Herbal medicinal products during pregnancy: are they safe?

Authors


Introduction

The use of herbal medicinal products, usually marketed as dietary (food, nutritional) supplements or ‘neutraceuticals’, is rapidly increasing. Between 1990 and 1997 herbal medicine product usage in the US general population rose by 380%1. In 1998 the total US sales of herbal medicine products amounted to $4 billion2. Virtually all survey data agree that users of herbal medicine products are predominantly female1,3,4. Thus one might assume that pregnant women frequently use herbal medicine products, particularly as such remedies are often perceived as being ‘natural and therefore free of risks’5.

This article is an attempt to review the usage of herbal medicine products by pregnant women and to discuss some of the implications arising from it. Its aim is to alert healthcare professionals to the fact that herbal medicine products are not entirely free of risks for pregnant or lactating women.

Methods

Medline and Embase searches (1966–August 2001) were conducted using the following key words: adverse effects, genotox, herbal, herbalism, herbal medicine, lactation, mutagen, phytotherapy, pregnancy, risk, safety, teratogen. In addition, my own extensive files were searched. The bibliographies of all articles thus found were further scanned for relevant articles. For inclusion, an article had to contain original data on either the prevalence of use or adverse effects of herbal medicine products during pregnancy or lactation. Papers relating to related subjects such as fertility treatments, contraception, the deliberate use of herbal medicine products as abortifacients or postpartum care were excluded. In vitro studies were also excluded. No language restrictions were applied. All articles thus retrieved were read by the present author. Data was extracted according to pre-defined criteria and are summarised in narrative form below.

Use of Herbal Medicine Products in Pregnancy

patients' Surveys

A series of surveys conducted in Finland suggested that, between 1985 and 1988, the use of ‘alternative drugs’ (mostly herbal medicine products) had risen from 4% to 15%6. The authors also confirmed that higher social classes seem to use herbal medicine products more frequently than pregnant women from lower classes. An Australian survey of 300 consecutive women attending antenatal clinics implied that 12% had taken herbal medicine products during their pregnancy(personal communication, G. Pinn). A survey conducted with 1200 pregnant Nigerian women demonstrated that 12% used native herbs7. Herbal medicine product use was more prevalent amongst nulliparous women (42%). A study conducted in South Africa showed that out of 229 pregnant women, 55% had used herbal medicine products during pregnancy8. Also in South Africa, a survey of 218 pregnant women demonstrated that 7% of the sample preferred traditional herbal medicine to conventional medicine9. Eight percent of these women thought that, in antenatal care, the former was more effective than the latter. A survey of 200 pregnant US women demonstrated that 15% used ‘home remedies’ (most commonly ginger, vitamin B6, chamomile, and cola) in an attempt to relieve morning sickness10. Two hundred and fifty pregnant US women attending antepartum visits were prospectively enrolled in a survey about the use of herbal and alternative medical therapies11. Two hundred and forty-two (97%) women completed the questionnaire. Of the respondents, 9.1% reported using herbal medicine products during the current pregnancy, 7.5% using these agents at least weekly. The most commonly used herbal medicine products during pregnancy were garlic, aloe, chamomile, peppermint, ginger, echinacea, pumpkin seeds, and ginseng. Herbal medicine product use during pregnancy was strongly associated with prior use of herbal supplements. There were trends towards greater use amongst white and better educated women. Similar findings were reported in another US survey12. Of the 150 pregnant women responding, 13% had used dietary supplements during pregnancy. The most common products were echinacea (9%), ‘pregnancy tea’ (9%) and ginger (7%).

Therapists' Surveys

A survey of 500 members of the American College of Nurse–Midwives suggested that more than half of them employed herbal medicine products for the purpose of inducing labour13. The herbal medicine products most frequently named were blue cohosh (64%), black cohosh (45%), red raspberry leaf (64%), castor oil (93%) and evening primrose oil (60%). It is interesting to note that for none of these herbal medicine products is there compelling evidence of efficacy for inducing labour14. Canadian researchers sent questionnaires to randomly selected groups of physicians (n= 157) and naturopaths (n= 194)15. Only one physician and 49% of the naturopaths recommended herbal medicine products to their pregnant patients. In a further study, questionnaires were mailed to all 120 licensed certified nurse–midwives in North Carolina16. The researchers requested information concerning their recommendations about ‘alternative’ treatments for pregnant women. Of the respondents, 94% reported recommending such therapies and 57% recommended them to more than 10% of their patients. Specifically, herbal medicine products were recommended by 73% of all respondents.

