You are on page 1of 6

Journal of Ethnopharmacology 140 (2012) 513518

Contents lists available at SciVerse ScienceDirect

Journal of Ethnopharmacology
journal homepage:

Pharmacovigilance of herbal medicine

Shaw Debbie a, , Ladds Graeme b , Duez Pierre c , Williamson Elizabeth d , Chan Kelvin e,f
Chinese Medicine Advisory Service, Medical Toxicology Information Services, Guys & St Thomas NHS Foundation Trust, Guys Hospital, London SE1 9RT, UK
Pharsafer 29 Frederick Sanger Road, Guildford, Surrey GU2 7YD, UK
Universit Libre de Bruxelles CP 205/9, Bd du Triomphe, 1050 Bruxelles, Belgium
The School of Pharmacy, University of Reading, Whiteknights, Reading RG6 6AJ, Berkshire, UK
Centre for Complementary Medicine Research, University of Western Sydney, NSW 2560, Australia
Faculty of Pharmacy, The University of Sydney, NSW 2006, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Pharmacovigilance is essential for developing reliable information on the safety of herbal medicines
Received 31 October 2011 as used in Europe and the US. The existing systems were developed for synthetic medicines and require
Received in revised form 31 January 2012 some modication to address the specic differences of medicinal herbs. Traditional medicine from many
Accepted 31 January 2012
different cultures is used in Europe and the US which adds to the complexities and difculties of even
Available online 9 February 2012
basic questions such as herb naming systems and chemical variability. Allied to this also is the percep-
tion that a natural or herbal product must be safe simply because it is not synthetic which means that
the safety element of monitoring for such medicines can be overlooked because of the tag associated
Herbal medicine
with such products. Cooperation between orthodox physicians and traditional practitioners is needed
Drug safety to bring together the full case details. Independent scientic assistance on toxicological investigation,
botanical verication can be invaluable for full evaluation of any case report. Systematic pharmacovigi-
lance is essential to build up reliable information on the safety of herbal medicines for the development
of appropriate guidelines for safe effective use.
2012 Elsevier Ireland Ltd. Open access under CC BY-NC-ND license.

1. Introduction or pharmacovigilance), (iii) hypersensitivity, allergic and idiosyn-

cratic reactions (detectable by pharmacovigilance), (iv) mid-term
Pharmacovigilance is dened as the study of the safety of mar- and long-term toxic effects including liver, renal, cardiac and neuro-
keted drugs under the practical conditions of clinical usage in toxicity also genotoxicity and teratogenicity (detectable by in vitro
large communities (Mann and Andrews, 2002). The objective is to and in vivo toxicological studies or by pharmacovigilance). As
extend safety monitoring and detect drug adverse events that have many herbal products on the market have not been thoroughly
previously been unrecognised despite evaluation in clinical trials. tested for their pharmacology and toxicology, pharmacovigilance
Although these methods were developed for monitoring pharma- has paramount importance in detecting unwanted reactions.
ceutical medicines they are also used for additional evaluation of In addition there is an ongoing problem with unexpected toxi-
the safety of other medicinal products including herbals, blood city of herbal products due to quality issues, including use of poor
products, vaccines and even medical devices. quality herbal material, incorrect or misidentied herbs, incor-
As the use of herbal medicines has increased, so too have rect processing methods, supply of adulterated or contaminated
the reports of suspected toxicity and adverse events. Such herbs or products (Shaw, 2010a). These quality issues can be
unwanted reactions can be due to (i) side effects (usually addressed to some degree by improved regulation requiring GMP
detectable by pharmacodynamics and often predictable); (ii) reac- standards for manufacturing. However medicinal herbs/products
tions occurring as a result of overdose, overduration, tolerance, come from many countries with differing manufacturing standards
dependence-addiction (detectable either by pharmacodynamics and enforcement of regulations so poor quality products are likely
to remain a problem.
The safety of herbal medicines has become an issue for the reg-
ulatory authorities, as serious effects have been reported, including
Corresponding author.
hepatotoxicity, renal failure and allergic reactions (Perharic et al.,
E-mail addresses: Debbie (D. Shaw),
(D. Pierre), (W. Elizabeth), 1995; Nortier and Vanherweghem, 2007). The World Health Organ- (C. Kelvin). isation, recognising the growing importance of the use of herbal

