Preoperative use of herbal medicines: a patient survey

Abstract

Background. There has been recent concern in the media over the possible detrimental effects of herbal medicines on the perioperative period. Perceived by the public as ‘natural’ and therefore safe, herbal remedies may have led to adverse events such as myocardial infarction, bleeding, prolonged or inadequate anaesthesia and rejection of transplanted organs. In addition, herbal remedies can interact with many drugs given in the perioperative period. In this article we summarize the potential perioperative complications that can occur.

Methods. In order to determine the extent of use of herbal medicines, we conducted a survey of patients presenting for anaesthesia. During a 3‐month period, patients were directly asked by anaesthetic staff if they were currently self‐administering herbal medication.

Results. Of 2723 patients, 131 (4.8%) were taking one or more herbal remedy. In only two cases was this recorded in the patients’ notes. Women and patients aged 40–60 yr were most likely to be taking a herbal product (P<0.05 and P<0.001 respectively). The most commonly used compounds were, in descending order, garlic, ginseng, ginkgo, St John’s wort and echinacea.

Conclusion. Self‐administration of herbal medicines is common in patients presenting for anaesthesia. Because of the potential for side‐effects and drug interactions it is important for anaesthetists to be aware of their use.

Br J Anaesth 2002; 89: 792–5

Accepted for publication: July 5, 2002

There has been increasing concern in the American medical press about the potential complications arising in the perioperative period as a result of patients’ use of complementary medicines.1 The assumption by patients and doctors that these products ‘are natural and therefore safe’ is clearly dangerous. Morbidity and mortality are more likely in the perioperative period because of the polypharmacy and assaults on normal physiology that occur.2 Such complications may include myocardial infarction,3 bleeding,45 prolonged6 or inadequate anaesthesia7 and organ transplant rejection.8

The exact degree of cause for concern remains unclear. Very few of the huge number of herbal remedies have been formally researched and therefore most information is pooled from case reports and other anecdotal evidence.

In Table 1 we summarize the most commonly used herbal remedies together with their possible modes of action and perioperative complications.5^–7 9–24

There appears to be increasing public interest in the use of complementary medicine. In the USA the self‐prescribing of herbal medicines in the presurgical patient has been studied. Tsen and colleagues reported the use of herbal remedies in 22% of presurgical patients25 and in a similar study Kaye and colleagues reported 32%.26 More than 70% of these patients failed to disclose their use of herbal medicines to their attending anaesthetist.

It remains to be seen if the enthusiasm for such remedies is as high in this country. The aim of this study was to quantify the use of herbal remedies in patients attending for anaesthesia in a British district general hospital.

Method

During a 3‐month period, operating department practitioners and anaesthetic nursing staff questioned all patients about their current use of herbal remedies immediately before administration of anaesthesia. The questioning formed part of the routine preoperative checklist and asked if the patient was presently taking, or at some point in the last 2 weeks had taken, herbal medication.

A list of 16 commonly used medications was made available to the patient and questioner to act as a guide, as not all the questioners were familiar with herbal remedies. However, if the patient was taking another form of alternative medicine not mentioned on the list this was also recorded. Vitamins and minerals were not included in this survey. Non‐English speakers were excluded from questioning.

The questionnaire recorded age, sex and the name(s) of the herbal medicines currently taken. In the case of patients taking such remedies, their notes were examined retrospectively for any mention of the herbal medicine in the preassessment clerking or drug information section of the anaesthetic form.

Results were tabulated and analysed using appropriate descriptive statistical analysis. Differences with respect to the influences of patient sex and patient age on the prevalence of herbal medication use were tested by using χ² analysis. A P value of <0.05 was considered statistically significant.

Results

During the 3‐month period, 3349 patients underwent anaesthesia at Wexham Park Hospital. Forms were received from 2723, giving a response rate of 81.3%.

A total of 131 patients reported the use of herbal medications (4.8%). Eighty‐three patients were taking a single herbal remedy and 48 patients were taking more than one. Of these patients, 42 reported using two herbal medicines, four reported using three, one reported using four and one reported using five remedies. Female patients (80 out of 1387) used herbal medicines more frequently than males (51 out of 1336; χ²=5.65, P<0.05) (Table 2).

Our data indicate an influence of age on the prevalence of herbal medication use (Table 2).

The most commonly used medications reported are shown in Table 3. Garlic, ginseng, ginkgo, St John’s wort and echinacea were the most frequently reported remedies in order of highest to lowest.

In only two cases was there written documentation of the patient’s use of herbal remedies in the medical notes. In both cases the documentation was in the preassessment clerking. There were no records of the patients’ use of herbal preparations on any anaesthetic forms.

Discussion

Our survey showed that substantially fewer patients are self‐administering herbal remedies when compared with similar surveys carried out in the USA. However, it remains that a significant proportion of the presurgical population are taking potentially harmful medications without the knowledge of their anaesthetist.

Although the pharmacodynamics and pharmacokinetics of the majority of these remedies have yet to be fully clarified, reports of adverse events in the perioperative period suggest their importance and certainly that anaesthetists should become more aware of their use.

In the UK the majority of herbal medicines are exempt from the licensing requirements set out in Section 12 of the Medicines Act 1968.27 Approximately 20% of companies choose to seek a licence as a sign of a higher quality product. This process, however, is long and expensive and it can prove difficult to meet conventional requirements to prove product efficacy. In addition, herbal medicines cannot be patented, and this further removes the incentive to undertake costly research. In effect, this leaves the vast majority of herbal remedies exempt from the safety and efficacy requirements and regulations that prescription‐only and over‐the‐counter drugs must fulfil (i.e. preclinical animal studies, controlled clinical trials and post‐marketing surveillance). This current lack of regulation of herbal medicines also allows the potency of the herbal content to vary from manufacturer to manufacturer.28 This can therefore lead to significant variations in the pharmacological effects of a given remedy.

