ADVERSE DRUG REACTIONS, HERBAL ADVERSE EVENT REPORTS, AND ADVERSE EVENTS REPORTING SYSTEMS

Several operational definitions exist regarding what constitutes adverse drug reactions (ADRs) of varying severity. According to the World Health Organization (WHO), an ADR is defined as “Any response to a drug which is noxious and unintended, and which occurs at doses normally used in man for prophylaxis, diagnosis, or therapy of disease, or for the modification of physiologic function.”⁹ For reporting purposes, the FDA categorizes a serious adverse event as one in which “the patient outcome is death, life-threatening (real risk of dying), hospitalization (initial or prolonged), disability (significant, persistent, or permanent), congenital anomaly, or required intervention to prevent permanent impairment or damage.”¹⁰ The American Society of Health-System Pharmacists (ASHP) defines a significant ADR as any unexpected, unintended, undesired, or excessive response to a drug that:

Consistent with this definition, allergic and idiosyncratic reactions are also considered ADRs.¹¹ Not technically classified as ADRs are side effects, which are defined by ASHP as “an expected, well-known reaction resulting in little or no change in patient management (e.g., drowsiness or dry mouth due to administration of certain antihistamines or nausea associated with the use of antineoplastics)” and that has “a predictable frequency and an effect whose intensity and occurrence are related to the size of the dose.” Also not categorized as ADRs are “drug withdrawal, drug-abuse syndromes, accidental poisoning, and drug-overdose complications.”¹¹

Following the thalidomide disaster of the 1950s, WHO established and has maintained an international adverse drug events reporting system. Since 1978, the WHO International Drug Monitoring Program (IDMP) has collected ADR reports from 60 participating nations, and now includes both pharmaceutical and botanical medicine information. Of the more than 2.5 million reports in their International Drug Information System (INTDIS) database, approximately 10,000 relate to herbal medicines, primarily involving multiple ingredients.¹² This demonstrates a remarkably low incidence of serious ADRs resulting from herbal products, especially when compared with that of pharmaceutical drugs and the much larger worldwide consumption of botanicals compared with conventional medications. Many of the reports were associated with negative interactions with conventional medications and most side effects were reported for people 60 to 69 years of age.

A retrospective review of adverse effects reports (AERs) due to herbal medicines made to the National Poisons Unit of the United Kingdom was conducted in 1991. Between 1983 and 1989 a total of 1070 inquiries were made. Twenty-five percent of reports were of subjects with acute symptoms. Most confirmed adverse events were due to herbal sedatives.¹³ Ernst conducted a survey of complementary and alternative medicine users in the United Kingdom. Of those who had reported use of herbal medicines, 8% reported having observed adverse effects, none of which were considered serious.¹³

Adverse events associated with dietary supplements are not effectively monitored in the United States.¹⁴ The main AER systems are local and national poison control centers and the FDA’s MedWatch program, all of which have serious limitations. Most reporting systems, including these, are passive systems with no criteria for submission, no verification of the authenticity or accuracy of reports, and no effective follow-up or investigation. In order to be meaningful, AERs must be critically reviewed and products analyzed. Yet, neither a systematic review of the patient or event is conducted, nor is the product involved in an event typically analyzed, making it impossible to establish a causal relationship between an event and an herb/herbal product. Further, recording practices are highly variable between systems and between centers within the same reporting system and there is often selection bias.¹⁴