0378-8741 2012 Elsevier Ireland Ltd. Open access under CC BY-NC-ND license.
514 D. Shaw et al. / Journal of Ethnopharmacology 140 (2012) 513518

medicines worldwide developed guidelines for the monitoring of is dened as any substance or combination of substances presented
herbal safety within the existing pharmacovigilance framework as having properties for treating or preventing disease in human
(WHO, 2004). beings. In contrast a food supplement cannot claim to treat or
prevent disease or contain a pharmacologically active substance.
2. Challenges of herbal pharmacovigilance This can be a complex area with the same herb being supplied as a
herbal medicine but also as an ingredient in a dietary supplement.
Herbal medicines in Europe come from all traditions including There are regulatory implications. In Europe, herbal medicines are
Chinese, Indian, north and south American and African systems as registered under two directives, either well-established use or
well as that of European systems. This diversity adds to the chal- traditional herbal medicinal products both of which have signi-
lenges of herbal pharmacovigilance including basic questions such cant requirements for quality (GMP) and safety (amongst others).
as dening the most appropriate herb naming system (botanical, Food supplements do not have the same legal requirements for
common, pharmaceutical name or herbal drug name) and valida- quality control.
tion of the botanical identity of the herbal ingredients. These are The denition of a herbal product as a food or medicine may
not normally an issue with monitoring synthetic medicines. Some therefore have considerable impact for pharmacovigilance.
of these questions, such as naming issues or adulterations, do not
t easily into the existing systems of pharmacovigilance or the
2.2. Nomenclature and what was used
electronic data systems that were developed for pharmaceuticals
(DynPage, UMC). However, although some modications may be
Adverse reaction reports, whether submitted to regulatory
needed, developing separate systems for herbals is not the answer
authorities or published in the medical literature, are meaning-
as this is likely to add complications and cause confusion if differ-
less if the medicinal herb(s) or ingredients in a product cannot be
ent forms or systems are used with the risk of reducing reporting
identied. Names for medicinal herbs include the Latin scientic
rates further (Menniti-Ippolito et al., 2008). As a good example of
name, the common or vernacular name, the pharmaceutical name
harmonisation, the Uppsala Monitoring Centre (UMC) takes ADR
or pharmacopoeal name or the specic herbal drug names (as used
reports from over 100 countries around the world and in 2010
in Traditional Chinese Medicine (TCM)). Herbal prescriptions, prod-
their database contained over 4 million reports, of which approx
uct packaging or labels may have one or more of these (occasionally
21,000 included herbal or natural products (UMC, 2011). These
no label) depending on source and regulatory status of the product.
are all incorporated in a single database with review of suspected
These have to be interpreted with care as even the scientic names
signals carried out by experts in relevant elds.
may have synonyms.
The common or vernacular name is the least precise and the
2.1. Specic challenges
same name may be used for plants from different genera or species
and so should be avoided if possible. In Europe and US however, the
Unlike synthetic medicines, herbal medicines are typically
common name is frequently the term of reference, although the EU
chemically rich and complex products and not isolated single com-
guidelines require the Latin scientic name on the labelling. If the
pounds. A number of factors can inuence the qualitative and
product is of pharmacopoeial quality then the identity of the species
quantitative chemical prole including:
and plant part will be dened in the European Pharmacopoeia
(PhEur). The common name may be misleading or confusing if used
Geographical origin climate, soil, photoperiod.
on crude plant material or unlicensed herbal medicines that are not
PhEur compliant. To avoid ambiguity it is desirable that the genus,
Parts of the plant leaves, stems, root, root bark, etc.
species and part of the plant are listed somewhere on the product
Harvesting time (year, season, time of day) and conditions.
or packaging of the crude material.
Storage, processing, extraction.
Unfortunately a botanically correct label does not necessarily
Combinations of herbs and/or processing of the combined herbs
conrm that the product contains what is listed on the label. In cases
as medicines.
of serious adverse reactions where specic toxins are suspected
then laboratory analysis of the product/herb may be advisable to
As standardisation of herbs relative to the active compounds is
verify the reports.
rarely performed, this intrinsic variability results in products that
may be quite different and not necessarily bioequivalent even when
from the same single herbal ingredients (Loew and Kaszkin, 2002). 2.2.1. Initiatives to address nomenclature and quality issues
Combining reports of adverse effects (or efcacy) requires careful There is currently no single reference list of medicinal plants
evaluation of the differences or similarities of the chemistry or bio- which presents an authoritative view on their current scientic
logical activity; however, if looking for safety signals for further name and linking all synonyms of those plants that are found in
investigation then a potentially useful approach might be to study the literature. The only names that are standardised are Latin sci-
groups of herbs containing similar compounds. entic names (e.g. Bupleurum chinense DC.); their standardisation is
achieved though the International Code of Nomenclature of algae,
2.1.1. Herbal medicines and dietary supplements fungi, and plants (ICN formerly ICBN).
The classication and regulation of herbal products may A new initiative, the Medicinal Plants Names Index (MPNI)
vary between different countries/jurisdictions. In the EU they underway at the Royal Botanic Gardens Kew will address
are classied as herbal medicines (regulatory implications) with this issue (
requirements for safety and quality standards. Some herbs may be projects/MPNI.htm). Working with a wide range of stakeholders,
supplied as food supplements. In the US, herbal products are clas- one of this projects main aims is to develop an authoritative index
sied as dietary supplements or botanicals, not medicines. Quality to scientic plant names mapped to frequently used vernacular,
will vary although GMP requirements were issued by the FDA in trade and pharmacopoeia names in order to support the develop-
2007. Pharmacovigilance reporting is not compulsory for manu- ment of global, industry-wide medicinal data standards. This is a
facturers. 3-year project (funded by Wellcome Trust) and will be accessed
There are subtleties in the legal differentiation between food freely over the internet with a range of other information services
supplement and herbal medicine. But broadly a medicinal product being developed tailored to different needs.
D. Shaw et al. / Journal of Ethnopharmacology 140 (2012) 513518 515

