Fatal Seizures Due to Potential Herb-Drug Interactions - Journal of ...

Journal of Analytical Toxicology, Vol. 29, October 2005
] Case Report [
Fatal Seizures Due to Potential Herb-Drug Interactions
with Ginkgo Biloba
Thomas Kupiec and Vishnu Raj
Analytical Research Laboratories, 840 Research Parkway, Suite 546, Oklahoma City, Oklahoma 73104
Abstract[
Alternative therapy including herbal drugs and complementary
medicine is becoming increasingly popular. However, the rise in the
incidence of herb-drug interactions is causing concern, especially in
the absence of warning labels addressing potential adverse effects.
We present the case of a 55-year-old male who suffered a fatal
breakthrough seizure, with no evidence of non-compliance with his
anticonvulsant medications. The autopsy report revealed sub-
therapeutic serum levels for both anticonvulsants Depakote | and
Dilantin | . Concomitant with his prescribed medications, the
decedent was also self-medicating with a cornucopia of herbal
supplements and nutraceuticals, prominent among which was
Ginkgo biloba. Ginkgo, an herbal extract from the leaves of the
Ginkgo biloba tree, has been used medicinally for centuries and has
been touted as a cure for a variety of medical conditions. The
induction of Cytochrome P450 enzymes by components of herbal
drugs has been known to affect the metabolism of various drugs.
Dilantin is primarily metabolized by CYP2C9, and secondarily
metabolized by CYP2C19. Valproate metabolism is also modulated
in part by CYP2C9 and CYP2C19. A recent study revealed
significant inductive effect of ginkgo on CYP2C19 activity.
CYP2C19 induction by ginkgo could be a plausible explanation for
the subtherapeutic levels of Dilantin and Depakote. Additionally,
ginkgo nuts contain a potent neurotoxin, which is known to induce
seizure activity. Evidence of other herbal drugs diminishing the
efficacy of anticonvulsant medication does exist; however, there has
been only one other documented instance of ginkgo potentiating
seizure activity in the presence of anticonvulsant therapy.
Highlighting the potential adverse effects and drug interactions of
ginkgo on the packaging of the drug may help prevent inadvertent
use in vulnerable individuals.
Introduction
Alternative therapy including herbal drugs and complemen-
tary medicine are becoming increasingly popular in the Western
world. Although some of these drugs have potential benefits,
there is an erroneous public perception that these herbal drugs
are benign and without adverse effects (1). However, as is being
documented, herb-drug interactions are becoming alarmingly
frequent, especially considering the absence of a label ad-
dressing the possibility of adverse effects. The incidence of
herb-drug reactions are generally based on two factors, that is,
factors related to drug and factors related to the individual.
Drug-related factors are dose, dosing regimen, route of admin-
istration, pharmacokinetics, and therapeutic range. Patient as-
sociated factors are genetic polymorphisms, age, gender, and
pathology (2). The following case is about a 55-year-old pa-
tient, compliant with his seizure medications, suffering a fatal
breakthrough seizure. The autopsy report revealed subthera-
peutic serum levels of both the anticonvulsants, Depakote and
Dilantin. Concomitant with his prescribed medications, the pa-
tient was also self-medicating with a cornucopia of herbal sup-
plements and nutraceuticals, prominent among which was
ginkgo biloba. Ginkgo is an herbal extract from the leaves of the
Ginkgo biloba tree, and has been used medicinally for cen-
turies (3). It has been credited with increasing cerebral blood
flow, and the herbal extract is used to treat a variety of condi-
tions including post4hrombotic syndrome, peripheral vascular
disease, difficulties with memory, confusion, anxiety, headache,
tinnitus, etc. (3-5). In 1998, more than $649 million was spent
on medicinal herbs in the United States (6), and ginkgo biloba
was the top seller with reported retail sales of $150 million. The
biologically active components of ginkgo biloba include the
ginkgo-flavone glycosides or flavonoids (e.g., bilobetin,
ginkgetin, sciadopitisin, quercetin, etc.) and the terpenoids
(e.g., bilobalides and ginkgolides) (2,5-7). Ginkgolides are po-
tent inhibitors of platelet aggregating factors, which explains
the purported pharmacologic actions of ginkgo (i.e., increased
vasodilation and peripheral blood flow rate) (2). Reports of se-
rious adverse effects such as prolonged bleeding times and
spontaneous subdural hematomas have been associated with
the anti-platelet aggregating factor activity of ginkgo (3,8).
Ginkgo has also been reported to interfere with platelet function
and has been associated with bleeding, even in the absence of
anti-coagulants (2). Some of the possible drug interactions of
ginkgo include aspirin, warfarin, ticlopidine, clopidogrel, and
dipyridamole (9). Although there has been documented evi-
dence of an herbal agent (shankapulshpi) causing diminished ef-
fectiveness of anticonvulsant medication (phenytoin), there has
been only one other reported instance of ginkgo potentiating
seizure activity in the presence of anticonvulsant medication
(3,8).
