Essentials of Complementary and Alternative Medicine (June 1999)

Step 1. The patient really wants to try acupuncture. She has tried smoking cessation classes and nicotine gum and patches without success. She wants to have a 
baby and stop smoking before she becomes pregnant.
Step 2. If properly delivered with disposable needles, ear acupuncture is unlikely to produce serious side effects. The cost and time commitment is considerable, but 
so is continuing to smoke.
Step 3. Her physician offers to search the literature. He logs onto the Internet and goes to the Cochrane Database of Clinical Trials in Complementary Medicine 
(available free on the Web page of the NIH National Center for Complementary and Alternative Medicine). Using the key words “smoking” and “acupuncture,” he finds 
27 controlled trials and 1 systematic review (
30
).
Step 4. The systematic review (published in 1990) is a criteria-based quality review of 15 trials. The review reports that the vast majority of these trials were small, of 
poor quality, and reported negative results. The French study provided by the acupuncturist was also on-line and, although it reports positive effects from 
acupuncture, it received a very poor score in the systematic review. A more recent study (1997) evaluated smoking cessation rates in groups given acupuncture, 
nicotine gum, both, or neither. About 6% remained off tobacco at 4 years, with no difference between any treatment (
31
).
Steps 5 through 9 are unnecessary because the current best evidence indicates that acupuncture is no more effective than placebo or nicotine gum for smoking 
cessation. The patient should be informed that using the acupuncturist for this purpose may be a waste of time and money. Once the patient's initial enthusiasm for 
the acupuncture treatments has subsided, a relapse to smoking is highly probable. Provided the acupuncture is delivered in a proper manner with sterile needles, the 
patient's main risk is the cost and time of the treatments as well as potential resignation from future attempts at smoking cessation should she return to smoking. This 
might prevent her from enrolling in a comprehensive smoking cessation program, which may increase her chances of success in smoking cessation. Thus, an 
additional risk of acupuncture in this case is substituting ineffective therapy for possible effective therapy.
C
HAPTER
 R
EFERENCES
1.
Stolberg M. Die Homoeopathie auf dem Pruefstein. Der erste Doppelblindversuch der Medizingeschichte im Jahr 1835. Muenchner Medizinische Wochenschrift 1996;138:364–366.
2.
Evidence-Based Medicine Working Group. Evidenced-based medicine: a new approach to teaching the practice of medicine. JAMA 1992;268:2420–2425.
3.
DiGiacomo S. Biomedicine as a cultural system: an anthropologist in the kingdom of the sick. Encounters with Biomedicine 1987:315–346.
4.
Collins HM. In: Collins HM, ed. Sociology of scientific knowledge: a source book. Bath, England: Bath University Press, 1982.
5.
McKinlay JB. From “promising report” to “standard procedure”: seven stages in the career of a medical innovation. Health and Society 1981;59:374–411.
6.
Berg AO. Clinical practice policies: believe only some of what you read. Fam Pract 1996;3:58–70.
7.
Slawson DC, Shaughnessy AF, Bennett JH. Becoming a medical information master: feeling good about not knowing everything. J Fam Pract 1994;38:505–513.