The UMC has worked with the Royal Botanic Gardens Kew The safety of herbs is mostly based on empirical experience and
and Uppsala University to develop appropriate standardisation and is effective in identifying acute toxicity with a rapid onset of symp-
cross-referencing of herbal names for their database to enable toms within hours or days of using any herbal medicines. However
international reporting of herbal ADRs. Review of potential signals this traditional experience is not effective at identifying herb(s) that
in the UMC database clearly shows the challenges and issues of cause cumulative, chronic or delayed toxicity. If the rst signs of
herbal pharmacovigilance, from identifying what was used (plant, adverse effects are not recognised until months or years after start-
plant part, extract, dose) to clinical information available from dif- ing or even stopping use of the herbs/drugs the use of the herbs is
ferent countries. The UMC has subsequently developed the WHO likely to be forgotten with such a delay.
Herbal Dictionary an international reference source of herbal prod- Aristolochic acid nephropathy (AAN) is a very good example
ucts designed to be used for coding and analysis of drug safety data of chronic toxicity (Type C). The effects are cumulative and renal
both pre and post marketing. symptoms can be delayed for 2 years after stopping use of the
There are a number of initiatives to address herbal identication herbs (Reginster et al., 1995). Various species of Aristolochia have
and set quality standards. In addition to individual country pharma- been used in the traditional medicines of many countries (Heinrich
copoeias (e.g. British Pharmacopoeia) the European Pharmacopoeia et al., 2009) (Aristolochia manshuriensis Kom and Aristolochia fangchi
has an extensive programme of herbal monographs under develop- Y.C. Wu ex L.D. Chou & S.M. Hwang in Chinese medicine) although
ment including herbs from China and the Indian sub-continent. In aristolochic acids were known to have the potential to cause renal
addition the European Medicines Agency is also developing herbal toxicity. It was only because of the cluster of cases in Belgium
monographs. Herbal monographs have also been published by the with detailed follow-up that the pattern of toxicity was recog-
American Herbal Pharmacopoeia, USP and the WHO (5 volumes). nised and characterised with full histological description, details of
These monographs can dene the standards for herb qual- progression of the disease and possible development of urothelial
ity however to be effective these require appropriate regulation carcinomas (Reginster et al., 1995; Lemy et al., 2008).
and enforcement to ensure implementation. In Europe, there is Idiosyncratic reactions (Type B) can occur within days or weeks
a requirement for GMP of all registered herbal products (well- of starting use of a medicine but are difcult to identify as these are
established use or traditional herbal products). Although this not predictable, neither dose nor time dependent and are not nec-
should improve the overall quality of herbal products available on essarily related to pharmacological activities but can result from
the market, less scrupulous dealers may sell poor quality products formation of reactive metabolites and immune-mediated reac-
through the internet, mail order or other unregulated supply routes. tions (Edwards and Aronson, 2000). Such reactions occur rarely
Products sold as foods or food supplements will also avoid the GMP (>1:10,000 and <1:1000) but are signicant as they may be serious
requirements. or even fatal (CIOMS, 1995). Drug-induced hepatotoxicity is often
due to an idiosyncratic reaction. It is difcult to conrm single case
reports of suspected herbal toxicity which may be due to idiosyn-
2.3. Sourceusers of herbal medicine cratic reactions as symptoms may be non-specic and there is no
clearly identiable toxic compound that can be measured in labo-
Surveys have shown that consumers tend to self-prescribe ratory analysis. Such rare adverse reactions will not be recognised
herbal medicines without consulting a professional herbal practi- without a formal reporting system.
tioner or other health professional (Barnes et al., 1998; Ipsos Mori,
2008). Products can be bought over-the-counter from pharmacies,
supermarkets, markets or the internet without any consultation 4. Pharmacovigilance methods
with a health professional. Herbal medicines are prescribed by
orthodox medical professionals in few European countries (e.g. A range of methods are used for post marketing monitoring
Germany). Consumers may not be aware that adverse effects of of drug safety including spontaneous reporting and prescrip-
herbal medicines can be reported to their general practitioner or tion event monitoring (DynPage UMC). These methods can be
know how to report to regulatory authorities. In addition, con- used for monitoring herbal safety but require modication to
sumers may not associate the herbal product with the effect. A address specic challenges such as botanical nomenclature, quality,
number of studies have shown that consumers are reluctant to adulteration, labelling issues, prescriber/reporter differences and
admit to their physician that they have been using herbal medicines under-reporting.
(Barnes, 2003).
Some consumers will consult herbal practitioners, but at present 4.1. Spontaneous reports
regulations as to who can supply herbal medicines vary consid-
erably around Europe. Training and practice standards vary from Safety of medicines is commonly monitored through sponta-
highly trained professionals to untrained and unregulated individ- neous reporting systems. There are subtle differences between
uals. countries but the principles are the same. Standardised forms are
used for reporting of suspected adverse reactions to the regulatory
authorities by medical professionals, including physicians, phar-
3. Identifying adverse reactions
macists, nurses and in some countries, by consumers (McLernon
et al., 2010). The reports are of suspected adverse reactions, and
The classication of types of adverse reactions is well estab-
a reporter does not have to conrm the association between drug
lished in orthodox medicine and applies equally to herbal medicine.
and effect. Such causality is assessed on a case-by-case basis by
Adverse reactions are classied as (Edwards and Aronson, 2000):
the reporting centres. Statistical methods are used to identify dis-
proportionate reporting rates which can lead to a safety signal. A
Type A (acute/augmented); dose related and explained by phar- signal only indicates an adverse effect of interest and one that
macology of herbs. requires further evaluation and investigation the link to the drug
Type B (bizarre/idiosyncratic); not dose related or predictable by or herb is not conrmed.
pharmacology. Spontaneous reports are more likely to be effective where
Type C (chronic/cumulative): cumulative effect. products are regulated as medicines and also where products are
Type D (delayed onset) carcinogenic, genotoxic. supplied by health professionals who are well informed in the use
516 D. Shaw et al. / Journal of Ethnopharmacology 140 (2012) 513518