A relatively recent survey of data from US poison control centers evaluated the incidence of AERs for all dietary supplements.¹⁴ In a 10-month period, 1466 reports of potential events due to ingestion of supplements were made. More than one-third of these (534) were unintentional ingestions; 471 reported no symptoms. Half (741) of the total reports produced symptoms. The reviewers reported that 66% (489) were associated with dietary supplement exposure with varying degrees of confidence and 27% (132) were considered to be either definitely or probably related, whereas 34% (166) were reported as definitely or probably unrelated. The balance of reports (39%; 190) represented a 50% chance for the event to be correlated to supplement use. Ninety of the subjects used supplements in an attempted suicide, whereas the others used supplements for various reasons, ranging from the treatment of specific conditions (35%) to enhanced athletic performance (10%) and dieting (14%), among others. Botanicals most frequently associated with adverse events included ma huang (Ephedra sinensis), St. John’s wort (Hypericum perforatum), guarana (Paullinia cupana), and ginseng (species not identified). Of a total of 401 calls specifically related to adverse events of dietary supplements, 286 events were classified as mild, 89 as moderate, and 22 as severe, with a total of four deaths. Subjects experiencing symptoms were taking as many as 44 ingredients concurrently and had more serious adverse events than did those taking fewer supplements long-term. However, 89% of these adverse events were considered mild (57%) or moderate (32%); 11% were reported as severe or resulted in death. Most acute reports (95%) were considered mild (77%) or moderate (18%), and 4% severe or death. Many of these reports suffer from the same limitations of not being subjected to critical review or product characterization. Of all the supplement adverse events reported, only 12 products were retrieved for analysis. In comparison, a total of 61,229 calls regarding adverse events due to other consumed substances, mostly pharmaceuticals, were made in the same time period.

CATEGORIES OF COMMON HERBAL ADVERSE EFFECTS

The Toxicologic Medical Unit of Guy’s and St. Thomas Hospital in London conducted a 5-year toxicologic review of adverse events caused by herbal drugs based on reports made to the National Poison Control Center.¹⁵ Drowsiness and dizziness followed by vomiting, diarrhea, abdominal pain, and nausea were the most commonly reported side effects. Other events of note included agitation and irritability, cardiac arrhythmias, psychological disturbances, and facial flushing. Interaction of botanical and conventional anticoagulants has resulted in reports of abnormal bleeding, and effects of interactions with herbs that rely on the cytochrome P450 enzyme system have resulted in elevations or decreases in serum drug concentrations affecting clearance times, efficacy, and toxicity. Reports of liver abnormalities were most commonly associated with Chinese herbal medicines.

Botanical Quality Standards and Reactions Caused by Contamination and Adulteration of Herbs

The safety of any medicinal product is partly dependent on the manner in which the product is produced. Although individuals, organizations, industry, and government are actively working to establish national quality control (QC) standards, current standards are not a guarantee for herbal products in the United States, or those imported from countries such as China and India. European herbal product standards are generally stricter than most other countries, including the United States; thus, European herbal products have a greater likelihood of safety and quality. The most important aspects of QC include the accurate identification, relative purity, and relative quality of raw material used in herbal products (Tables 4-3 and 4-4).

TABLE 4-3 Potential Toxic Plant Adulterations Causing Safety Concerns

TABLE 4-4 Potential Contaminants of Herbal Products

* Current law prohibits the presence of toxic microorganisms in herbal products. Contaminated products can be removed from the market. To date, there have been no published reports of herbal products toxicity from such microorganisms.

** Although some of these pesticides have been banned in the United States for decades, they may still be in use in other countries. Conversely, some pesticides may be approved for use in common foods but lack approval for use on herbal products.

† These sterilization processes, although approved for use in certain foods, are not approved for use on herbal products.

‡ Heavy metals in food and water is a problem worldwide but have been specifically noted to occur in numerous herbal medicine products manufactured in China.

§ It is prohibited to combine conventional drugs in products marketed as dietary supplements.

The safety of plants may also be affected by the presence of contaminants not naturally occurring in the plants, including heavy metals (taken up from the soil naturally, resulting from pollution, or their addition as part of traditional formulation or processing, as may be the case with TCM or Ayurvedic preparations and prescriptions), microbial contamination (owing to animal feces or poor handling under unhygienic conditions), or processes that are applied to the plant (e.g., sterilization with ethylene oxide gas, pesticides, fungicide, or gamma irradiation). By law, herbal products that enter the market must be safe for consumption as a food and be free of toxic microbial contamination. Similarly, certain treatments, such as with pesticides and irradiation, are not permitted on herbal products unless expressly approved. Nonetheless, because herbs are obtained from a variety of sources worldwide, many of which routinely use such treatments, herbal products may contain one or more of these contaminants.

