Journal of Analytical Toxicology 2015;39:572 576

doi:10.1093/jat/bkv071 Advance Access publication July 2, 2015 Case Report

A Non-Fatal Self-Poisoning Attempt with Sildenafil

Veerle Matheeussen1*, Kristof E. Maudens2, Kurt Anseeuw3 and Hugo Neels1,2
1
Laboratory of Toxicology, Ziekenhuis Netwerk Antwerpen, Lange Beeldekensstraat 267, 2060 Antwerp, Belgium, 2Toxicological
Centre, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk, Belgium, and 3Emergency Department, Ziekenhuis Netwerk
Antwerpen, Lange Beeldekensstraat 267, 2060 Antwerp, Belgium

*Author to whom correspondence should be addressed. Email: [email protected]

The phosphodiesterase type 5 inhibitor, sildenafil, is not generally dose-related increase in the frequency of visual adverse events,
known for its use as a self-poisoning drug. However, intoxication but no clear relationship between dose and maximum decreases
cases with lethal outcome have been described. The case presented in blood pressure and no clinically signicant changes in electro-
here is of a 56-year-old man who claimed to have undertaken an cardiogram. Due to the small number of cases, a relationship
unsuccessful suicide attempt by mono-ingestion of 65 tablets of between sildenal overdose and increased cardiac risk could
100 mg sildenafil. He arrived at the emergency department 24 h not be excluded. The highest reported single dose in their post-
after intake with severe vomiting and symptoms of blurred vision. marketing survey was of a 33-year-old man who took 24 tablets of
Clinical examination revealed no priapism. Of note was a sinus tachy- 100 mg sildenal. He was diagnosed with annular scotoma,
cardia of 100 bpm without signs of hypotension. To quantify the defective color vision, vascular retinal dilatation, visual eld
sildenafil concentration in serum, an high-performance liquid chroma- defect and papilledema. He recovered from all except for visual
tography photo-diode array method was developed and validated eld defect and annular scotoma (4). Although experience with
according to European Medicines Agency guidelines. The intoxicated sildenal overdose is limited, a few cases (6, 8 13) are reported
patient had a serum concentration of 22.2 mg/mL sildenafil, at the in the literature including two with a fatal outcome (Table I).
time of presentation, which is far above the therapeutic peak concen- One due to an omental varix rupture in a 41-year-old man with
tration. The serum concentration further declined to 9.2 and advanced alcoholic cirrhosis (9) and another due to sudden
2.3 mg/mL, respectively, 5 and 14 h later, revealing a biological cardiac death in a 56-year-old man, with pre-existing risk factors
half-life of 4.2 h. To the best of our knowledge, this patient took for coronary artery disease such as hypertension, diabetes
the highest sildenafil dose, which resulted in the highest serum con- mellitus and excessive alcohol consumption. Furthermore, he
centration, ever reported. In this subject, sildenafil showed good received a multidrug treatment for essential hypertension
tolerability because few symptoms occurred and only moderate sup- and an anxio-depressive disorder (6). The symptoms in the
portive therapy was needed for full recovery without sequelae. other non-fatal overdose cases included optic neuropathy and
cilioretinal artery obstruction resulting in complete blindness
(8), prolonged priapism (10, 12), mild tachycardia (13) and rhab-
domyolysis (11). Aortic and vertebral artery dissection have been
associated with chronic sildenal abuse (14, 15).
Introduction
Sildenal or 1-[[3-(6,7-dihydro-1-methyl-7-oxo-3-propyl-1H-
pyrazolo[4,3-d]pyrimidin-5-yl)-4-ethoxyphenyl]sulfonyl]-4-met- Case history
hylpiperazine was originally investigated as an antihypertensive A 56-year-old man contacted the emergency services after an
agent and is used for the treatment of pulmonary hypertension attempted suicide with 65 tablets of 100 mg sildenal citrate
under the trade name Revatiow (Pzer) (1), although its use (Viagra or Kamagra, the latter being available for online purchase
as Viagraw in the setting of erectile dysfunction is generally without prescription) 24 h earlier. He experienced severe
better known (2). Its mechanism of action is the inhibition of vomiting and reported symptoms of blurred vision and difcul-
phosphodiesterase type 5 (PDE5), which results in an increase ties in distinguishing facial expressions. The subjective visual
in nitric oxide-stimulated release of cyclic guanosine monophos- perception included a dark view with occasional light ashes.
phate and subsequent smooth muscle relaxation and vasodila- Clinical examination revealed a pulse of 100 beats/min, a blood
tion. Inhibition of PDE5 in the corpus cavernosum smooth pressure of 125/70 mm Hg and normal body temperature. Sinus
muscle facilitates erectile response to sexual stimulation (3). tachycardia of 118/min was found on electrocardiogram and
Sildenal has a good safety prole in the registered once daily head computed tomography showed dilatation of the ventricular
doses of 25 100 mg (4). Adverse effects associated with silden- system and subarachnoid spaces. Lab results were only minimally
al therapy are transient, moderate and dose-dependent effects abnormal with a slightly diminished kidney function (glomerular
linked to vasodilation, including headache, facial ushing and itration rate: 53.8 mL/min/1.73 m2) and an increase in inam-
nasal congestion (4 7). Occasional visual disturbances like blu- matory parameters (elevated C-reactive protein: 10.6 mg/L
ish haze or blurred vision are explained by its action on PDE6 (normal: ,5 mg/L) and leucocytosis (13.9 109/L, reference
which is present in the retina (3, 6). An overdose of sildenal range: 3.45 9.76 109/L)). A transient increase in troponin I
may theoretically result in hypotension, tachycardia and cardiac was also noted, with no clinical signs or echocardiographic
arrest (5). However, a post-marketing safety analysis of overdose abnormalities. Supportive care was given and an high-performance
cases revealed that exceeding the maximum recommended daily liquid chromatography photo-diode array method was optimized
dose or dose frequency only results in an increased occurrence and validated in order to quantify the sildenal serum
and severity of the known adverse drug reactions, including a concentration.
# The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