of this reporting system. Consumers may not be aware of the impor- Hospital also undertakes multidisciplinary investigation of cases of
tance of reporting adverse effects (McLernon et al., 2010). herbal toxicity.
In the US where herbal medicines/natural products are supplied Well established use or products registered under the Tradi-
as dietary supplements, health professionals and consumers can tional Herbal Medicinal Product Directive (THMPD) will have brand
report suspected adverse reactions to the FDA MedWatch scheme. name and the correctly listed constituents. Getting accurate lists
In the UK the spontaneous reporting system is referred to as the of ingredients is problematic for unregulated/unlicensed products.
yellow card scheme; other countries such as Australia have blue Poor quality products are a continuing concern as there is no assur-
cards. In the UK the yellow card was modied to allow for the inclu- ance that the product contains the ingredients listed on the label.
sion of herbals in 2000. However there are still problems regarding Herbal products adulterated with pharmaceutical medicines for
accurate ingredient lists, botanical naming of the medicinal herbs, slimming (e.g. sibutramine), inammatory conditions (steroids) or
processing and product quality. erectile dysfunction (sildenal) are an international problem. In
In the UK the Medicines and Healthcare Products Regulatory a review of herbal safety alerts for 2010, we found that pharma-
Agency receive about 20,000 yellow card reports per annum but ceutical contamination or adulteration accounted for 336 of 390
this includes only about 100 herbal reports. Despite efforts to warnings that were issued by regulatory authorities in UK, US,
improve reporting by extending to nurses, pharmacists and con- Canada, Singapore, Hong Kong and Australia.
sumers there has not been any signicant increase in herbal reports. With low numbers of ADR reports in a single country, then
Countries including Sweden and Italy have carried out studies on herbal signals of interest may not be identied, especially for reac-
herbal ADRs (Menniti-Ippolito et al., 2008; Jacobsson et al., 2009). tions that are rare. The WHO Collaborating Centre for Monitoring
Because there are few yellow card herbal reports in the UK, it Drug Safety (UMC) seeks to address this problem by collating
is relatively easy to identify adverse effects of concern by evalua- ADR reports from over 100 countries around the world. By early
tion of the individual reports without waiting for statistical signal 2011 their database contained over 6 million drug and herbal
detection. reports (UMC, 2011). This is the largest repository of such reports.
Where medicines are regulated (e.g. licensed as well- They have attempted to address the nomenclature issues as herbal
established or traditional herbal medicinal product in Europe) then reports come from countries with different systems of traditional
manufacturers have pharmacovigilance obligations under Euro- medicine.
pean directives and perhaps additional National regulations. These However because of the variability of the composition of differ-
requirements are the same for orthodox and herbal medicines. ent herbal products caution is needed when combining reports on
This includes timeframes and other reporting requirements for a single herb/product. But groups of herbs with similar chemical
informing the regulatory authorities of any reports of unwanted composition can be used to identify signals for further study.
or unexpected side effects of their products. Unlicensed or unregu-
lated products or dietary supplements do not have to comply with
this directive. 5. Monitoring for herbdrug interactions