Despite the exemption from regulatory laws, in October 1996 the Yellow Card Scheme [the voluntary, spontaneous adverse drug‐reaction reporting scheme run by the Committee on Safety of Medicines (CSM/Medicines Control Agency UK)] was extended to include reporting of suspected adverse reactions to unlicensed herbal medicines. As of May 28, 2002, the CSM/Medicines Control Agency had received 1260 reports of adverse reactions that were suspected to be associated with herbal medicines via the Yellow Card Scheme. Of these, 31 had a fatal association (CSM, personal communication).

In addition, at the international level, 5000 suspected reactions were reported to the WHO before 1996,29 and between 1993 and 1998 a further 2621 adverse events, including 101 deaths, were reported to the US Food and Drugs Administration.30 Of concern is that there is no doubt that these figures are grossly underestimated, as medical staff are often ignorant of the pharmacology of these medications.

In our survey there was no documentation of herbal medications on the anaesthetic forms of patients using such products. We must assume that this was because either the patients or their anaesthetists did not consider them to be of importance. We believe that anaesthetists should elicit and document a full drug history, including the use of herbal remedies, in every patient.

Because pharmacokinetic data are lacking, the American Society of Anesthesiologists recommends that patients discontinue use of herbal medications 2–3 weeks before surgery.31 32 However, it remains that many patients will still be taking herbal remedies before surgery, either because they are unaware of this recommendation, or because they are presenting for non‐elective surgery. This means that all anaesthetists must familiarize themselves with the potential perioperative complications that can occur with the commonly used remedies. Anticipating a possible reaction is better than reacting to an unexpected condition.

Acknowledgements

We acknowledge the help of Miss Claire Davies (senior pharmacovigilance scientist) and Miss Lesley Curwen (scientific assessor, Pharmacoviligance Group Post Licensing Division) of the Medicines Control Agency.

Table 1

Pharmacological effects and potential perioperative complications of eight commonly used herbal remedies

Name of herb	Common uses	Pharmacological effects	Potential perioperative complications
Echinacea, purple coneflower root	Prophylaxis and treatment of viral, bacterial and fungal infections	Stimulation of the immune system.9 With long term use may be immunosuppressive10	Reduced effectiveness of immunosuppressants.11 Potential for wound infection with long‐term use.10 May cause hepatotoxicity especially when used with other hepatotoxic drugs12
Ephedra, ma huang	Diet aid	Indirectly and directly acting sympathomimetic13	Dose‐dependent increase in heart rate and blood pressure with potential for perioperative myocardial infarction and stroke.14 Arrythmias with halothane. Tachyphylaxis with intraoperative ephedrine
Garlic, ajo	Antihypertensive, lipid‐lowering agent, anti‐thrombus‐forming	Inhibits platelet aggregation (partially irreversibly) in a dose‐dependent manner.15 Lowers serum lipid and cholesterol levels16	May potentiate other platelet inhibitors. Concerns for perioperative bleeding. Concerns for neuraxial blockade
Ginkgo, maidenhair; fossil tree	Circulatory stimulant. Used to treat Alzheimer’s disease, peripheral vascular disease and erectile dysfunction	Inhibits platelet activating factor.17 Antioxidant.18 Modulates neurotransmitter activity	Concerns for perioperative bleeding.5 19 May potentiate other platelet inhibitors
Ginseng, ajo	To protect the body against stress and restore homeostasis	Poorly understood. Possible similar mechanism to steroid hormones. Inhibits platelet aggregation (partly irreversibly).20 Prolongs activated partial thromboplastin time.21	Potential to increase perioperative bleeding. Potential for hypoglycaemia
Kava, ava pepper	Anxiolytic	Possible potentiation of γ‐aminobutyric acid (GABA) transmission22	Potentiates sedative effects of anaesthetic agents.6 Possible withdrawal syndrome after sudden abstinence. Recent reports of kava‐induced hepatotoxicity to the medicines control agency
St John’s wort, goatweed, amber, hardhay	Treatment for depression and anxiety	Central inhibition of serotonin, noradrenaline and dopamine.23 Induction of cytochromes 450 3A4 and P450 2C97	Decreased effectiveness of cyclosporin, alfentanil, midazolam, lignocaine, calcium channel blockers and digoxin7
Valerian, vandal root, all heal	Anxiolytic and sleep aid	Potentiation of GABA neurotransmission24	Potentiates sedative effects of anaesthetic agents. Withdrawal‐type syndrome with sudden abstinence

Table 2

Age groups

Age (yr)	Taking herbal medication (A)	Not taking herbal medication	Total (B)	% (A/B)
≤30	18	655	673	2.7%
31–40	20	283	303	6.6%
41–50	34	358	392	8.7%
51–60	32	348	380	8.4%
61–70	17	435	452	3.8%
>70	10	513	523	1.9%

Table 3

Types of herbal medication used

Medication	Number using	% of all patients	% of total number of herbal medications used
Arnica	3	0.11	1.6
Echinacea	19	0.70	10.1
Ephedra	3	0.11	1.6
Garlic	43	1.58	22.9
Ginkgo	30	1.10	16.0
Ginseng	34	1.25	18.1
Kava	1	0.03	0.5
Primrose	11	0.40	5.9
Red clover	2	0.07	1.1
Saw palmetto	5	0.18	2.7
St John’s wort	27	0.99	14.4
Valerian	10	0.37	5.3
Total	188	4.81

Author notes

1Royal Berkshire Hospital, Reading, UK. 2Department of Anaesthesia, Wexham Park Hospital, Wexham Street, Slough, Berkshire SL2 4HL, UK*Corresponding author