Downloaded from https://academic.oup.com/jat/article-abstract/39/7/572/819563

by guest
on 11 November 2017
 Table I
Sildenafil Intoxication Reports Ranked by Increasing Dose

Dose Age (years) Gender Symptoms Lethal Serum concentration Reference

(M/F)
Unknown 41 M Variceal rupture Yes Not reported (9)
75 mg (5 mg/kg) 2 M Facial flushing, transient penile engorgement (bilateral rhonchi, diarrhea) No Not reported (12)
200 mg 54 M Combined nonarteritic, anterior ischemic optic neuropathy and cilioretinal artery occlusion No Not reported (8)
250 mg 45 M Rhabdomyolysis, subjective changes in visual perception No Not reported (11)
6  50 mg (30 mg/kg) 1.5 M Mild facial flushing, asymptomatic tachycardia, prolonged priapism No 3.9 mg/mL after 7 h (10)
600 mg 56 M Cardiomegaly with dilated cardiomyopathy, diffuse coronary atherosclerosis and extensive Yes 6.3 mg/mL (6)
myocardial fibrosis
2000 mg 42 F Flushing, headache, mild tachycardia No Not reported (13)
2400 mg 33 M Defective color vision, vascular retinal dilatation, visual field defect, papilledema, annular scotoma No Not reported (4)