Safety of Specific Herbal Medicine Products: Case Reports

Agnus Castus

Agnus castus is a plant with oestrogen-like activities used for a variety of gynaecological problems, particularly on the European continent. One report described a woman who, while undergoing in vitro fertilisation, took this remedy during an unstimulated cycle17. She showed considerable derangement of gonadotrophin and ovarian hormone levels. The authors believed that agnus castus may lead to ovarian hyperstimulation and may increase the risk of miscarriage.

Alcoholic Herbal Tincture

A case of ‘fetal alcohol syndrome’ was associated with the use of a herbal tonic by a 29 year old Chinese woman during her pregnancy18. The herbal medicine product in question contained 19% alcohol, and the mother had used it extensively during early pregnancy. She denied the exposure to alcoholic beverages or drugs. The authors thus believe that the herbal medicine product caused the syndrome. At three-year follow up, the boy still showed signs of motor and mental retardation.

Blue Cohosh

US paediatricians reported the case of an infant whose mother had taken blue cohosh to promote uterine contractions19. The infant suffered myocardial infarction, profound congestive heart failure and cardiovascular shock. After remaining critically ill for several weeks, he eventually recovered. The authors believe that cardiotoxic alkaloids contained in blue cohosh were the cause of this infant's health problems.

A woman ingested an unspecified amount of blue and black cohosh to induce labour20. The infant was hospitalised shortly afterwards with seizures, kidney damage, and the need for mechanical ventilation. The authors suggested that caulosaponin, a constituent of blue cohosh, caused the problems. Caulosaponin constricts coronary blood vessels and causes myocardial toxic reactions.

Chinese Herbal Medicine

A case has been reported of a 15 month old Chinese girl with an accessory phallic urethra (about 1cm in diameter and 2cm long) arising to the right of the anus21. The authors discussed the possibility that this unusual abnormality was related to the fact that the mother took some Chinese herbal medicines during pregnancy. The nature of the herbs was not noted and causality in this case seems uncertain.

Dong Quai

Dong quai is recommended in traditional Chinese medicine for dysmenorrhea, irregular menstruation, anaemia, postpartum weakness, and other problems. A 32 year old woman, three weeks postpartum, developed acute headache, weakness, light-headedness, and vomiting22. Her blood pressure was 195/85mmHg. She had taken Dong quai for postpartum weakness and said that she had not been taking any other medicines. Her three-week-old son's blood pressure was also raised at 115/69. Dong quai medication of the mother and breastfeeding of the child were discontinued and the blood pressure normalised in both patients within 48 hours.

Ginseng

A 30 year old mother had taken panax ginseng (650mg 2×/day) throughout pregnancy and during lactation of her two-week-old baby23. The boy (birthweight 3.3kg) was noted to have thick black pubic hair, hair over the entire forehead, and swollen, red nipples. During the first 3.5 weeks of life, he gained 1.1kg while being breastfed and subsequently received a formula. His pubic and forehead hair began to fall out after he was two weeks old and was scant at 7.5 weeks. He had enlarged testes. Endogenous androgen production was ruled out and the authors suspected that this case of androgenisation was caused by the hormonal effects of the herbal medicine product. A subsequent exchange of ‘letters to the Editor’ cast doubt on the notion that the herbal medicine product contained panax ginseng and suggested it contained Siberian ginseng24.

Polygonum Multiflorum

Investigators from Hong Kong reported the case of a 31 year old Chinese woman who developed signs and symptoms of hepatitis after taking the Chinese remedy ‘Shou-Wu-Pian’ prepared from Polygonum multiflorum25. The obvious causes for her condition were excluded, and the authors are confident that it was caused by the hepatotoxicity of the herbal medicine product. The clinical outcome was not reported.

Pyrrolizidine Alkaloids

A five-day old Swiss girl was admitted with symptoms of jaundice, massive hepatomegaly and abdominal effusion26. The mother had had a normal pregnancy. The baby's abdomen was distended, and she displayed symptoms of jaundice. Extensive biochemical tests were performed to identify the cause of neonatal liver failure. An open liver biopsy was performed on the 26th day of life, and the baby died 11 days later. The biopsy showed lesions typical of veno-occlusive disease. It turned out that the mother had consumed large amounts of herbal tea throughout the course of her pregnancy. TLC analysis found that the herbal tea contained 0.6mg of pyrrolizidine alkaloids which are known to be hepatotoxic.