Most herbs and botanical products are not tested before importation into the United States. It is impossible for consumers and health professionals to determine if such contaminations are present without subjecting the product to extensive testing. Again, any presence of a known pathogen (e.g., Escherichia coliform) or prohibited treatment (certain pesticides or gamma irradiation) renders a product adulterated and subject to removal from the market by the FDA. However, regulatory oversight in this area is almost nonexistent. The best way for practitioners to avoid potentially contaminated or adulterated products is to become knowledgeable of the practices of suppliers and manufacturers, and identify those with high quality standards (see Identifying Quality Botanical Medicine Products). Companies that are run by or employ qualified herbalists on their staff and that serve the needs of health professionals are often the most reliable, but this in itself is not a guarantee of quality.²⁰

Traditional Chinese Herbal Medicines Mixed with Conventional Pharmaceutical Drugs

It is not uncommon for Chinese herbal products made in Asia, including those sold in the United States, to be adulterated with conventional pharmaceutical drugs.²¹ Adulterants have included caffeine, paracetamol, indomethacin, hydrochlorothiazide, prednisolone, barbiturates, and corticosteroids.²² The purpose for the inclusion of pharmaceutical drugs in botanical products is presumed increased efficacy. However, inclusion of such substances is not consistent with principles of traditional Chinese medicines. In a survey evaluating herbal products used in hospitals in Taiwan, approximately 24% of 2406 herbal drugs tested contained pharmaceuticals.

Significantly, in most cases the pharmaceutical adulterants are not disclosed on the product label or packaging. Herbal products adulterated with conventional medications have been found to be manufactured in Mainland China, Hong Kong, and Taiwan.²¹ In the United States, under dietary supplement regulations, such products are considered to be adulterated and are subject to removal from the market. Thus far, there have been no reports of the addition of pharmaceuticals in domestically made Chinese herbal products. Although there are undoubtedly high-quality herbal products manufactured in Asia, it is currently almost impossible for consumers and most health professionals (even trained TCM practitioners) to differentiate between adulterated and nonadulterated products. Thus, patients using some traditional Chinese herbal medicine products in combination with conventional medications may be also be unknowingly subjected to unexpected drug–drug interactions via a pharmaceutical adulterant.

Ayurvedic Herbs and Heavy Metal Contamination

In traditional Ayurvedic medicine, the addition of metals is common in selected herbal preparations. These are considered to contribute specific desired activities and are processed in a manner in which it is believed to reduce their toxicity, although this latter assertion has not been critically documented. Additionally, if indeed processing does reduce the toxicity of heavy metals, there is no guarantee that all manufacturers apply these techniques appropriately. The prevalence of heavy metals in Ayurvedic preparations has been reported.

In a study conducted by Saper et al., it was found that one of five Ayurvedic herbal products produced in South Asia and sold in Boston-area South Asian grocery stores contained potentially harmful levels of lead, mercury, and/or arsenic. If taken as recommended by the manufacturers, 20% of products could result in heavy metal intakes above published regulatory standards.²³ The products selected, however, were not necessarily representative of typical Ayurvedic products, as most included heavy metals as ingredients, whereas the majority of Ayurvedic herbal products do not. The Indian government has recently made testing for heavy metals compulsory and labeling for heavy metals within permissible limits mandatory for Ayurvedic medicine products produced in India and destined for export beginning January 1, 2006, as part of an attempt to address this problem.²⁴

Reports of Interactions

The literature consists mainly of case reports and pharmacologic data that have not been assessed for clinical relevance.²⁷ Patients or physicians make most individual case reports, usually after having experienced or observed an unexpected adverse reaction. Case reports, which typically are not subjected to critical review, are among the most unreliable body of information from an evidence-based perspective. There have been a few critical reviews that provide some meaningful guidance regarding herb–drug interactions (Box 4-2). Researchers Izzo and Ernst conducted a systematic review of herb–drug interactions of seven of the most popularly used botanicals: garlic, ginkgo, St. John’s wort, saw palmetto, kava kava, Asian ginseng, and echinacea. Clinical data were collected from standard databases, recent articles and books, and interviews with herbal product manufacturers,¹⁰ herbal experts,⁸ and organizations related to medical herbalism.²⁴ In a 21-year period (1979–2000), a total of 41 case reports or case series in 23 publications and 17 clinical trials were obtained. Of these, only 12 interactions could be considered life threatening (mostly associated with a dangerous reduction in blood plasma levels of cyclosporine in organ transplant patients using St. John’s wort) owing to a putative effect on cytochrome P450 systems.²⁸