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission. 755

Case History
The decedent, a 55-year-old male, was swimming at a health
club with an attendant walking alongside as he swam, when he
suddenly went underwater and lost consciousness. The two life-
guards at the pool immediately pulled him out, and one of them
initiated CPR; paramedics were summoned. There was an initial
report of a seizure in the pool, but it could not be corrobo-
rated. During the time that the paramedics attempted resusci-
tation, they encountered a pulseless rhythm, ventricular fibril-
lations, and ventricular tachycardia. The patient received
defibrillation and intravenous advanced cardiac life support
medications including epinephrine, bicarbonate, and normal
saline. At the ER, he was observed to be lifeless and cyanotic,
with an absence of spontaneous movements or respirations.
Pulmonary sounds were symmetric, with bilateral coarse rales
and rhonchi. The pupils were fixed and dilated with an absence
of corneal and doll's eye reflexes. The nail beds were cyanotic,
and the patient was flaccid with an absence of neurological re-
sponses. The patient was pronounced dead, and an autopsy was
indicated. Autopsy results revealed evidence of an old cerebral
infarct and severe coronary vascular disease, as well as an old
myocardial infarct and prior bypass surgery. The toxicological
analyses involved a broad-based gas chromatography (GC)
screen for acidic/neutral drugs, with confirmation by GC-mass
spectroscopy (MS). Trinders' test was used to check for salicy-
lates, and headspace GC was used for analysis of volatiles. Chem-
Elut columns were used for GC. The toxicological analysis re-
vealed a femoral blood valproic acid (Depakote) level of less
than 26 IJg/mL and a phenytoin (Dilantin) level of 2.5 IJg/mL.
Caprillic acid and para-methyl phenytoin were used as internal
standards for valproic acid and phenytoin, respectively. Both
the anti-epileptic drugs were considered to be subtherapeutic at
these levels. The cause of death was ruled to be a seizure disorder
due to an old cerebrovascular event. According to his medical
records, the patient had been in sound health until a year prior
to his demise, when he experienced a cerebrovascular accident
following coronary artery bypass surgery, after which he devel-
oped a seizure disorder. His medical records revealed that, on
three occasions, his anticonvulsant (Dilantin, Depakote) serum
drug levels had dipped below the therapeutic range for no ap-
parent reason. Table I provides serum Dilantin levels from the
medical records of the deceased. On one occasion, he experi-
Table I. Fluctuations in the Dilantin Levels
Date Dilantin Level Therapeutic Range
4/24/00 4.1 mg/L
5/3/00 10.4 mg/L
7/21/00 14.0 mg/L
8/4/00 9.6 mg/L
8/14/00 10.6 mg/L
8/25/00 13.7 mg/L
9/29/00 21.2 mg/L
10/20/00 14.1 mg/L
11/30/00 13.9 mg/L
756
10 to 20 mg/L
Journal of Analytical Toxicology, Vol. 29, October 2005
enced a breakthrough seizure with a therapeutic Dilantin
level. Following his demise, it was discovered that apart from
his prescription medications, the deceased was taking copious
nonprescription vitamins, supplements, and herbal nutraceuti-
cals, apparently without the knowledge of his physician. He
had begun taking the supplements approximately a year prior
to his death. The most common herbal ingredient in these
preparations was ginkgo, an extract from the leaves of the
Ginkgo biloba tree. Among the other ingredients were ginseng
and saw palmetto extracts. The possible seizure-potentiating
role of these herbal products was explored in an extensive liter-
ature study.
Discussion
Herbal products have long been popular, and some of their
modern derivatives constitute highly prescribed medications
(e.g., the evolution of digoxin from foxglove). Although many
herbs have been evaluated in randomized controlled trials, the
limitations of poor design, small samples, and uncertain com-
position of products make any reliable interpretation impossible.