St John's Wort

Two women who took St John's wort during pregnancy, in order to avoid potential harmful effects of synthetic antidepressants to the fetus, also discontinued their prescribed medications without discussing their decisions with their doctor27. Because the effects of St John's wort (or most other herbal medicine products) on the fetus are not known, the authors cautioned against the use of St John's wort under these circumstances and argued that tricyclic antidepressants or fluoxetine would be the safer form of antidepressive drug therapy.

Tripterygium Wilfordii

Tripterygium wilfordii is used for rheumatoid arthritis and for male contraception, particularly in Japan. Its high level of toxicity has been described repeatedly. A Japanese woman took the remedy during early pregnancy for rheumatoid arthritis28. After an uncomplicated delivery at 38 weeks, the infant suffered from occipital meningoencephalocele and cerebellar agenesis. The authors considered that Tripterygium wilfordii was the most likely cause of this infant's anomalies.

Case Series and Retrospective Surveys

Castor Oil

Four hundred and ninety-eight pregnant women were asked about herbal medicine product-use before artificial rupture of membranes29. Meconium passage was significantly more common in patients who had recently taken either castor oil or a local (South African) herbal medicine product called ‘Sihlambezo’. Caesarean section was significantly more frequent in pregnancies complicated by fetal meconium passage.

Echinacea

A study from Canada investigated the safety of echinacea use during pregnancy30. Four hundred and twelve pregnant women contacted a teratogen information service between 1996 and 1998 with concerns about the safety of consuming echinacea during pregnancy. Two hundred and six of them had already taken echinacea during pregnancy, while the other 206 (the control group) had subsequently decided not to take it. In the echinacea group, 112 women (54%) reported taking the herb in the first trimester of pregnancy, and 17 (8%) used echinacea throughout their pregnancies. No significant differences were noted between the echinacea and the control groups in the rate of major or minor birth defects, nor were there any differences in pregnancy outcome, delivery method, maternal weight gain, gestational age, infant birthweight, or fetal distress. In the echinacea group, six major and six minor malformations occurred; of these, there were four major and two minor malformations in babies of women who took echinacea during the first trimester. By comparison, the researchers observed seven major and seven minor malformations in control group infants. Thirteen miscarriages were documented in the echinacea group, compared with seven in the control group.

German Herbal Mixture

A similar study from Germany (published as an abstract only) retrospectively analysed 762 pregnancies31. Malformations, stillbirths and miscarriages were noted and set in relation to use of Sinupret, a complex herbal remedy licensed for sinusitis. The authors provide few details and no data but conclude that this study “does not indicate the possibility of any teratogenic or embryotoxic effect of this herbal medicine product”.

Montanoa Tomentosa

Eight pregnant women drank infusions of Montanoa tomentosa during labour32. Their newborns showed cardiorespiratory depression requiring intensive care. They improved during the first minutes with an average Apgar score at one minute of 4.5, and 7.4 at 10 minutes. A negative correlation was found between the number of ingested infusions and the Apgar score at one minute. The mechanism underlying the adverse effects is still unknown. The authors suggest that it may be similar to oxytocin or ergot alkaloids, with which Montanoa shares uterine or systemic effects.

Raspberry Leaf

A small retrospective study of raspberry leaf tea in childbearing women suggested a decreased likelihood of premature or overdue labour and of medical intervention in labour33. The study sample comprised only 51 women who had taken the herbal medicine product during pregnancy; thus it had insufficient power to detect adverse events. The herbal medicine product was not associated with any childbirth complications, and no evidence of long term teratogenic effects were found. In view of its stimulant effect on the uterus, the herbal medicine product should, however, not be recommended during pregnancy.

Various South African Herbal Medicines

In a study carried out in two hospitals in South Africa, 20 children were identified with hepatic veno-occlusive disease, probably due to the administration of herbal medicine products34. The predominant clinical presentation was ascites of various degrees and hepatomegaly. Nine children died. Of those survivors who could be followed up, progression to cirrhosis and portal hypertension was frequent. Pyrrolizidine alkaloid poisoning was implicated to be the cause of the problem. In four cases an on-admission urine specimen was available and in all of these the presence of pyrrolizidine alkaloids was confirmed.

