.B. Moreland, I.I. Goldstein, N.N. Kim, and A. Traish. Sildenafil range from 10 to 26 ng/mL following the maximum daily citrate, a selective phosphodiesterase type 5 inhibitor. Trends recommended dose of 20 mg (1). The blood value deter- Endocrinol. Metab. 10" 97-104 (1999). mined in the case presented here is significantly higher than 4. G. Altiokka, Z. Atkosar, E. Sener, and M. Tuncel. FIA of sildenafil what is expected based on the reported concentrations ob- citrate using UV-detection. J. Pharm. Biomed. Anal. 25: 339- served following maximum daily recommended dosage. An 342 (2001). 5. C. de Mey. Oppurtunities for the treatment of erectile dysfunction exceptionally high blood concentration could indicate one or by modulation of the NO axis--alternatives to sildenafil citrate. more of the following: an inappropriately high dosage, Curr. Med. Res. Opin. 14:187-202 (1998). metabolism deficiency, hepatic impairment, or possible drug 6. F. Giuliano, C. Donatucci, F. Montorsi, S. Auerbach, G. Karlin, interaction with erythromycin or kctoconazole (16). Another C. Norenberg, M. Homering, T. Segerson, and I. Eardley. Varde- possibility tbr the seemingly high blood value could be post- nafil is effective and well-tolerated for treating erectile dysfunction in a broad population of men, irrespective of age. BJU Int. 95: mortem redistribution. The volume of distribution for varde- 1] 0-1 ] 6 (2005). nafil is high at 208 L/kg. This high distribution value suggests 7. J.S. Kalsi, G. Bahadur, A. Muneer, O. Ozturk, N. Christopher, the possibility that this compound could easily redistribute D.J. Ralph, and S. Minhas. Novel PDE5 inhibitors for the treatment between fluids and tissues with high water content after death. of male erectile dysfunction. Reprod. Biomed. Online 7:456--461 An accurate interpretation of the high blood concentration (2003). 8. E. Bischoff. Potency, selectivity, and consequences of nonselec- found in this case is difficult because of the lack of pub- tivity of PDE inhibition. Int. J. Impot. Res. 16 Suppl 1:$11-$14 lished scientific research on this subject and, thus, its sig- (2004). nificance is not known. 9. Sildenafil (Viagra). TOXI-NEWS 18:1 (1999). 10. D.J. Borrillo. Dangers of Viagra use in pilots. FederalAir Surgeons Medical Bulletin 98:9-10 (1998). Table III. Concentrations of Vardenafil Found in the 11. E. Bischoff. Vardenafil preclinical trial data: potency, pharmaco- Victim of a Fatal Aviation Accident dynamics, pharmacokinetics, and adverse events. Int. J. Impot. Res. 16 Suppl 1:$34-$37 (2004). Specimen Vardenafil 12. X. Zhu, S. Xiao, B. Chen, F. Zhang, S. Yao, Z. Wan, D. Yang, and H. Han. Simultaneous determination of sildenafil, vardenafil and tadalafil as forbidden components in natural dietary supplements Heart blood 291 ng/mL for male sexual potency by high-performance liquid chromatog- Bile 1665 ng/mL raphy-electrospray ionization mass spectrometry. J. Chromatogr. Liver 86 ng/g A 1066:89-95 (2005). 13. S.R. Gratz, C.L. Flurer, and K.A. Wolnik. Analysis of undeclared Kidney 15 ng/g synthetic phosphodiesterase-5 inhibitors in dietary supplements Skeletal muscle 8 ng/g and herbal matrices by LC-ESI-MS and LC-UV. J. Pharm. Biomed. Heart muscle 26 ng/g Anal. 36" 525-533 (2004). 14. R.J. Lewis, R.D. Johnson, and C.L. Blank. Quantitative determi- Lung 234 ng/g nation of sildenafil (Viagra)and its metabolite (UK-103,320)in fluid and tissue specimens obtained from six aviation fatalities.

332 Journal of Analytical Toxicology,Vol. 31, July/August2007

J. Anal. Toxicol. 30:14-20 (2006). 16. http://www.Levitra.Com/hcp/clinical_information/vardenafil_ 15. R.D.Johnson, R.J. Lewis, D.V. Canfield, and C.L. Blank. Accurate safety.Htm, December 2005. assignment of ethanol origin in postmortem urine: liquid chromatographic-mass spectrometric determination of serotonin Manuscript received March 1,2007; metabolites. J. Chromatogr. B 805:223-234 (2004). revision received April 16, 2007. Downloaded from https://academic.oup.com/jat/article/31/6/328/682815 by guest on 27 September 2021

333