Experimental Table II
Reagents and chemicals Chromatographic Conditions
Sildenal citrate was purchased from Sigma-Aldrich (Bornem, Time (min) % Solution A % Solution B Flow (mL/min)
Belgium) and dissolved in methanol at a stock concentration of
0 95 5 0.625
2 mg/mL. Prazepam (LGC Standards, Molsheim, France) was 19 0 100 0.595
used as internal standard (IS) in a 0.75 mg/mL methanol working 23 0 100 0.625
24 95 5 0.625
solution. All other reagents were purchased from Merck (VWR, 30 95 5 0.625
Haasrode, Belgium). Methanol and acetonitrile were of HPLC
grade.
interference. Carry-over effects were studied by measuring blank
Sample preparation samples after analysis of samples with high sildenal concentration.
Verication of the linearity was performed by analyzing seven cal-
Venous blood samples were collected in Ethylenediaminetetra-
ibration standards (0.025, 0.05, 0.1, 0.25, 0.5, 1 and 2.5 mg/mL) on
acetic acid (EDTA) serum geltubes, centrifuged and stored at
six different days. A calibration curve was constructed by plotting
48C until analysis. To 350 mL serum, 35 mL methanol, 35 mL IS
the peak area ratio (Y-axis) of the target compound to the IS versus
working solution, 100 mL 1 M carbonate buffer pH 9.5, 400 mL
the nominal concentrations (X-axis) with a least-squares linear re-
ethyl acetate and 950 mL n-hexane were added. After mixing vig-
gression. The method was considered linear if the coefcient of de-
orously on a vortex shaker for 30 s, the samples were rotated for
termination (r 2) was equal to or better than 0.99. The lower limit of
5 min at a speed of 20 rpm, again vortexed during 5 s and centri-
quantication (LLOQ) was dened as the lowest concentration on
fuged at 10,000 g for 5 min. The upper layer was transferred to
the calibration plot with an imprecision and bias of ,20%. To eval-
another vial and dried under a nitrogen stream. The residue
uate the accuracy and precision of the method, quality control
was reconstituted in 200 mL methanol and again dried after a
(QC) samples at four different concentrations are analyzed in
30 s vortex step and centrifugation at 5,000 g during 2 min.
6-fold on one day (intra-day accuracy and precision) or on six con-
The residue was resolved in 70 mL of a watermethanol mixture
secutive days (inter-day accuracy and precision). The following QC
(75:25, v/v). Twenty microliters of this solution were injected
concentrations were used: 0.025 mg/mL (LLOQ), 0.075 mg/mL
into the HPLC system.
(low QC), 0.25 mg/mL (medium QC) and 2 mg/mL (high QC).
To meet the EMA accuracy criterion, the mean value of the calcu-
Chromatography lated concentrations should be within 15% of the actual fortied
HPLC was carried out on an Agilent 1100/1200 series HPLC value. The precision is reported as the variation coefcient (%)
equipped with an autosampler, quaternary pump, column oven and should not exceed 15% at every concentration. The recovery
and photodiode array (PDA) detector. System management and was assessed by comparing the peak area ratios (sildenal/IS) be-
data acquisition were performed with the Agilent Chemstation tween samples in which sildenal was added before and after ex-
software. The analytical HPLC column was a C8 ZORBAX traction. This was repeated six times for both the low and high QC
Eclipse Plus (Agilent), 150 mm  3.0 mm internal diameter concentration. A 10-fold dilution of serum was also tested (n 6)
with 3.5 mm particle size. The column temperature was held at due to the high expected sildenal concentration in the patient
408C. The run was performed with a gradient shown in serum samples. Data analysis was generated using GraphPad
Table II, and the mobile phase was a mixture of solution A Prism software version 6.05 for Windows.
(10 mM phosphate buffer, pH 2.3) and solution B (10 mM phos-
phate buffer, pH 2.3: acetonitrile (20:80)). Detection was per-
formed at 225 nm. Results
Method validation
Method validation The analytical peaks of sildenal and IS are well resolved with re-
The quantication of sildenal in serum was validated according to tention times of 10.3 and 16.6 min, respectively. Chromatograms
European Medicines Agency (EMA) guidelines (16). Ten sources of of blank and calibrator are presented in Figure 1a and b. No inter-
blank serum were individually analyzed and evaluated for fering peaks were observed in the extracts of 10 different blank

A Non-Fatal Self-Poisoning Attempt with Sildenal 573

Downloaded from https://academic.oup.com/jat/article-abstract/39/7/572/819563

by guest
on 11 November 2017
 Figure 1. Chromatograms of blank serum (a), 0.25 mg/mL sildenafil calibrator (b) and 10-fold diluted patient sample (24 h after intake) (c). The retention times of sildenafil and IS are
10.3 and 16.6 min, respectively.