8.
Wennberg J. Dealing with medical practice variations: a proposal for action. Health Aff (Millwood) 1984;3:6–32.
9.
Chassin MR, Brook RH, Park RE, et al. Variations in the use of medical and surgical services by the Medicare population. N Engl J Med 1986; 314:285–290.
10.
Eddy DM. Clinical decision making: from theory to practice. A collection of essays from JAMA. Boston: Jones & Bartlett Publishers, 1996.
11.
Ellenberg JH, Nelson KB. Sample selection and the natural history of disease. Studies of febrile seizures. JAMA 1980;243:1337–1340.
12.
Feinstein AR. Clinical judgment revisited: the distraction of quantitative models. Ann Intern Med 1994;120:799–805.
13.
Linde K, Worku F, Stör W, et al. Randomized clinical trials of acupuncture for asthma–a systematic review. Forsch Komplementärmed 1996;3: 148–155.
14.
Jonas WB. Alternative medicine. J Fam Pract 1997;45:34–37.
15.
Linde K, Clausius N, Ramirez G, et al. Are the clinical effects of homeopathy all placebo effects? A meta-analysis of randomized, placebo controlled trials. Lancet 1997;350:834–843.
16.
Physicians for the 21st century: the GPEP Report. Washington, DC: Association of American Medical Colleges, 1984 and 1994.
17.
Haynes RB, Sackett DL, Gray JM, et al. Transferring evidence from research into practice: 1. The role of clinical care research evidence in clinical decisions. ACP J Club 1996;125:A14–16.
18.
Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York: Churchill Livingstone, 1997.
19.
Gill P, Dowell AC, Neal RD, et al. Evidence-based general practice: a retrospective study of interventions in one training practice. BMJ 1996; 312:819–821.
20.
Haynes RB, Sackett DL, Gray JA, et al. Transferring evidence from research into practice: 2. Getting the evidence straight. ACP J Club 1997;126:A14–16.
21.
Schneider CJ, Jonas WB. Are alternative treatments effective? Issues and methods involved in measuring effectiveness of alternative treatments. Subtle Energies 1995;5:69–92.
22.
Wiesenauer M, Ludtke R. A meta-analysis of the homeopathic treatment of pollinosis with galphimia glauca. Forschende Komplenemtarmedizin 1996;3:230–234.
23.
Jonas WB. Clinical trials for chronic disease: randomized, controlled clinical trials are essential. J NIH Res 1997;9:33–39.
24.
Frank JD. Non-specific aspects of treatment: the view of a psychotherapist. In: Shepherd M, Sartorius N, eds. Non-specific aspects of treatment. Bern: Huber, 1989:95–114.
25.
Beecher HK. The powerful placebo. JAMA 1955;159:1602–1606.
26.
Benson H, Epstein MD. The placebo effect: a neglected asset in the care of patients. JAMA 1975;232:1225–1227.
27.
Moerman DE. General medical effectiveness and human biology: placebo effects in the treatment of ulcer disease. Med Anthropol Q 1983;14(3): 13–16.
28.
Roberts AH, Kewman DG, Mercier L, Hovell M. The power of nonspecific effects in healing: implications for psychological and biological treatments. Clin Psychol Rev 1993;13:375–391.
29.
Circo A, Tosto A, Raciti S, et al. First results of an anti-smoke outpatient unit: comparison among three methods. Riv Cardiol Prev Riabil 1985; 3:147–151.
30.
ter Riet G, Kleijnen J, Knipschild P. A meta-analysis of studies into the effect of acupuncture on addiction. Br J Gen Pract 1990;40:379–382.
31.
Clavel CF, Paoletti C, Benhamou S. Smoking cessation rates 4 years after treatment by nicotine gum and acupuncture. Prev Med 1997;26:25–28.