There is a perception that herbal medicines are safe, even if

4.2. Problems with spontaneous reports
taken at the same time as prescription drugs (Delgoda et al., 2004).
Herbs may be used to treat the primary condition or to reduce
Underreporting is a well known problem with spontaneous
the side effects of their conventional treatment. Under-reporting
reporting systems (Hazell and Shakir, 2006). It is thought that this
of suspected interactions between herbs and drugs is of increasing
is likely to be a greater problem with herbal medicines. Factors
concern and arises from the same reasons as under-reporting of
contributing to under-reporting of herbal ADRs include:
herbal ADRs.
The particular problems that need to be addressed are those that
Lack of association between herb and adverse effect. may affect specic patient groups where the incidence of combin-
Patient stops using the herbal medicine when they feel unwell. ing orthodox and herbal medicine use is thought to be high, and the
Physician/patient unaware that herbal ADRs should be reported. risk of interaction signicant, such as in cancer patients. However
Physician unaware of the use of herbal medicines as patient does any patients who are on drug regimes involving potent medicines
not consider herbal and nutritional products to be medicines metabolised by cytochrome P450 enzymes or where bioavailability
and does not disclose use (Barnes, 2003; Williamson et al., 2009). is affected by P glycoprotein are at increased risk of experiencing
herbdrug interactions.
The usefulness of any report of adverse effects is dependent on Adverse reactions due to interactions may not be recognised
knowing what herb/product was used. This relies on the accurate if the physician or other health professional is not aware of the
identication of the product/herb by the reporter. If inadequate concomitant use of medicinal herbs (Giveon et al., 2004; Goldstein
names are used this can lead to confusion. For example if only gin- et al., 2007). Adverse reactions due to drug interactions are rela-
seng is used in a report this may refer to Panax ginseng C.A. Mey tively common although largely preventable. The UMC has recently
or Panax quinquefolius (Burk.) F.H. Chen. But it may also refer to a undertaken studies to show that pharmacokinetic and pharma-
range of other plants that are also labelled as ginseng Eleuthero- codynamic drug interactions can be detected in their database
coccus senticosus (Rupr. & Maxim) Maxim (Siberian ginseng) or the (Strandell et al., 2011). As this study develops it is hoped that
unrelated Withania somnifera (L.) Dunal (Indian ginseng). herbdrug interactions may also be investigated.
The quality of ADR reports can be improved by collaboration
with pharmacognosy departments, botanic gardens or other tox-
icological units. In the UK the Chinese Medicine Advisory Service 6. Other methods of monitoring
assists with enquiries on suspected herbal ADRs. Through collab-
oration with the Royal Botanic Gardens Kew, the identity of the Prescription event monitoring (PEM) is a non-interventional
Chinese medicinal herbs can be veried. When ADRs are published hypothesis generating method for studying a drug once it is placed
in medical journals or submitted to the authorities then there is on the market, through monitoring of individual prescriptions. In
some condence in the botanical identity. In Hong Kong the Hospi- the UK a modied protocol for using PEM for herbal medicines
tal Authority Toxicology Reference Laboratory at Princess Margaret has been developed based on monitoring prescriptions from herbal
D. Shaw et al. / Journal of Ethnopharmacology 140 (2012) 513518 517

