With some justification, the public is convinced that modern medicine is good at treating acute severe problems, such as infection, but not so good at treating or preventing chronic medical conditions, such as chronic back pain. Epilepsy, by its nature, straddles the divide between acute and chronic conditions. Some patients use complementary and alternative medicines (CAM) in the hope that they will control their seizures. Other patients use them in an attempt to cope with comorbid conditions, including some medication-related side effects. It is not uncommon for patients to use herbal treatments for lack of energy, depression, or sleep problems.
Patients often do not mention their herbal drug use to their physicians. Sometimes, these alternative treatments are not thought to be medically relevant or may simply be forgotten during a rushed appointment. Other patients undoubtedly avoid mentioning their use of CAM because they suspect that their physicians would not approve. We, as physicians, are also often uncomfortable talking about natural remedies. When a patient asks about herbal remedies, we may be dismissive, answer in broad and uninformative yet discouraging terms, or simply admit our ignorance. Our lack of comfort stems partly from the limited coverage of such topics in medical literature. However, the bigger problems are a lack of substantiated studies, non-standardized products, and an herbal industry that operates more on myth than scientific rigor. As I can also attest from reviewing the literature, most herbal remedies are backed up only by spotty and anecdotal evidence.
Nevertheless, epilepsy brings patients back to us to seek advice. The following is a brief summary of possible answers to some of the most common questions:
Q: Are there any SAFE natural or alternative treatments that can cure my epilepsy without medications or surgery?
A: Epilepsy, like most chronic medical conditions, is often treated symptomatically – that is, by treating clinical seizures. Seizures occur intermittently, so their control is a test of the observer's patience. Some patients' epilepsy has an underlying curable cause, which often means an epileptic focus amenable to brain surgery, such as a tumor or mesial temporal sclerosis. Most other patients are never entirely cured. Many patients may enter periods of remission while on drugs. Sometimes, the seizures disappear altogether and do not recur when medication is withdrawn – there are certainly some younger patients who “outgrow” their epilepsy. For most patients, however, the most realistic hope is not a “cure,” but complete and often life-long seizure control through medication.
The ketogenic diet is the only non-medication/non-surgical treatment shown to be effective in some patients.1 It may reduce seizures in some patients and in rare instances control seizures completely, but it does not treat or cure the underlying epilepsy. This diet has adverse effects2 and is hard to follow for anyone with free access to vending machines or a refrigerator. It is sometimes offered to young children whose access to food can be controlled more strictly. The long-term safety of the diet has not been adequately studied in adults. Some limited cases series suggest, but certainly do not prove, that a modified Atkins diet may also be helpful.3
The other non-pharmacological approach is to implant a vagal nerve stimulator. However, this treatment is not non-invasive, and though it reduces seizures in some patients, it rarely eliminates them.
Q: I want to try such-and-such non-pharmacological treatment for my epilepsy. Is it safe or effective?
A: The short answer to both questions is “I don’t know.” In truth, some approaches are so benign that they cannot be too harmful even if their efficacy is unproven. These methods include relaxation techniques,4 yoga,5 hypnosis, acupuncture,6 biofeedback, and other treatments that promise to provide patients with some degree of control by using the body’s natural powers to produce a mild reduction in seizure symptoms.7 However, safety can still be a concern — even the use of magnets has been reported to carry a risk of seizures — and anecdotal reports of success hardly prove that these modalities can accomplish any biological result.
Q: I want to try such-and-such herbal or natural treatment for my epilepsy. Is it safe?
A: There are many reports of the use of herbal and natural remedies to treat epilepsy, but evidence of efficacy is lacking. Some agents have significant potential for harm. Fortunately, resources are now available to clinicians to review the basics of some herbal remedies. I have mostly used the Natural Medicine Database for this review,8 although a number of other good resources are available (references at bottom of article). Any promise of efficacy should be verified thoroughly. Given the lack of data about these treatments' teratogenic potential and their effects on children, herbal remedies should be avoided in children and women with the potential to bear children.
For many drugs, appropriate safety data is entirely lacking. Several remedies suggested for epilepsy are frankly poisonous, even in small quantities. Table 1 lists some substances with a reported risk of toxicity. These drugs should be avoided altogether. For other remedies with scant proof of safety or efficacy (Table 2), a similarly cautious approach is warranted. Other substances may be more widely used but have been reported to be rarely, but fatally, toxic (Table 3).