CHAPTER 6. T
HE
 S
AFETY OF
 H
ERBAL
 P
RODUCTS
Essentials of Complementary and Alternative Medicine
CHAPTER 6. T
HE
 S
AFETY OF
 H
ERBAL
 P
RODUCTS
Peter A.G.M. De Smet
Introduction
 
Nontraditional Hazards
 
Excipients
Consumer-Related Determinants of Adverse Effects
Situational Risks
Adverse Effects
 
Individual Herbs
 
Recreational Herbs
 
Traditional Chinese Medicines
 
Individual Herbs
 
Traditional Indian and Pakistani Medicines
Chapter References
INTRODUCTION
A consumer of an officially licensed herbal medicine may not have to be concerned about the correct identity of the ingredients, but an individual who goes into the 
field to collect his or her own herbs should. Austrian physicians described a case of a young boy who developed venoocclusive disease of the liver after long-term 
consumption of a tea prepared from Adenostyles alliariae. The boy's parents had erroneously gathered this plant instead of coltsfoot ( Tussi-lago farfara), and A. 
alliariae contains a much higher level of hepatotoxic pyrrolizidine alkaloids than does coltsfoot (
1
, 
2
). Botanical identity can also be problematic within the context of 
commercially available materials. For example, German researchers exposed that Sarothamni scoparii flos does not always originate from  Sarothamnus scoparius 
(Besenginster) but may also come from Spartium junceum (Spanischer Ginster) (
3
). This adulteration could be clinically relevant because the flowers of  S. junceum 
are rich in cytisine-type quinolizidine alkaloids (
4
, 
5
).
The botanical quality of prepackaged herbal products may also cause problems, especially in countries (such as the United States) that do not categorize these 
products as medicines (
6
). As a result, these products remain exempt from governmental approval processes, and their quality may remain essentially uncontrolled (
7
, 
8
). An example in the United States concerned a South American product labeled as “Paraguay Tea.” This product was associated with an outbreak of anticholinergic 
poisoning. On chemical analysis, the product yielded belladonna alkaloids instead of the xanthine derivatives that were expected in a preparation from  Ilex 
paraguarariensis (
8
).
Herbal products should be free not only from toxic botanical adulterants, but also from other contaminants (
Table 6.1
) (
9
), such as substantial residues of pesticides 
(
Table 6.2
) (
10
). The need to prevent contamination with pathogenic microorganisms was illustrated by the case of a bone marrow transplant recipient, who probably 
acquired hepatic mycosis from the ingestion of a naturopathic medicine contaminated with a  Mucor fungus (
11
). There is also evidence that medicinal plant materials 
from India and Sri Lanka can be contaminated with toxigenic fungi ( Aspergillus, Fusarium). Because aflatoxin B has sometimes been recovered from these materials 
in potentially unsafe amounts, it is prudent to improve their storage conditions (
9
, 
12
, 
13
).
Table 6.1. Potential Contaminants to Account for in Quality Control of Herbal Medicines
Table 6.2. Pesticide Residue Limits in Herbal Medicines*
Another practical concern is the presence of toxic metals (e.g., lead, arsenic) or conventional pharmaceuticals (e.g., corticosteroids, nonsteroidal antiinflammatory 
drugs, benzodiazepines) in certain herbal medicines of Asian origin. These hazards have been denounced for more than two decades, but they continue to pose an 
occasional threat to public health (
9
). Although most of the recent reports on the undeclared presence of Western pharmaceuticals involve Chinese herbal medicines, 
the contamination of herbal medicines with pharmaceuticals is not necessarily limited to products of Oriental origin. We analyzed Dutch herbal drops for weight 
reduction that were declared to contain  Ephedra and 14 other ingredients because a Dutch professional cyclist had tested positive for norpseudoephedrine at a 
doping control urinalysis. The level of norpseudoephedrine in the investigated product was substantially higher than that of ephedrine, which is normally not the case 
in Chinese Ephedra plants. The manufacturer later admitted that his product had been spiked (
14
).
Although the safety of herbal medicines can be compromised by deficient product quality, some herbal products become more dangerous when they have excellent 
quality. Yohimbe products rich in yohimbine will be less safe for over-the-counter use than products containing no or negligible amounts of this alkaloid (
15
, 
16
).
Nontraditional Hazards
Safety claims cannot always be based on long-standing traditional herbal experience: not all herbal medicines have firm roots in traditional practices, and this seems 