practitioners. It is a useful method for investigating specic safety For manufactured herbal products they include product name,
concerns on frequently used medicinal herbs (Barnes, 2003). producer, batch number, type and concentration of extract and
Intensive monitoring schemes can be an extension of the spon- concentration of any standardised constituents. Botanical iden-
taneous reporting schemes and are used to stimulate reporting on tication only includes Latin scientic name, plant parts and
specic medicines. In Thailand there is an extensive list of regis- preparation (herb or extract). These guidelines are however not
tered herbal medicines that are also used in hospital settings. They adequate for herbal medicines supplied as complex formulas or for
used intensive monitoring on 9 different herbal products where most traditional medicines such as Chinese or Ayurvedic medicine
there was a need for further safety information (Saokaew et al., as used in Europe.
2011). Any ADR report should list the name as given on the
Another source of pharmacovigilance and safety information on label/prescription. If there is additional information on identity
herbal medicines is the Poisons Control Centres (Dennehy et al., based on morphological or chemical investigation then the source
2005; Gryzlak et al., 2007). In Europe and US these centres take of this detail should be included. Traditional herbal formulas (and
enquiries where there are concerns about the safety of a prod- even products) do not necessarily include the botanical name. The
uct or where there is a suspected poisoning. These enquiries are Latin scientic name cannot be assumed from the common or drug
not necessarily formal reports so supporting information, includ- name and is only accurate if the plant material has been authen-
ing product details, time course and dosage may be lacking. Often ticated. If the plant identity has been veried, then the scientic
the patients may have taken an acute or chronic overdose and are name, author, plant part and processing should be given. Details
seeking medical attention; this may not yield useful information of processing of the herbal material before use (e.g. in Chinese
on medium- or long-term toxicities. Poisons Control Centres are medicine, stir frying, steaming) are required as this can dramati-
an important source for dietary supplement ADRs in the US as was cally change the chemical prole or bioavailability of the herbs.
shown by a study carried out in 2008 which revealed that the pri- Other useful details for any evaluation are reason for use and
mary reporting portal MedWatch, received fewer reports than the diagnosis by the herbal practitioner. This should be based on sup-
poison centres (Gardiner et al., 2008). portive information from the practitioner. Involving the herbal
Other pharmacoepidemiological methods for detailed investi- practitioner in any assessment of suspected adverse reaction can
gation of drugs that can also be used for herbal medicine safety provide useful background or understanding of the use of the herbs.
include case control and cohort studies. These can be used to test Accurate adverse reaction reports are required so that real
hypotheses developed after signals are detected using spontaneous adverse events can be identied and appropriate warnings and
reports. One such signal was identied from the reports of possi- guidelines can be given to practitioners and patients. If suspected
ble liver injury associated with the use of Chinese herbs (Perharic adverse reactions are not properly evaluated or the herbs have been
et al., 1995; McRae et al., 2002). A pilot case control study was used incorrectly identied then safe or useful herbs may be incorrectly
to demonstrate that there was no increased association with liver restricted. Accurate identication of the medicinal herbs used is