574 Matheeussen et al.

Downloaded from https://academic.oup.com/jat/article-abstract/39/7/572/819563

by guest
on 11 November 2017
 Table III
Analytical Performance (Precision and Accuracy)

Target Intra-day (n 6) Inter-day (n 6)

concentration
(mg/mL) Concentration CV% % Concentration CV% %
(mg/mL) Error (mg/mL) Error
0.025 0.029 8.3 16.5 0.025 12.9 21.3
0.075 0.068 7.8 29.7 0.071 9.2 25.8
0.250 0.226 7.6 29.8 0.240 7.7 23.9
2.000 1.831 10.2 28.5 2.008 6.9 0.4

Figure 3. Elimination of sildenafil on a semi-logarithmic plot.

detected but the concentration was below the LOQ and the sam-
ple was thus discarded from further analysis. Linear regression
analysis of the data in Figure 3 resulted in the following equation
y 20.16x 7.0 with an r 2 value of 0.9995. Extrapolation
revealed a sildenal half-life in this patient of 4.2 h which corre-
sponds to the previously reported half-lives between 1.4 and
4.5 h (5).
Figure 2. Elimination of sildenafil on a linear plot.

Discussion
plasma samples and no carry-over from high concentrated sam-
ples was found. The limit of detection (LOD) was 0.008 mg/mL. The pharmacokinetics of sildenal, when used in therapeutic
The sildenal calibration curve ranging from 0.025 to 2.5 mg/mL, doses, has been studied extensively in the past. For example in a
exhibited a good linearity with a coefcient of determination study of Nichols et al., oral administration of sildenal in adults re-
(r 2) of 0.999. Both intra- and inter-day variation coefcients sulted in peak plasma concentrations of 0.13, 0.27 and 0.56 mg/mL
were below 15% and error percentages were between 215% after 25, 50 and 100 mg sildenal, respectively (7), which
and 15% (Table III). The recovery rates were 79 + 8% for the is comparable to the 0.08, 0.16 and 0.33 mg/mL reported
low QC concentration and 80 + 6% for the high concentration. by Milligan et al. after the same doses of sildenal (17). Peak
A 10-fold dilution of high sildenal-concentrated serum resulted concentrations were reached 1 h postdose. The results of both
in concentrations of 97 + 10% of the expected values. pharmacokinetic studies imply that plasma concentrations are pro-
portional to the administered sildenal dose (7, 17). However, ac-
cording to the latter study of Milligan et al., there is evidence for a
nonproportionality at higher doses since ingestion of 200 mg
Patient results sildenal resulted in a Cmax of 0.90 mg/mL, which was a 30%
The diagnosis of sildenal intoxication was conrmed by the increase in bioavailability relative to the other doses (17).
analysis of multiple serum specimens of the patient during 3 Little is known about the relation between nontherapeutic
days in which the clearance of the drug could also be studied. sildenal doses, sildenal serum concentrations and clinical tox-
The rst sample was drawn at the time of hospital admission, icology. The highest sildenal concentration found in our patient
which was 24 h after the sildenal intake, and showed the was 22.2 mg/mL, which is about three times higher than the
highest concentration (22.2 mg/mL). The chromatogram of this 6.3 mg/mL that was detected in one of the fatal case reports
sample (10-fold diluted) is shown in Figure 1c. Since the patient (6). After studying the few reported overdose cases (Table I), it
presented himself hours after the intoxication, the initial con- becomes clear that there is also not a good correlation between
centration had probably been higher (about 900 mg/mL). the administered dose and the clinical outcome, which might
Moreover, as adsorption of sildenal to the separator gel in the be explained by pre-existing risk factors or pharmacokinetic
serum collection tubes cannot be excluded, in vivo concentra- differences. Especially these last differences are hard to unravel
tions might have been even higher. The subsequent measure- because sildenal concentration measurements were not per-
ments revealed a rapid decline in sildenal concentration to 9.2 formed in most case reports.
and 2.3 mg/mL, respectively, 29 and 38 h after intake. Sildenal In the case described by Hung et al. (13), only mild symptoms
clearance showed rst-order kinetics in which the plasma con- (facial ushing, headache and mild tachycardia) occurred, while
centrationtime prole during the elimination phase decreases the case reported by Tracqui et al. (6) was fatal after ingestion of
exponentially in the plot with linear axes (Figure 2) and is linear a 3-fold lower sildenal dose (12  50 mg versus 20  100 mg).
if plotted on a semi-logarithmic scale (Figure 3). In a fourth However, one might assume that the pre-existing risk factors for
serum sample, taken 64 h after ingestion, sildenal could be coronary artery disease in the deceased patient will have