A number of other herbal remedies are somewhat more ubiquitous and appear safe on the surface. Many have been used in traditional remedies for hundreds of years. Others occur in commonly consumed herbs or are naturally present in small quantities in our diets. The FDA, for example, assigns the designation GRAS, or “Generally Regarded As Safe,” to some substances without formal safety testing, although this designation is usually accompanied by additional comments or restricted to a particular use of a substance. Nevertheless, some of the same substances may have toxic effects at higher doses. A good example is the ubiquitous caffeine, which, while safe for most people, can be fatal in extremely high doses.9 Caffeine at sublethal doses can also exacerbate seizures,10 a risk that is very low but is hard to quantify for individuals. The same can be said of many vitamins and herbs. Many common herbs, for example, contain minute amounts of toxic substances. Essential oils of these herbs can exacerbate seizures or even be fatal. Table 4 lists many herbal treatments and supplements that are thought to carry a risk of toxicity at high doses. A number of herbal drugs can exacerbate seizures. Table 6 lists remedies that are thought to increase seizure frequency in some patients. Curiously, some of the remedies in the table are also the ones purported to be helpful in some epilepsy syndromes. This is not surprising. Among traditional epilepsy medications, for example, we know that carbamazepine may make absence seizures worse and that diazepam can cause some Lennox-Gastaut patients to have more drop attacks. Unfortunately for herbal remedies, the anecdotal reports of efficacy do not describe the seizure types for which the herb is said to help (e.g., complex partial vs. generalized onset seizures).
Making matters more complicated are the pharmacological interactions between herbal preparations and conventional medications. Many herbal remedies (particularly some of the popular ones that appear first in Google searches) appear to mediate their effect via binding to GABAA receptors, which suggests that these substances could have effects similar to barbiturates and benzodiazepines and raise theoretical concerns (see section below). The first concern is that pharmacodynamic interaction with GABAA modulators, such as lorazepam, diazepam, or phenobarbital, could exacerbate adverse effects. The second concern is that many GABAA active drugs produce tolerance, are habit-forming, and pose a risk of withdrawal seizures. While this effect has not been proved for most herbal remedies, one has to exercise caution in starting and stopping the herbal remedies, particularly after prolonged use. It is not surprising that most anecdotal studies of the benefits of these herbal remedies report only short-term efficacy data, not long-term results. Table 5 lists some herbal remedies that have GABAA-ergic effects. The possible pharmacokinetic drug interactions of herbal remedies are discussed separately below.
It should not be assumed that all herbal remedies are pure and of appropriate quality. These substances are, in general, not tested to prove that they contain what they promise, and purity data is usually not available. In some cases, remedies have contained toxic substances (lead, mercury, and arsenic11,12) or toxic herbs that resemble the herbs advertised. Some recent studies have identified traditional preparations that contained common antiepileptic drugs (often phenytoin or phenobarbital) at inappropriate doses.13
Having said that, there are reputable suppliers of many supplements. Products from countries in which the herbal industry is regulated (such as Germany) may be more certain to contain the promised ingredients. Plants, however, do not heed government regulation and often produce different levels of active substances. The differences depend not only on the growers of the plants, but also on whether leaves are harvested in fall, spring or summer, not to mention whether the plant is grown in shade or full sun. There are also many reports that some of the desired active compounds may be concentrated during the packaging and drying process, while others are destroyed. Therefore, the therapeutic and toxic effects of a dose of a herbal preparation may differ significantly among various preparations.
The bottom line is that many of us find it unethical to prescribe herbs with uncertain pharmacological and therapeutic properties when there are known medicines that do not have these disadvantages. Herbal remedies should not ever be used for first-line treatment of epilepsy.
Q: I have symptom X and want to use the natural treatment Y to help relieve it. Given my epilepsy, is it safe to do so?
A: I probably have someone call with this question every month. Given the number of available herbal medicines and the number of seizure drugs, mentioning every possibility is rather hard. As mentioned, the lack of adequate data about the safety, efficacy, and pharmacological effects of herbal and alternative remedies is the main issue. Most pertinent is the inadequacy of our information about whether any of these remedies have the potential to make seizures worse.
Assuming that preparations are not obviously toxic, the main concerns include toxicity at high doses, the risk of seizure exacerbation, and pharmacological interactions. Again, Tables 1–6 above summarize the areas of concern. Herbal drugs can certainly have harmful interactions with antiepileptic drugs by pharmacokinetic interaction. Some of these are listed in Table 7.
Substances used for depression include St. John’s Wort, S-adenosyl-L-methionine (SAMe), and 5-hydroxytryptophan (5-HTP). Hyperforin, the main active ingredient in St. John’s Wort, induces the activity of the Cytochrome P450 (CYP) 3A4 isozyme and the p-glycoprotein transporter, resulting in several clinically significant drug interactions. St John's Wort is the most common herb involved in suspected herbal drug interactions. However, there are no clinically significant interactions with antiepileptic drugs. Even though carbamazepine is metabolized by CYP3A4, carbamazepine auto-induces CYP3A4 itself, and the addition of St. John’s Wort does not increase the induction. Little is known about the interactions of SAMe with epilepsy. In one preparation, 5-HTP was associated with fatal eosinophilia-myalgia syndrome. It may cause seizures in some populations (e.g., Down Syndrome15).