The paucity of incidences of herb–drug interactions for these seven botanicals should also be viewed within the context of their total use. Definitive data regarding actual herb use are lacking, but available published marketing reports for ginkgo, kava, and St. John’s wort provide some guidance. Ginkgo is by far the most widely used and prescribed medicinal herb in the world. According to a review of DeFeudis, from 1979 to 1998, approximately 110 million daily doses of ginkgo leaf extract were sold in France alone. According to market data of Dr. Willmar Schwabe GmbH, the makers and world’s leading marketer of the most clinically tested ginkgo extract (EGb 761), in the period from 1979 to 1997, it was estimated that more than 2 billion daily doses of ginkgo extract were distributed worldwide. A more recent report estimates that from 1997 to 2002, 1.375 billion daily treatments of ginkgo single preparations have been administered in Germany alone. Up until 2003, a total of only four case reports of potential interactions between ginkgo and conventional medications had been published.²⁹ Two of these were associated with bleeding events. This has led to significant concerns raised regarding its potential to increase the effects of anticoagulants. However, such events, if causal, are rare. More than 40 clinical trials with several thousand subjects have been conducted using the EGb 761 ginkgo extract with no reports of increased bleeding. A number of studies have specifically looked at its potential for increasing the anticoagulant effects of aspirin and warfarin with mostly negative results; it has been included in pharmacovigilance reviews throughout Europe for decades. Nonetheless even in cases in which an interaction has not been definitively confirmed, caution should be advised.

As of 2001, St. John’s wort was the primary antidepressant used throughout Europe with approximately 66 million daily doses in Germany, outselling most conventional antidepressants. Up until the 2000 review of Izzo and Ernst, a total of 29 potential interactions were reported. Most of those were associated with decreased blood cyclosporine levels due to enhanced hepatic enzyme induction (cytochrome P450). Others were associated with an additive effect of St. John’s wort wi selective serotonin reuptake inhibitors (SSRIs). Again, considering the prevalence of St. John’s wort use, such incidences of adverse effects are rare. With increased understanding that St. John’s wort does affect P450 enzyme systems, such interactions can be more readily predicted and therefore minimized (see Drugs Metabolized via Cytochrome P450 System).²⁸

Categories of Herb–Drug Interactions

Use of Herbs Prior to Surgery

A number of botanicals should be avoided, or used with care, prior to or when undergoing surgery, largely due to theoretical or known risks of interactions with coagulation therapies or anesthesia. Patients must also be queried regarding their use of botanicals at this time. As noted, there are a number of botanicals with blood-thinning potential, such as ginkgo, garlic, dong quai, and red clover, whose use is so common that patients might not report these in a standard medical history intake. There are also botanicals that have been reported to interact with general anesthesia, most notably, St. John’s wort, which may intensify or prolong the effects of anesthesia. The American Society of Anesthesiologists (http://www.asahq.org/patientEducation/herbal.htm) recommends that patients stop taking herbal supplements 2 to 3 weeks prior to surgery because they may affect anesthesia and bleeding times, and cause dangerous fluctuations in blood pressure. If there is not enough time prior to surgery to stop, patients are recommended to bring the products to their primary care physician so an assessment of any danger can be made.