South African investigators randomly selected 229 women presenting in early labour35. Of these, 55% gave a positive history of herbal medicine product use and 45% had not used herbal medicine products during pregnancy. Grade II to III meconium staining of liquor was 56% in the former and 15% in the latter subgroup. Caesarean section had to be employed in 39% of the former and 22% of the latter population.

A further study from South Africa investigated the incidence and aetiology of acute renal failure in pregnancy in patients requiring haemodialysis36. Forty-two women were included. The authors noted that, in this population, the use of herbal medicine products was associated with an increased rate of septic abortions.

One of the above-mentioned surveys related herbal medicine product use in South Africa to pregnancy outcome8. Fifteen percent of the women not using herbal medicine products and 55.6% of those using herbal medicine products had had grade II to III meconium staining of liquor. The rates of caesarean section were 22% and 38.5%, respectively. The authors concluded that herbal medicine product-use during pregnancy may lead to fetal distress.

Clinical Trials of Safety

Evening Primrose Oil

There is a paucity of clinical trials of herbal medicine products in pregnant women which specifically report on adverse effects. One exception is a US controlled study of evening primrose oil on the length of pregnancy in low risk nulliparous women37. Fifty-four women taking this herbal medicine product were compared with 54 women who did not. Oral administration of evening primrose oil from the 37th gestational week until birth did not shorten gestation or decrease duration of labour. It was, however, associated with an increase in the incidence of prolonged rupture of membranes, oxytocin augmentation, arrest of descent and an increased frequency of vacuum extraction.

Raspberry Leaf

One hundred and ninety-two low risk, nulliparous women were randomised to receive either raspberry leaf tablets (2 × 1.2g/day) or placebo38. Medication started during the 32nd week of pregnancy until labour. No adverse effects for mother or baby were noted. The herbal medicine product did not shorten the first stage of labour but a small shortening of the second stage (10 minutes on average) and less (19%vs 30%) forceps deliveries were observed in the treatment compared with the control group.

Epidemiological Studies

Glucocorticoids have long been suspected to be responsible for low birthweight. Finnish authors tested whether maternal consumption of glycyrrhizin (an inhibitor of cortisol metabolism) in licorice affects birthweight in humans. A sample of 1049 women and their healthy singleton infants was studied in 199839. Glycyrrhizin intake was calculated from detailed questionnaires on licorice consumption. Glycyrrhizin exposure was grouped into three levels; low (<250mg/week; n= 751), moderate (250–499mg/week; n= 145) and heavy (≥500mg/week; n= 110). Birthweight and gestational age were obtained from hospital records. Babies with heavy exposure to glycyrrhizin were not significantly lighter at birth, but they were significantly more likely to be born prematurely. The odds ratio for being born before 38 weeks of gestation was 2.5 (95% CI 1.1, 5.5; P= 0.03).

Comment

Although preliminary, the evidence summarised above shows that herbal medicine products have been associated with risks to pregnant women and their babies. Rather than being an in-depth review of the subject, the main purpose of this review was to alert healthcare professionals to the fact that herbal medicine products are not entirely risk free.

Worldwide there are virtually thousands of herbal medicine products in use. An attempt has been made to systematically review the literature, yet the evidence assembled above is almost certainly incomplete. Under-reporting can be assumed to be high for several reasons. Consumers often think that herbal medicine products are risk free and rarely inform healthcare professionals about using herbal medicine products1. In most countries herbal medicine products are not regulated as medicines. This means that adverse effect monitoring is either non-existent or inefficient. Moreover, many articles of adverse effects of herbal medicine products appear in obscure journals or are published as ‘letters to the editor’ and are thus not retrievable with standard search techniques. In this context, it is noteworthy that the majority of the above evidence was not found through computerised literature searches but through searching my personal files. Table 1 is an attempt to summarise information regarding the potential risks of those herbal medicine products which are popular in western countries. This evidence has been collated from a range of up to date authoritative texts14,40–43. In spite of its impressive length, it does not claim to be complete.