Furthermore, the absence of advisory warning labels addressing
possible adverse effects of these nutraceutical agents, make it
imperative to study and document any possible drug interac-
tions. The U.S. Food and Drug Administration's Special Nutri-
tionals Adverse Effect Monitoring System (SN/AEMS) described
seven reports of seizures associated with ginkgo. Four reports
were associated with multi-ingredient products and three were
associated with single-ingredient preparations, each from dif-
ferent manufacturers (10). Several reports in the literature in-
dicate that consumption of ginkgo nuts might induce seizure ac-
tivity. After the consumption of ginkgo nuts, a previously healthy
36-year-old woman without any past or family history of epilepsy
developed vomiting and generalized tonic and clonic convul-
sions (11). Ginkgo nuts contain a potent neurotoxin, 4'-0-
methoxypyridoxine (4'-MPN), which is known to indirectly in-
hibit the enzyme activity of glutamate decarboxylase, resulting
in decreased levels of y-amino-butyric acid (GABA)(12). GABA
is an inhibitory neurotransmitter, and a reduced concentration
of GABA may lead to seizure induction (11). The leaves of the
ginkgo tree are a source of extracts for commercial ginkgo
products. Although the majority of commercial ginkgo products
may not contain sufficient 4'-MPN to cause seizures, depending
on the harvest season and the potential introduction of con-
taminants, 4'-MPN may be present in amounts sufficient to
cause seizures in a vulnerable populace, such as infants or in-
dividuals with known seizure disorders (13). Arenz et al. (14)
documented the occurrence of 4'-MPN in ginkgo leaves and
some commercially available ginkgo preparations. Manocha et
al. (15,16) proposed an alternative mechanism for the putative
seizure activity of ginkgo biloba~ They found that ginkgo, a
platelet-aggregating factor (PAF) antagonist, facilitated strych-
nine-induced convulsions in mice. This study emphasized the
possible modulating effect of ginkgo biloba on the inhibitory
neurotransmitter glycine because strychnine competes with
glycine, thus exerting a central stimulant effect through in-

Journal of Analytical Toxicology, Vol. 29, October 2005
hibitory blockade (15,16). Additionally, the prior administration
of ginkgo to mice potentiated the seizure activity of picrotoxin,
a GABA antagonist, in the presence of the anti-convulsant drugs
sodium valproate and carbamazepine. This was possibly be-
cause of the anti-GABAergic activity of ginkgo components
(15,16).
Another putative hypothesis in the current case was the
induction of the Cytochrome P450 (GYP) enzyme system by
nutraceutical ingredients, causing extensive metabolism of
the anticonvulsant mediations, thus leading to seizure induc-
tion due to the subtherapeutic levels of these drugs. Zhou
et al. (2) describe the induction and inhibition of CYP enzyme
system by various components of herbal drugs. If the CYP
enzymes metabolizing a certain drug are induced, they
will hasten its elimination from circulation, thereby dimin-
ishing its therapeutic benefit (17). Dilantin is primarily metab-
olized by CYP2C9 and secondarily metabolized by CYP2C19
(18). Additionally, valproate is also metabolized by CYP2C9
and CYP2C19; however, these enzymes account for a relatively
minor portion of its excretion (19). A recent study by Yin
et al. (20) found significant inductive effect of ginkgo on
CYP2C19 activity. They concluded that co-administration of
ginkgo with a CYP2C19 substrate may significantly reduce
the effect of the drug. CYP2C19 induction by ginkgo could be
a plausible explanation for the subtherapeutic levels of
Dilantin and Depakote.
Granger (1), in what is probably the only other documented
case of ginkgo precipitating seizure activity in the presence of
anticonvulsant medications, describes two patients with well-
controlled epilepsy, compliant with their anti-epileptic medi-
cation (sodium valproate), who developed seizures within two
weeks of using ginkgo products. After cessation of the herbal
remedy, both the patients remained seizure-free without any in-
crease in the dosage of antiepileptic medication (1). Other re-
ports of ginkgo include an elderly female with elevated blood
pressure who was treated with a thiazide diuretic and later
began taking ginkgo biloba. After one week, her blood pressure
increased further and remained increased for a few weeks. After
stopping ginkgo and the diuretic, blood pressure returned to
pretreatment level. Furthermore, in patients with epilepsy,
there have been also reports of breakthrough seizures occurring
with the combination of valproate and phenytoin (13). The ab-
sence of federal mandate and the lack of a warning label may en-
danger individuals afflicted with epilepsy. Miller (8) advises
against the use of ginkgo products in epileptic patients, as the
ginkgo neurotoxin 4'-MPN could nullify the prophylactic effects
of anticonvulsants. Tyagi et al. (21) suggest that the use of
ginkgo may lower the seizure threshold, and they advise against
its use with antiepileptic medication. Ginkgo is known to pro-
vide some therapeutic benefit in cerebral ischemia, memory de-
ficiency, and other cerebral disorders; however, a co-morbid
seizure disorder may also be present in these individuals. For
them, any possible benefit with ginkgo could be offset by po-
tential interactions with anticonvulsive medications. In his re-
port, Granger (1) emphasizes the need for highlighting the
seizure potential and other possible adverse effects of ginkgo on
the packaging of the drug.