A Non-Fatal Self-Poisoning Attempt with Sildenal 575

Downloaded from https://academic.oup.com/jat/article-abstract/39/7/572/819563

by guest
on 11 November 2017
 contributed signicantly to the evocation of sudden cardiac library/EPAR_-_Product_Information/human/000638/WC500055840.
death with this lower sildenal dose (6). Our patient had no note- pdf (accessed 23 December, 2014).
worthy medical history. 2. European Medicines Agency. Viagra Summary of Product
Characteristics. http://www.ema.europa.eu/docs/en_GB/document_
The sildenal dose of the second fatal case (9) is unknown and library/EPAR_-_Product_Information/human/000202/WC500049830.
the serum concentrations were not measured, but again the pdf (accessed 23 December, 2014).
existence of a predisposing risk factor, advanced alcoholic cir- 3. Rang, H., Dale, M., Ritter, J., Moore, P. Pharmacology, 5th edition.
rhosis, might have contributed to the fatal outcome as the liver Churchill Livingstone: Edinburgh, UK, 2003; p. 444.
is the predominant route of elimination of sildenal (7). Serum 4. Giuliano, F., Jackson, G., Montorsi, F., Martin-Morales, A., Raillard, P.
aspartate transaminase (AST) activity is one of the covariates (2010) Safety of sildenal citrate: review of 67 double-blind placebo-
controlled trials and the postmarketing safety database. International
that signicantly inuence the sildenal clearance, together
Journal of Clinical Practice, 64, 240255.
with age and co-administration of potential CYP3A4 inhibitors. 5. Baselt, R.C. Disposition of Toxic Drugs and Chemicals
For every 10 units increase in AST the clearance of sildenal in Man, 10th edition. Biomedical Publications: Seal Beach, CA,
decreases by 6% (17). The AST concentrations in the cirrhotic 2014; pp. 1846 1848.
patient were not reported (9), but it is clear that the diminished 6. Tracqui, A., Miras, A., Tabib, A., Raul, J.S., Ludes, B., Malicier, D. (2002)
metabolization capacity will have had an effect on the clearance Fatal overdosage with sildenal citrate (Viagra): rst report and re-
view of the literature. Human & Experimental Toxicology, 21,
of sildenal. Our patient had a well-preserved liver function,
623629.
shown by normal levels of AST, alanine transaminase, alkaline 7. Nichols, D.J., Muirhead, G.J., Harness, J.A. (2002) Pharmacokinetics of
phosphatase and gamma-glutamyl transferase. sildenal after single oral doses in healthy male subjects: absolute bio-
Sildenal is metabolized by CYP3A4 and is therefore potential- availability, food effects and dose proportionality. British Journal of
ly sensitive to multiple drug interactions. Our patient reported Clinical Pharmacology, 53(Suppl. 1), 5S12S.
the regular use of both lormetazepam and haloperidol. 8. Akash, R., Hrishikesh, D., Amith, P., Sabah, S. (2005) Case report: asso-
Although the latter drug is a potent CYP2D6 inhibitor, both ciation of combined nonarteritic anterior ischemic optic neuropathy
(NAION) and obstruction of cilioretinal artery with overdose of