Table 1.  Herbal medicinal products and potential adverse effects during pregnancy.
Common nameLatin nameRelevant adverse effects
  1. The information contained in this table has been collated from Ernst et al.14, Blumenthal et al.40, Brinker et al.41, Fetrow & Avila42 and Lepik43.

AlfalfaMedicago sativaMay cause uterine stimulation.
Aloe veraAloe veraStimulation of uterine muscle activity. Possible abortifacient and emmenagogue.
AngelicaAngelicaEmmenagogue effects.
AsafoetidaFerula asafoetidaEmmenagogue effects.
AshwagandhaWithamia somniferaAbortifacient properties.
BarberryBerberis vulgarisUterine stimulant.
BasilOcimum basilicumEmmanagogue, abortifacient, mutagenic.
BearberryArctostaphylos ursiOxytocic action.
Bitter lemonMomordica charantiaEmmanagogue and abortifacient effects.
Black cohoshCimicifuga racemosaOestrogenic activity, suppresses endogenous luteinising hormone secretion (in rats) and binds to uterine oestrogen receptors, reduces circulating luteinising hormone levels. Emmenagogue effects.
Blood rootSanguinaria canadensisEmmenagogue and uterine stimulant.
Blue cohoshCaulophyllum thalictroidesGastrointestinal symptoms, stimulates contraction of uterine muscle, causes arterial constriction, inhibits embryo implantation (in rats), alleged to induce menstruation and promote abortion.
BonesetEupatorium perfoliatumAbortifacient effects.
BorageBorago officinalisMutagenic (contains pyrrolizidine alkaloids).
BroomCytisus scopariusContains sparteine, a powerful oxytocic compound, traditionally used to induce labour.
BuckthornRhamnus catharticusAbortifacient, mutagenic, genotoxic effects.
BugleweedLycopus virginicusAntigonatotrophic and antithyrotropic activity.
BurdockArctium lappaOxitocic and uterine stimulant action.
ButterburPetasites hybridusEmmenagogue, hepatotoxic, genotoxic, & carcinogenic effects.
ButtercupRamunculusUterine stimulant.
CalamusAcorus calamusEmmenagogue and genotoxic activity.
CalendulaCalendula officinalisEmmenagogue and abortifacient effects.
CamphorCinnamomum camphoraEmmenagogue and uterine stimulant.
Cascara sagradaRhamnus purshinanaAbortifacient, mutagenic and genotoxic action.
Cassia cinnamonCinnamomum aromaticumEmmenagogue and abortifacient effects.
Castor beanRicinus communisEmmenagogue and abortifacient effects.
CatnipNepeta catariaEmmenagogue and abortifacient effects.
CelandineChelidonium majusUterine stimulant.
CeleryApium graveolensUterine stimulant, abortifacient and emmenagogic action.
Chamomile (Roman)Chamaemelum nobileEmmenagogue and abortifacient effects.
Chaste treeVitex agnus castusEmmenagogue effects.
ChicoryCichorium intybusEmmenagogue and abortifacient effects.
CinchonaCinchonaAbortifacient, uterine stimulant, oxytocic, teratogenic effects.
CinnamonCinnamonium verumEmmenagogue effects.
ColaCola nitidaLow birthweight, birth defects, premature birth.
ColtsfootTussilago fararaContains hepatotoxic pyrrolidizine alkaloids, risk of fatal hepatic veno-occlusive disease, abortifacient effects.
ComfreySymphytum officinaleContains hepatotoxic pyrrolizidine alkaloids, risk of foetal hepatic veno-occlusive disease, hepatotoxic and carcinogenic in animals.
Echinacea (Corn flower)E. augustifolia or E. purpurea or E. pallidaWeak oxytocic effect.
Ephedra (Ma Huang)Ephedra sinica, E. equisetina (and others)Contains ephedrine and related alkaloids, increases blood pressure, heart rate and causes CNS activity, stimulates uterine muscle.
FennelFoeniculum vulgareEmmenagogue effects.
FeverfewTanacetum partheniumMay promote menstruation and induces abortion.
FlaxLinum usitatissimumEmmenagogue effects.
FrangulaRhamnus frangulaEndometrial stimulation, mutagenic and genotoxic effects.
GarlicAllium sativumEmmenagogue effects.
GingerZingiber officinaleAbortifacient, emmenagogue and mutagenic effects.
GoldensealHydrastis canadensisUterine stimulant.
Gotu kolaCentella asiaticaEmmenagogue effects.
GuaranaPaullinia cupanaLow birthweight, birth defects, premature birth.
Hemp agrimonyEupatorium cannabinumEmmenagogue and abortifacient effects.
HibiscusHibiscus rosa sinensisEmmenagogue effects.
HorehoundMarrubium vulgareEmmenagogue and abortifacient effects.