injury and any individual herb (Shaw, 2010b). As yet these methods essential when publishing case reports in the medical literature to
have been under-utilised for herbal medicines. prevent misunderstandings.
As the use of herbal medicines has spread around the world
there is lack of information on the safety of herbs as used in
7. Herbal practitioners
different patient populations that may differ in essential param-
eters such as pharmacogenomics and metabolization proles or
Herbal practitioners are potentially a useful source of informa-
gut microora composition and bioactivity. Effective pharmacovig-
tion on ADRs but with varying levels of professional regulation
ilance is essential to build up reliable information on the safety of
in Europe they are not necessarily recognised as ADR reporters.
herbal medicines for the development of appropriate guidelines for
Some herbal practitioner organisations have established their own
safe effective use.
reporting schemes but these are not necessarily linked to ofcial
agencies. There are benets to reporting by trained herbal practi-
tioners. They are educated in the use of the medicinal herbs and Acknowledgement
should know actions and potential toxicity of the herb and be
able to identify unexpected effects of the treatment. Herbal pre- Authors wish to thanks Christine Leon, Royal Botanic Gardens
scriptions are routinely modied to reduce side effects or improve Kew for advice and discussion on botanical nomenclature.
responses. Although some effects may be minor sometimes non-
specic symptoms may indicate a more serious illness (such as References
tiredness and liver dysfunction). In Europe they may treat rela-
Barnes, J., Mills, S.Y., Abbot, N.C., et al., 1998. Different standards for reporting ADRs
tively low numbers of patients so may not recognise rare adverse to herbal remedies and conventional OTC medicines: face-to-face interviews
events. In countries where there is limited recognition or lack of with 515 users of herbal remedies. British Journal of Clinical Pharmacology 45,
professional standing of herbal practice then practitioners are likely 496500.
Barnes, J., 2003. Pharmacovigilance of herbal medicines: A UK perspective. Drug
to be less condent about reporting suspected ADRS. Clear deni- Safety 26, 829851.
tions of reportable ADRs need to be dened and harmonised across CIOMS Working Group III, 1995. Good safety information practices. In: Guidelines
jurisdictions. for Preparing Core Clinical Safety Information on Drugs. WHO, Geneva.
Delgoda, R., Ellington, C., Barrett, S., Gordon, N., Clarke, N., Younger, N., 2004. The
practice of polypharmacy involving herbal and prescription medicines in the
treatment of diabetes mellitus, hypertension and gastrointestinal disorders in
8. Minimum requirements for ADRs
Jamaica. West Indian Medical Journal 53, 400405.
Dennehy, C.E., Tsourounis, C., Horn, A.J., 2005. Dietary supplement-related adverse
Various groups have produced guidelines for reporting adverse events reported to the California poison control system. American Journal of
Health-System Pharmacy 62, 14761482.
events and clinical trials (Gagnier et al., 2006; Kelly et al., 2007)
DynPage, 2011.
including herbal products. The requirements for case reports mn2=7252.
include essential case details: patient demographics (age, gen- Edwards, I.R., Aronson, J.K., 2000. Adverse drug reactions: denitions, diagnosis and
der) relevant medical history, symptoms, abnormal laboratory management. Lancet 356, 12551259.
Gagnier, J., Boon, H., Rochon, P., Moher, D., Barnes, J., Bombardier, C., for the CONSORT
results, drug identication, reason for use, dose, time course (dura- Group, 2006. Controlled trials evaluating herbal interventions: implementing
tion of use, onset of symptoms) and details of adverse event. the consort statement. Drug Safety 29, 9111010.
518 D. Shaw et al. / Journal of Ethnopharmacology 140 (2012) 513518