an underestimated issue. When traditional source plants are extracted in a nontraditional way (e.g., by resorting to a nonpolar solvent, such as hexane), the question 
can be raised whether this nontraditional extract is just as safe as the traditional one. Until recently, the ostrich fern ( Matteuccia struthiopteris) was generally 
considered a nontoxic, edible plant with a history of use as a spring vegetable that went back to the 1700s. However, recent observations of serious gastrointestinal 
toxicity following the consumption of lightly sauteed or blanched ostrich fern shoots suggest that this vegetable is safe only when thoroughly cooked before use (
17
). 
A similar example is the recent outbreak of bronchiolitis obliterans in Taiwan, which was associated with the ingestion of  Sauropus androgynus. This herb normally is 
cooked before being eaten as a vegetable, but in this case the numerous victims had all consumed uncooked leaf juice as an unproven method of weight control (
18
).
It is also possible that an herbal ingredient may have no medicinal tradition at all, and its route of administration or dose level may be quite different from that used in 
a traditional setting. The question could be raised, for instance, to which extent the excellent oral safety record of certain traditional herbs is applicable to the herbal 
cigarettes, which are nowadays available in Western health food stores. After all, there is evidence to suggest that certain respiratory risks attributed to tobacco 
smoking may extend to the smoking of nontobacco herbal products, particularly marijuana (
19
, 
20
, 
21
, 
22
, 
23
 and 
24
).
Excipients
An inconspicuous source of adverse reactions to herbal medicines are the excipients (
25
). This has been illustrated by the careful evaluation of a case of contact 
dermatitis caused by a commercial ointment containing a  Centella asiatica extract. Instead of automatically assuming that this ingredient was responsible, the 
investigators decided to seek confirmation by patch testing with the individual components. It was thus discovered that the reactions were caused mainly by the 
presence of propylene glycol. A true allergic response to the botanical components remained unproven (
26
). Lanolin is among the notorious contact allergens that can 
be present as excipient in topical applications (
27
). In an evaluation of the contact sensitization potential of five commercial herbal ointments in 1032 consecutive or 
randomly selected visitors to patch test clinics, 2 of the 11 patients with a positive response had been sensitized by lanolin (
28
). There have also been reports about 
the presence of pesticide residues in lanolin (
29
, 
30
).
Another noteworthy excipient is alcohol. Psychological and physical dependence on herbal medicines with a high alcohol content seem rare, but they have been 
reported (
31
). There is also an association between a case of suspected fetal alcohol syndrome with a history of maternal ingestion of a herbal tonic (containing 14% 
alcohol) daily for the first two months (
32
). The U.S. Food and Drug Administration has ruled that the alcohol concentration in over-the-counter drug products intended 
for oral ingestion should not exceed 10% for adults and children over 12 year of age, 5% for children between 6 and 12 years of age, and 0.5% for children under 6 
years of age (
33
).
CONSUMER-RELATED DETERMINANTS OF ADVERSE EFFECTS
The chance of an adverse reaction to an herbal medicine depends not only on the product's actual composition and manner of use, but also on consumer-bound 
parameters, such as age, genetics, and concomitant diseases. For instance, the risk that the alkaloid berberine in Chinese  Coptis spp. elicits jaundice seems to be 
most substantial in infants who are deficient in glucose-6-phosphate dehydrogenase (
34
, 
35
 and 
36
). Another example is that slow metabolizers of the quinolizidine 
alkaloid sparteine will be more prone to the oxytocic potential and other toxic effects of  Cytisus scoparius than rapid metabolizers (
35
, 
36
 and 
37
). This latter example 
illustrates the general principle that pharmacokinetic information can help to predict certain types of consumer-dependent adverse effects. However complex the 
composition of an herbal medicine may be, its constituents are chemical entities which, besides having pharmacodynamic properties, must obey the same 
pharmacokinetic rules that apply to conventional drug molecules. In other words, insight into the ways in which the pharmacokinetics of complementary medicines are 
modified by factors such as hepatic or renal insufficiency provides a rational tool for predicting and avoiding dose-related adverse effects (
38
).
Another intriguing finding is that Chinese subjects are more sensitive to the effect of atropine on heart rate than are Caucasian subjects, who in their turn seem to be 
more sensitive than Black individuals (
39
). A theoretical implication of such interracial differences is that local experience with a traditional herbal remedy cannot 
always be extrapolated indiscriminately to societies with another ethnic make-up.
Concurrent use of other drugs must also be considered. For instance, consumers of caffeine-containing herbs ( Cola, Ilex, and Paullinia) may have an increased risk of 
adverse effects, such as tremors or tachycardia, when they concurrently ingest pipemidic acid, ciprofloxacin, or enoxacin, because these antibacterial quinolones 
inhibit the hepatic metabolism of caffeine (
40
, 
41
, 
42
 and 
43