Conclusions
Currently, there are no means to ascertain that the ginkgo
product is free of the seizure-inducing neurotoxin. In the pre-
sent case, the fluctuations in the concentrations of Dilantin
and Depakote could not be definitively attributed to herb-drug
interactions. However, in the absence of an alternative hypoth-
esis, it is our belief that the nutraceutical product contributed
to breakthrough seizures in the decedent. Although the associ-
ation is not conclusive, the literature reports have sufficient pro-
bative value to suggest caution in the use of ginkgo products by
individuals with a history of convulsive disorders. A balance is
sought between the proven benefits and the suspected adverse
effects of ginkgo. However, without a qualifying label or a stan-
dardized quality control process, use of some nutraceutical
products may prove to have hazardous consequences in sus-
ceptible individuals.
References
1. A.S. Granger. Ginkgo biloba precipitating epileptic seizures. Age
Ageing 30:523-525 (2001).
2. S. Zhou, Y. Gao, W. Jiang, M. Huang, A. Xu, and J.W. Paxton. In-
teractions of herbs with cytochrome P450. Drug Metab. Rev.
35(1): 35-98 (2003).
3. J. Rowin and S.L. Lewis. Spontaneous bilateral subdural
hematomas associated with chronic Ginkgo biloba ingestion. Neu-
rology46(6): 1775-1776 (1996).
4. J. Kleijnen and P. Knipschild. Ginkgo biloba. Lancet 340:
1136-1139 (1992).
5. M.K. Kottke. Scientific and regulatory aspects of nutraceutical
products in the United States. Drug Dev. Ind. Pharm. 24(12):
1177-1195 (1998).
6. E.P. Elovic and R.D. Zafonte. Ginkgo biloba: applications in trau-
matic brain injury. J. Head Trauma Rehabil. 16(6): 603-607 (2001).
7. X.F. Yang, N.P. Wang, W.H. Lu, and F.D. Zeng. Effects of Ginkgo
biloba extract and tanshinone on cytochrome P-450 isozymes
and glutathione transferase in rats. Acta Pharmacol. Sin. 24(10):
1033-1038 (2003).
8. L. Miller. Herbal medicinals: selected clinical considerations fo-
cusing on known or potential drug-herb interactions. Arch. In-
tern. Med. 158(20): 2200-2211 (1998).
9. M.J. Cupp. Herbal remedies: adverse effects and drug interac-
tions. Am. Faro. Physician 59(5): 1239-1245 (1999).
10. P.J. Gregory. Seizure associated with Ginkgo biloba? Ann. Intern.
Med 134(4): 344 (2001).
11. H. Miwa, M. lijima, S. Tanaka, and Y. Mizuno. Generalized con-
vulsions after consuming a large amount of ginkgo nuts. Epilepsia
42(2): 280-281 (2001).
12. M. Shannon, E.A. McEIroy, and E.L. Liebelt. Toxic seizures in chil-
dren: case scenarios and treatment strategies. Pediatr. Emerg. Care
19(3): 206-210 (2003).
13. Drugdex Drug Evaluations (MICROMEDEX) - Dilantin | De-
pakote | and Strychnine.
14. A. Arenz, M. Klein, K. Fiche, J. Gros, C. Drewke, T. Hemscheidt,
and E. Leistner. Occurrence of neurotoxic 4'-O-methylpyridoxine
in Ginkgo biloba leaves, ginkgo medications and Japanese ginkgo
food. Planta Medica 62:548-551 (1996).
15. A. Manocha, K.K. Pillai, and S.Z. Husain. Influence of ginkgo
biloba on the effect of anti-convulsants. Indian J. Pharmacol. 28:
84-87 (1996).
16. A. Manocha, K.K. Pillai, S.Z. Husain. Effect of ginkgo biloba on
757

chemoshock in mice. Indian J. Pharmacol. 29:198-200 (1997).
17. E.M. Williamson. Drug interactions between herbal and prescrip-
tion medicines. DrugSaf. 26(15): 1075-1092 (2003).
18. The Worldwide Physiologist: cytochrome P450. http://www.anaes-
thetist.com/physiol/basics/metabol/cyp/p.htm#phenytoin.
19. Goodman and Gilman's The Pharmacological Basis of Therapeu-
tics, 10th ed., J.G. Hardman, L.E. Limbird, and A.G. Gilman, Eds.
758
Journal of Analytical Toxicology, Vol. 29, October 2005
McGraw-Hill, New York, NY, 2001.
20. O.Q. Yin, B. Tomlinson, M.M. Waye, A.H. Chow, and M.S. Chow.
Pharmacogenetics and herb-drug interactions: experience with
ginkgo biloba and omeprazole. Pharmacogenetics 14(12):
841-850 (2004).
21. A. Tyagi and N. Delanty. Herbal remedies, dietary supplements,
and seizures. Epilepsia 4.4(2): 228-235 (2003).