drugs will not interfere with sildenal metabolization. Oral bioa- Viagra. Journal of Ocular Pharmacology and Therapeutics, 21,
vailability of sildenal normally averages 41%, but food causes 315317.
small reductions in rate (tmax prolonged by 1 h) and extend 9. Finley, D.S., Lugo, B., Ridgway, J., Teng, W., Imagawa, D.K. (2005) Fatal
(29% reduction of Cmax) of systemic exposure (7). Food intake variceal rupture after sildenal use: report of a case. Current Surgery,
is not reported in any of the overdose cases. 62, 55 56.
To the best of our knowledge, the sildenal intoxication case 10. Wills, B.K., Albinson, C., Wahl, M., Clifton, J. (2007) Sildenal citrate
ingestion and prolonged priapism and tachycardia in a pediatric
that we presented ingested the highest sildenal dose which re-
patient. Clinical Toxicology, 45, 798800.
sulted in the highest sildenal serum concentration ever report- 11. Oh, D. (2014) Sildenal overdose can cause rhabdomyolysis and
ed. The serum concentration exceeded those of both fatal (6) subjective visual perception changes. Nephrology, 19, 258.
and non-fatal cases (10) and is far above the peak concentration 12. Cantrell, F.L. (2004) Sildenal citrate ingestion in a pediatric patient.
after therapeutic doses. The ingested dose was well tolerated, Pediatric Emergency Care, 20, 314 315.
with the expected increase in severity of the normal adverse 13. Hung, D., Yang, D. (2001) Sildenal overdose in a female and patient.
Clinical Toxicology, 39, 423 424.
effects, especially for the visual disturbances, but none of the
14. Dersch, R., Anastasopoulos, C., Hader, C., Stich, O. (2013) Vertebral
other reported toxicity symptoms, like rhabdomyolysis (11) or artery dissection associated with sildenal abuse. Journal of
blindness (8), occurred. Although our patient, who did receive Clinical Neuroscience, 20, 742.
psychological follow-up, recovered well from his sildenal intox- 15. Tiryakioglu, S.K., Tiryakioglu, O., Turan, T., Kumbay, E. (2009) Aortic
ication, the unofcial sale of sildenal and variants and the expo- dissection due to sildenal abuse. Interactive Cardiovascular and
nential increase in online pharmacies remain a major concern, Thoracic Surgery, 9, 141 143.
especially for patients with pre-existing comorbidities or multi- 16. European Medicines Agency Guideline on bioanalytical method
validation-2011 Version. http://www.ema.europa.eu/docs/en_GB/
ple drug intake. document_library/Scientic_guideline/2011/08/WC500109686.
pdf (accessed 14 September, 2014).
17. Milligan, P.A., Marshall, S.F., Karlsson, M.O. (2002) A population phar-
References macokinetic analysis of sildenal citrate in patients with erectile dys-
1. European Medicines Agency. Revatio Summary of Product function. British Journal of Clinical Pharmacology, 53(Suppl. 1),
Characteristics. http://www.ema.europa.eu/docs/en_GB/document_ 45S 52S.

576 Matheeussen et al.

Downloaded from https://academic.oup.com/jat/article-abstract/39/7/572/819563

by guest
on 11 November 2017