). Likewise, the oxidative metabolism of sparteine, a quinolizidine alkaloid in Scotch broom ( Cytisus 
scoparius), can be inhibited by the simultaneous intake of haloperidol (
44
), moclobemide (
45
), or quinidine (
46
). This latter example shows that the interfering drug 
does not have to be a synthetic compound but can also come from nature. According to a comment on the numerous adverse events associated with Ma Huang 
(Ephedra) consumption in the United States, the adverse effects of combined ephedrine and caffeine may be greater than those from the consumption of either 
compound alone (
47
). Besides the possibility that the toxicity of an herbal compound is increased by a conventional drug, there is also the possibility that the effects 
of a conventional medicine are enhanced by an herbal medicine. For example, the Indian herbal drug karela ( Momordica charantia) has hypoglycemic properties (
48
) 
and can thereby interfere with conventional antidiabetic treatment (
49
).
For an extensive overview of adverse drug interactions between conventional medicines and herbal products, the reader is referred to a publication elsewhere (
50
).
SITUATIONAL RISKS
Herbal medicines that are generally safe under normal conditions can be hazardous in specific circumstances. For example, psoralen-rich herbal preparations may 
produce phototoxic burns in visitors to tanning salons or in patients undergoing PUVA therapy (
51
, 
52
). Therefore, it is important to provide the consumer of herbal 
medicines with adequate product information, in which such situational hazards are clearly indicated.
Among the safety issues that need attention in this respect is the risk that herbal products make their consumers less fit for driving. This hazard should be taken into 
consideration for any plant-derived drug with central depressant activity, such as  Rauvolfia (
53
) and tetrahydropalmatine (
54
).
An even more inconspicuous risk is that top athletes may unwittingly take a doping agent in the form of a herbal product because the list of doping substances, which 
is issued by the International Olympic Committee, comprises several substances that occur naturally in herbs (
55
). Among these doping agents are ephedrines and 
caffeine, which are regularly present in Western health food preparations (
56
, 
57
 and 
58
).
ADVERSE EFFECTS
Individual Herbs
The following provides an overview of adverse effects of herbal preparations that have been reported in the literature. Unless otherwise specified, the presented data 
have been derived from previous reviews (
59
, 
60
 and 
61
). For additional information about this subject, the reader is referred to a rapidly growing list of detailed 
textbooks on herbal medicines (
62
, 
63
, 
64
, 
65
, 
66
, 
67
, 
68
, 
69
, 
70
, 
71
, 
72
, 
73
, 
74
 and 
75
). Herbs, which are primarily used for recreational purposes, are reviewed 
separately, as are herbs primarily employed within the context of traditional Chinese and Indian medicine.
Sometimes, this review does not focus on the crude herb but on one or more components. This approach is acceptable because, in principle, the toxic potential of an 
herbal product does not depend on its natural origin but on the pharmacological characteristics and dose levels of its bioactive constituents. Although it is conceivable 
that the toxicity of an individual herbal constituent can be modified by one or more of the other constituents, this is by no means some iron-clad rule: a complex 
composition does not always protect the consumer against the toxicity of a single constituent, and it might also result in a more toxic remedy.
A potential hazard of most medicinal herbs is allergic contact dermatitis, but this risk is much more substantial for certain herbs (e.g., those containing sensitizing 
sesquiterpene lactones) than for others (
27
, 
76
). Another common adverse effect ascribed to various herbs is gastrointestinal disturbances (e.g., caused by the 
presence of tannins or irritating saponins). More serious problems can be elicited by herbs containing well-known botanical substances with toxic potential (e.g., 
cardiac glycosides, podophyllotoxin, reserpine). It should be noted that such classical drug substances may reside in a less familiar botanical source. For example, 
cardiac glycosides occur not only in such well-known source plants as foxglove ( Digitalis) and oleander ( Nerium oleander), but are also found in pleurisy root 
(Asclepias tuberosa) (
77
). Serious health risks are also possible with certain obsolete herbs, which were formerly employed in medicine but have now been 
superseded by other, less dangerous alternatives (e.g., chenopodium oil, male fern). In addition to such classical toxic agents, several herbs and herbal constituents 
have been repeatedly associated in recent years with new adverse effects of such a serious nature that their unrestricted internal use as phytotherapeutic agents no 

longer seems acceptable. Aristolochia, Teucrium, Larrea tridentata, and herbs rich in pyrrolizidine alkaloids all are part of this latter category.
The following is a summary of the known and reported adverse effects of medicinal herbs and products. When faced with a patient using these products, find the 
summary of known adverse effects of those products. This will provide an indication of the adverse effects known for that plant and help guide the practitioner in 
monitoring for possible adverse effects in that patient. It will also help the practitioner inform the patient about possible risks.