Gardiner, P., Sarma, D., Low Dog, T., Barrett, M., Chavez, M., Ko, R., Mahady, G., Marles, Loew, D., Kaszkin, M., 2002. Approaching the problem of bioequivalence of herbal
R., Pellicore, L., Giancaspro, G., 2008. The state of dietary supplement adverse medicinal products. Phytotherapy Research 16, 705711.
event reporting in the United States. Pharmacoepidemiology and Drug Safety Mann, R.D., Andrews, E.F. (Eds.), 2002. Pharmacovigilance. Wiley, Chichester.
17, 962970. McLernon, D.J., Bond, C.M., Hannaford, P.C., et al., 2010. Adverse drug reaction report-
Giveon, S.M., Liberman, N., Klang, S., Kahan, E., 2004. Are people who use natural ing in the UK. Drug Safety 33, 775786.
drugs aware of their potentially harmful side effects and reporting to family McRae, C.A., Agarwal, K., Mutimer, D., Bassendine, M.F., 2002. Hepatitis associated
physician. Patient Education Counsel 53, 511. with Chinese herbs. European Journal of Gastroenterology and Hepatology 14,
Goldstein, L.H., Elias, M., Ron-Avraham, G., Biniaurishvili, B.Z., Madjar, M., Kamar- 559562.
gash, I., Braunstein, R., Berkovitch, M., Golik, A., 2007. Consumption of herbal Menniti-Ippolito, F., Mazzanti, G., Santuccio, C., Moro, P., Calapai, G., Firenzuoli,
remedies and dietary supplements amongst patients hospitalized in medical Valeri, F., Raschetti, R., 2008. Surveillance of suspected adverse reactions to
wards. British Journal of Clinical Pharmacology 64, 373380. natural health products in Italy. Pharmacoepidemiology and Drug Safety 17,
Gryzlak, B.M., Wallace, R.B., Zimmerman, M.B., Nisly, N., 2007. National surveillance 626635.
of herbal dietary supplement exposures: the poison control center experience. Nortier, J., Vanherweghem, J.L., 2007. For patients taking herbal therapy lessons
Pharmacoepidemiology and Drug Safety 16, 947957. from aristolochic acid nephropathy. Nephrology Dialysis and Transplantation
Hazell, L., Shakir, S.A.W., 2006. Under reporting of adverse drug reactions: a system- 22, 15121517.
atic review. Drug Safety 29, 385396. Perharic, L., Shaw, D., Leon, C., De Smet, P.A., Murray, V.S., 1995. Possible associa-
Heinrich, M., Chan, J., Wanke, S., Neinhuis, C., Simmonds, M., 2009. Local uses of Aris- tion of liver damage with the use of Chinese herbal medicine for skin disease.
tolochia species and content of nephrotoxic aristolochic acids 1 and 2 a global Veterinary and Human Toxicology 37, 562566.
assessment based on bibliographic sources. Journal of Ethnopharmacology 125, Reginster, F., Jadoul, M., van Ysersele de Strihou, C., 1995. Chinese herbs nephropathy
108144. presentation, natural history and fate after transplantation. Nephrology Dialysis
Ipsos Mori, 2008. Public Perceptions of Herbal Medicines General Public Qualitative and Transplantation 10, 157160.
and Quantitative Research, Saokaew, S., Suwankesawong, W., Permsuwann, U., et al., 2011. Safety of herbal
po/documents/news/con036073.pdf (accessed October 2011). products in Thailand. Drug Safety 34, 339350.
Jacobsson, I., Jonsson, A., Gerden, B., Hagg, S., 2009. Spontaneously reported Shaw, D., 2010a. Toxicological risks of Chinese herbs. Planta Medica 76, 20122018.
adverse reactions in association with complementary and alternative Shaw, D., 2010b. Investigation of liver toxicity of Chinese herbal medicine: pilot
medicine substances in Sweden. Pharmacoepidemiology and Drug Safety 18, case-control study. Drug Safety 33, 918919.
10391047. Strandell, J., Caster, O., Bate, A., Norn, G.N., Edwards, I.R., 2011. Reporting patterns
Kelly, W.N., Arellano, F.M., Barnes, et al., 2007. Guidelines for submitting adverse indicative of adverse drug interactions: a systematic evaluation in VigiBase. Drug
event reports for publication. Drug Safety 30, 367373. Safety 34, 253266.
Lemy, A., Wissing, K.M., Rorive, S., Ziotta, A., Roumeguere, T., Muniz Martinez, M.C., UMC. 2011. <> (accessed 24.10.11.).
Decaestecker, C., Salmon, I., Abramowicz, D., Vanherweghem, J.L., Nortier, J., World Health Organisation, 2004. WHO Guidelines on Safety Monitoring of Herbal
2008. Late onset of bladder urothelial carcinoma after kidney transplantation Medicines in Pharmacovigilance Systems. WHO, Geneva.
for end-stage arisotolochic acid nephropathy: a case series with 15-year follow Williamson, E.M., Driver, S., Baxter, K. (Eds.), 2009. Stockleys Herbal Medicines
up. American Journal of Kidney Diseases 51, 471477. Interactions. Pharmaceutical Press, UK.