Among the sedatives and hypnotics, valerian is commonly used to treat anxiety and induce sleep and has a reasonable safety track record for short-term use, with a large number of patient exposures. It even carries the GRAS designation. However, when valerian is used to treat epilepsy, there are some concerns. It may have significant GABA effects.16 It may also partially block the P450 pathway, with a risk of drug interaction. Kava, another herbal anxiety treatment, may also have significant GABA effects.17 There is also evidence of tolerance to this herb and a rare risk of liver toxicity with chronic use.18 Acute toxicity has been reported when kava is used with other sedatives such as lorazepam or alcohol19. Passionflower, another herbal sedative, is similarly thought to have effects on the GABA pathway.20 There are reports of rare life-threatening toxicity from this herbal remedy as well, including cases of vasculitis21 and cardiac abnormalities.22 Ethanol, although not prescribed by herbal practitioners, belongs in this family, albeit with an even higher risk. These sedative herbs, though reported to have benefits for seizures, are not shown to be effective for epilepsy, and all carry risks with long-term use. Nevertheless, aside from the concerns noted above, there is little evidence of harm from these herbs if used occasionally, on an as-needed basis, at modest doses, and in patients at low risk for drug interactions.
Stimulants and other cognitive enhancers, by contrast, are more likely to exacerbate seizures and pose problems with drug interactions. They may be used by epilepsy patients in an attempt to treat the cognitive problems associated with seizures and antiepileptic drugs. The stimulant ephedra (known in Chinese medicine as ma huang) has in several cases been reported to precipitate seizures.23 Coca (a source of cocaine) may also result in seizures. The effects of coffee and tea are probably modest due to the very small doses of active substances — theophylline, theobromine, and caffeine — in these beverages. However, risks may be present with large doses or when these substances are combined with other stimulants such as ephedra.24 The cognitive “enhancer” ginkgo biloba has been associated with seizures,25,26 but this is not well documented. Ginkgo biloba is also suspected of having effects on antiepileptic drug metabolism, although this effect is likewise uncertain. There are some concerns about the effects of ginseng on the metabolism of warfarin, which is also metabolized by the P450 system. Whether this interaction occurs in the P450 isozyme relevant to antiepileptic drugs or via another mechanism is unclear. Many herbal products have been shown to affect drug metabolism via the cytochrome P450 isozymes when studied in vitro. However, the majority of in vitro interactions do not translate to clinically significant interaction in vivo.
Q: I tried a natural remedy for a while for my seizures and it seemed to work. Can you prescribe it for me?
I get this question most commonly regarding marijuana. Many users of cannabis feel that their seizures are less significant when they smoke it. The data on the long-term toxicity of smoked marijuana in adults is well known (it is probably worse than cigarettes but not as bad as chronic heavy alcohol use). There are more significant concerns about its use in teenagers (and probably young adults). Paranoia and effects on mood and motivation are quite concerning. Data on short-term effects on seizures are mixed, as cannabis is reportedly helpful in some but may make seizures worse in others.27 Long-term safety and efficacy data for epilepsy are lacking. Given the rates of depression, other psychiatric problems such as paranoia, and unemployment among epilepsy patients, one should show extreme caution in advocating the use of marijuana in this population. Most of us discourage its use, and I do not prescribe it.
This question also comes up regarding one of the non-pharmacological alternative treatments discussed above. In particular, we are occasionally asked to fill out a statement of need and candidacy for a service dog to aid in seizure detection. Some insurance companies may pay for biofeedback training if it is prescribed by a health care provider. While I am not sold on the benefits of these interventions, I think that they are safe enough for a patient to try even if they are unlikely to be cost effective.
Q: I may need to take my seizure drugs for life. Do I need to take/avoid any supplements to prevent/alleviate other health problems in the future? Are there any foods to avoid or consume?
A: Here, there is probably more information than patients recognize. I usually spend a large portion of my initial visit going over some of these concerns. First, some healthy habits may be useful. Most importantly, lack of sleep and chronic stress28 are likely harmful for epilepsy. Who could argue against a healthy diet, adequate sleep, or exercise? Many other substances are best avoided. The effects of alcohol use — its interaction with seizure drugs, risks of seizure exacerbation and liver toxicity — are well known to clinicians. Regular use of alcohol may also affect the metabolism of seizure drugs. The harms of smoked tobacco and marijuana are higher in patients with epilepsy due to a greater risk of bone density loss and gum disease, particularly when combined with phenytoin. Cocaine and amphetamines also carry a risk of seizures. Caffeine intake may be unhelpful at high doses.