HorseradishArmoracia rusticanaAbortifacient effects.
HyssopHyssopus officinalisEmmenagogue and abortifacient effects.
IpecacCephalis ipecacuanhaUterine stimulant.
Joe-pye weedEupatorium purpureumAbortifacient effects.
JuniperJuniperus communisAllergenic, cathartic in large doses, diuretic, increases uterine tone; possible anti-implantation, abortifacient and emmenagogue effects.
KavaPiper methysticumLoss of uterine tone.
KhellaAmmivisnagaEmmenagogue and uterine stimulant.
Knot grassPolygonum aviculareAbortifacient effects.
LavenderLavendula officinalisEmmenagogue effects.
LeptandraVeronicastrum virginicumTeratogenic effects.
LicoriceGlycyrrhiza glabraEmmenagogue effects.
Life rootSenica aureusEmmenagogue and teratogenic effects.
LobeliaLobelia infataLoss of uterine tone.
LovageLevisticum officinaleEmmenagogue effects.
Madagascar periwinkleVinca rosaAbortifacient effects.
MadderRubia tinctorumGenotoxic and emmenagogue effects.
Male fernOryopteris filix-masAbortifacient effects.
MarjoramOriganum marjoranaEmmenagogue effects.
Marsh teaLedum palustreAbortifacient effects.
MasterwoodHeracleum lanatumEmmenagogue effects.
MateIlex paraguayensisLow birthweight, birth defects, premature birth.
MistletoeViscum albumUterine stimulant.
MotherwortLeomurus cardiacaEmmenagogue effects.
MugwortArtemisia vulgarisEmmenagogue and abortifacient effects.
MyrrhCommiphora myrrhaEmmenagogue and abortifacient effects.
NutmegMyristica fragransAbortifacient and mutagenic effects.
PapainCarica papayaEmmenagogue and abortifacient effects.
PareiraChondodendron tomentosumEmmenagogue and abortifacient effects.
ParsleyPetroselinium sativumEmmenagogue and abortifacient effects.
PassionflowerPassiflora incarnataUterine stimulant.
Peach pitPrunus persicaEmmenagogue and abortifacient effects.
PennyroyalHedeoma pulegioides or Mentha pulegiumTraditionally used as an abortifacient, hepatotoxic and neurotoxic.
PeonyPaeonia officinalisEmmenagogue effects.
PeppermintMentha piperitaEmmenagogue effects.
PinePinusAbortifacient effects.
Pleurisy rootAsclepias tuberosaUterine stimulant.
PomegranatePunica granatumEmmenagogue and uterine stimulant effects.
Prickly ashZanthoxylum americanumEmmenagogue effects.
PulsatillaAnemone pulsatillaUterine stimulant.
Queen Anne's laceDancus carotaEmmenagogue and abortifacient effects.
RaspberryRubus idaeusStimulates contraction in strips of pregnant human uterus, antigonatrophic activity.
RhubarbRheum palmatumUterine stimulant, mutagenic, genotoxic effects.
RosemaryRosmarinus officinalisEmmenagogue and abortifacient effects.
RueRuta graveolusEmmenagogue and abortifacient effects.
SafflowerCarthamus tinctoriusEmmenagogue and abortifacient effects.
SaffronCrocus sativusEmmenagogue and abortifacient effects.
SageSalvia officinalisEmmenagogue and abortifacient effects.
SandalwoodSantalkum albumAbortifacient effects.
SassafrasSassafras albidumEmmenagogue effects.
SavinJuniperus sabinaAbortifacient effects.
Scotch broomCytisus scopariusAbortifacient effects.
ScullcapScutellaria laterifoliaMay inhibit pituitary and chorionic gonadotropins, as well as prolactin, liver damage in humans.
SenegaPolygala senegaEmmenagogue and uterine stimulant effects.
SennaCassiaEndometrial stimulation, mutagenic and genotoxic effects.
Shepherd's purseCapsella bursa-pastorisEmmenagogue and abortifacient effects.
St John's wortHypericum perforatumEmmenagogue and abortifacient effects.
Stinging nettleUrticaEmmenagogue and abortifacient effects.
TansyTanacetum vulgareEmmenagogue and abortifacient effects.
ThymeThymusEmmenagogue effects.
TurmericCurcuma longaEmmenagogue and abortifacient effects.
ValerianValeriana officinalisStimulates uterine contraction.
WatercressNasturtium officinaleEmmenagogue and abortifacient effects.
Wild cherryPrunus serotinaTeratogenic effects.
Wild gingerAsarum canadenseEmmenagogue and abortifacient effects.
Wild marjoramOriganum vulgareEmmenagogue and abortifacient effects.
Wood sorrelOxalis acetosellaEmmenagogue effects.
Worm seedChenopodium ambrosioidesEmmenagogue and abortifacient effects.
WormwoodArtemesia absinthiumEmmenagogue and abortifacient effects.
Yellow cedarThuja occidentalisEmmenagogue and abortifacient effects.