Use of Herbs in Pregnancy and Lactation

There are many conditions associated with pregnancy for which expectant mothers seek natural therapies, sometimes in the hopes of avoiding what they perceive to be more toxic conventional medications. Such conditions include morning sickness, threatened miscarriage, vaginal infections, anemia, varicose veins, hemorrhoids, depression, anxiety, and sleeplessness. One survey of emergency room visits reported that 14.5% of women used herbal remedies during pregnancy.⁴² In addition to unsupervised herb use by consumers, approximately 50% of midwives and naturopathic physicians routinely prescribe herbal medicines during pregnancy, including a prominent use of botanicals such as black and blue cohosh (Actaea racemosa, Caulophyllum thalictroides), castor oil (Ricinus communis), and evening primrose oil (Oenethera biensis) for inducing labor. Considering whether to use an herbal medicine during pregnancy requires skill and sound judgment on the part of the practitioner, or in the case of self-medication, careful consideration on the part of the consuming woman. Minimally, the relative health of the expectant mother, the indication, and the appropriateness or inappropriateness of other medications or therapies must be considered. Ultimately, with herbal medicines, the choice of medications should be made by the expectant mother with full knowledge of the potential consequences. There are also significant legal liabilities that are incurred when using herbs in pregnancy. A list of botanicals that should not be used in pregnancy except under the care of a qualified health care professional is provided in Chapter 12, where the special considerations of herb safety during pregnancy are extensively addressed in Chapter 12.

Emergency Room Personnel

Emergency room attendants should be aware of the conditions they are most likely to encounter in patients who have reported a reaction thought to be correlated with consumption of an herbal product. These include the following:

Oncologists

Use of herbal medicines is prevalent among cancer patients. The seriousness of the disease leads patients to seek a variety of therapies regardless of the level of evidence supporting their use. Unfortunately, there are few well-designed trials examining the usefulness of botanical supplements for the direct treatment of cancer or in conjunction with conventional therapies, and what information exists is conflicting. Most commonly used are herbal therapies to enhance immune function, directly combat cancer, and yet others to provide palliative relief from side effects associated with conventional cancer therapies. There are two primary safety considerations regarding botanical therapies in cancer: (1) ensuring that potentially ineffective therapies are not substituted for potentially effective conventional therapies, and (2) making sure that botanical therapies do not lessen the effects of conventional therapies. Very few data regarding either exist.

Botanicals That Should Only Be Used by Qualified Health Professionals

The majority of botanicals that have remained on the commercial market have persisted because of their relatively high degree of safety when used by consumers without the guidance of a health professional. Most do not present a significant enough health risk to warrant prescription-only status. However, historically, relatively toxic botanicals (e.g., aconite, digitalis, and gelsemium) have been used by medical herbalists. Most of these have not remained in the commercial market as dietary supplements. The use of toxic such herbs should be limited to well-trained medical herbalists.⁴⁸ Recommendations regarding the use of botanicals that should only be used by skilled individuals can be found in the Botanical Safety Handbook (CRC Press).

ADDRESSING ETHICAL CONSIDERATIONS

Jeremy Sugarman, in the JAMA article, Physicians’ Ethical Obligations Regarding Alternative Medicine, suggests that physicians facing ethical dilemmas about integrating CAM into their practices, or at least trying to decide how to address CAM use by patients, consider applying a broad “set of inherent ethical principles of the medical profession: respect for persons, nonmaleficence, beneficence, and justice” into their decision-making process.⁵⁴ These principles, well known to physicians, are:

GUIDELINES

Once practitioners have decided that their patients have an ethical right to information on the safe and effective use of botanical therapies, or possibly even botanical products, how are they to advise patients about them? Wayne Jonas suggests that health practitioners follow a simple set of practical rules he calls the “four Ps”: protect, permit, promote, and partner.⁵⁵ He defines these as follows:

Karen Adams, in Ethical Considerations of Complementary and Alternative Medical Therapies in Conventional Medical Settings, suggests consideration be given to the following: (1) whether evidence supports both safety and efficacy; (2) whether evidence supports safety but is inconclusive about efficacy; (3) whether evidence supports efficacy but is inconclusive about safety; or (4) whether evidence indicates either serious risk or inefficacy.⁵³ She advises the following steps as guidelines:

The risk–benefit analysis shown in Box 4-7 should be considered when there is insufficient evidence for or against a particular treatment.

There are numerous medical practices that were once considered fringe, such as biofeedback, that are now a routine part of conventional medicine. As practitioners gain increased experience and confidence with a modality, and as both clinical and pharmacologic studies are done that continue to demonstrate the safety and efficacy of the modality, the less fringe a practice seems. Botanical medicines have always been part of human and medical history; their use should not seem entirely foreign today.

CHARACTERISTICS OF QUALITY BOTANICAL PRODUCTS

Aviva Romm