All women of child-bearing age with epilepsy are encouraged to take supplemental folic acid. This is due to concerns about some overlapping associations: 1) a higher incidence of neural tube defects in fetuses of women with epilepsy than other women,29 2) a higher incidence of neural tube defects in women with low dietary folic acid,30 and 3) the fact that some seizure medications are folic acid antagonists.31 Note that the critical issue of whether folic acid supplementation reduces rates of birth defects in the offspring of epilepsy patients is unanswered. Nevertheless, many of us advocate from at least 400 mcg daily (the amount in a prenatal multivitamin) to as much as 4 mg administered twice daily (a dose used safely in the treatment of folate deficiency associated neural tube defects).
Whether other groups of patients benefit from folate supplementation is unknown. There is a higher risk of strokes and heart disease in epilepsy patients than in the general public.32 This risk is associated with elevated blood levels of homocysteine,33 which may possibly be reduced by folate supplementation. It is unknown whether folate supplementation leads to better cardiovascular health in patients. Nevertheless, low dose supplementation may be helpful and carries little risk.
A concern is that folate supplements may delay diagnosis of B12 deficiency by masking megaloblastic anemia until neurological symptoms of B12 deficiency develop. Whether patients should be screened for B12 deficiency when taking folate is unclear, but this is unlikely to be cost effective. A simple substitution of B-complex for straight folic acid pills would not alleviate this concern because pernicious anemia is most often caused by problems with vitamin absorption rather than dietary deficiency.
Anecdotally, some patients report fewer side effects from their antiepileptic drugs when given B-vitamins. It has been claimed that pyridoxine may alleviate irritability from levetiracetam.34 Riboflavin has been advocated for prophylaxis for migraines,35 a common problem among epilepsy patients. As mentioned earlier, these supplements are safe at low doses.
The need for supplementation of other vitamins in patients with epilepsy is a bit unclear. Older literature has suggested that epilepsy patients tend to have low vitamin E and D levels, but it is unclear if this is significant. Whether a multivitamin is sufficient or even necessary is unknown. Given the high incidence of osteopenia, the use of vitamin D and calcium is probably prudent. A related concern is structural bone weakness in patients with epilepsy. This may be measured as osteopenia using DEXA (dual-energy X-ray absorptiometry). However, bones may also be of lower quality without frank osteopenia. For example, the diameter of axial bones may be reduced in patients who develop epilepsy at a young age. Weight-bearing exercise and sunshine may help in building stronger bones. Exercise may have the additional benefit of warding off depression and seasonal affective disorder.
For some other patients on particular drugs, certain nutritional supplements may be helpful. For example, levocarnitine levels may be low in patients on valproic acid.36 Patients with progressive epilepsy syndromes may also have problems with mitochondrial function. There are some dietary supplements that may help with these conditions. In these cases, however, there is the potential for high cost and even toxicity. Levocarnitine supplements, for example, may be associated with an increased risk of seizures. Supplements with narrower benefit should be treated like any other prescribed drug: they should be started one at a time under a physician’s guidance and titrated slowly with the hope of achieving a particular measurable response. Fortunately, the data for the safety and efficacy of these supplements are generally of higher quality.
Lastly, some foods at typical consumed quantities have the potential for interactions with seizure drugs. Grapefruit juice in particular contains substances that inhibit one primary CYP isozyme, CYP3A4, and may lead to toxicity of carbamazepine.38 However, none of the other antiepileptic drugs are significantly affected by this interaction.
Q: I want only natural remedies for my epilepsy. Is there any effective natural remedy you can offer me?
Some patients, unfortunately, simply lack faith in the practice of Western medicine. The short answer to this question is that there are no proven natural remedies effective for epilepsy, particularly for long-term use. Many antiepileptic drugs have short- and long-term toxicity and side effects, but there is no proof of safety for natural remedies at pharmacological doses, either.
Several natural remedies are listed as having “possible” or “proven” efficacy in the Natural Medicine Database. On close inspection, these treatments have only a small window of relevance. Pyridoxine is useful for the rare infant with pyridoxine dependent epilepsy.39,40 The use of N-acetylcysteine (Mucomyst) is supported only by a small case series in a single family of patients suffering from Unverricht-Lundborg disease.41 Medium-chain triglycerides, a group of naturally occurring fatty acid derivatives, have been thought to be helpful for patients with epilepsy,42 with as many as 55 patients reporting benefits in one study. In the late 1980s, however, it was discovered that patients who had been on the diet for more than two years were developing liver abnormalities,43 leading to the diet falling somewhat out of favor, although it is still sometimes used.
What should doctors do with the patient who refuses to take Western medicine based on a philosophical opposition? One may try to reconcile Western medicine with their beliefs. In any case, the physician’s responsibility is to inform the patient about the safety concerns, drug interactions, and adverse effects of herbal products, and the lack of adequate evidence that these agents are effective in the treatment of epilepsy, which is a dangerous condition.