In many of the above case reports and case series, a cause–effect relationship has not been established beyond reasonable doubt. We are often dealing with isolated instances for which neither independent confirmation nor plausible mechanisms of action are available. With virtually all of the retrospective investigations quoted above, there is obviously the possibility of confounding by a wide range of factors. This begs the question whether such reports are really reliable indications of true safety problems or false positive accounts with no clinical relevance. On the basis of the data available to date, it seems impossible to provide a conclusive answer to this question. In spite of this uncertainty, this author believes that it is wise to err on the safe side, investigate this area systematically, and take the existing evidence (although it is undeniably flawed) seriously.

A further pertinent question relates to the attitude of herbal medicine and other experts towards herbal medicine product use during pregnancy and lactation. These risks are judged differently by different authors. A recent literature review, for instance, noted that 6% of the literature sources cited chamomile and peppermint as unsafe while 12% and 15% judged ginger and raspberry leaf as unsafe44. Thus it is not surprising that some conventional healthcare professionals employ or recommend herbal medicine products during pregnancy. The need for sound information, it seems, is urgent.

The potential benefits of herbal medicine products for pregnant or lactating women are difficult to assess and very few trials have been published45–47. Table 2 lists preliminary data from efficacy trials of herbal medicine products carried out in pregnant women33,48–55. Almost invariably, the results are positive but far from convincing; some of the findings, however, might merit further investigation. It is noteworthy that these trials were conducted mostly in developing countries almost exclusively with Asian herbal medicine products. The only reasonably well-researched herbal medicine product is ginger, which has been shown in a recent systematic review to be an effective treatment for nausea and vomiting of various aetiologies including morning sickness56. But even for this herbal medicine product there is insufficient positive safety data to justify recommendations for use during pregnancy14.

Table 2.  Controlled clinical trials of herbal medical products (HMP) in pregnant women. PIH = pregnancy-induced hypertension.
ReferenceHMPIndicationDirection of result
Zhang et al. (1994)48RhubarbPIHPositive
Liu et al. (1994)49Salvia miltiorrhizae LigustrazinePIHPositive
Takakuwa et al. (1996)50Sairei-To (Chang Ling-tang)Recurrent abortionPositive
Takakuwa et al. (1997)51Sairei-To (Chang Ling-tang)Recurrent fetal wastagePositive
Bian et al. (1998)52Complex Chinese herbal mixtureBlood group incompatibilityPositive
Parsons et al. (1999)33Raspberry leafShortens labourPositive
Sun et al. (1999)53Shou-Tai-TangPrevention of miscarriagePositive
Fujii et al. (1999)54Sairei-to Tokishakuyaku-sanRecurrent abortionPositive
Mustefa et al. (2001)55Khameera MarwareedPalpitations during pregnancyPositive

With such a degree of uncertainty, firm conclusions are difficult. The evidence available to date implies that some herbal medicine products are associated with risks. Yet only a few attempts have been made to define the risk of specific herbal medicine products and those that have been published lack statistical power to produce conclusive results30,31. It therefore seems imperative that this area be rigorously investigated. Until definitive data emerge, the best advice is to consider all herbal medicine products contraindicated during pregnancy/lactation and to inform our patients accordingly.

Ancillary