Editorials

lar complications in diabetes are fully prevented, but

the risk increases steadily with rising levels of risk
factors. The new analysis of the UKPDS data confirms
this notion for both glycaemia and blood pressure.7 10
The findings from the observational analysis and the
trial effects are concordant, which is reassuring and
emphasises the need for more effective control of glucose concentrations and blood pressure in patients
with type 2 diabetes. The lower the level of blood
glucose, HbA1c, or blood pressure, the lower the risk of
complications. Thus, artificial target levels are not necessarily useful since they may sometimes incorrectly
lead both patients and physicians to think that
reaching such levels fully protects against late complications of diabetes.
We know that it is difficult to maintain reductions in
glucose concentrations and blood pressure even when
using multiple pharmacological agents that in short
term trials have produced excellent results. This was
also confirmed in the UKPDS. Thus, the alternative
possibility would be to start treatment at lower levels
than those currently used as thresholds. The guidelines
for antihypertensive treatment have been drastically
shifted in this direction over the past decades.11 Guidelines for the treatment of hyperglycaemia should be
also evaluated from this perspective.
Disappointing results from lowering high concentrations of blood glucose may be due to the use of
inappropriate diagnostic tests. A large European
epidemiological study showed that the postprandial
glucose concentration is a better predictor of mortality
than is fasting glucose.12 Mortality was already
increased in people with impaired glucose tolerance.
The present UKPDS data show that the lower the concentration of glucose the lower risk of complications.11
Therefore, we must seriously ask whether treatment to
lower raised blood glucose should be started much
earlier. Perhaps impaired glucose tolerance should be

an indication for treatment. There is a need to carry

out controlled clinical trials to find out whether lowering glucose concentrations at the levels of impaired
glucose tolerance will reduce microvascular and
macrovascular complications.
Jaakko Tuomilehto professor
Diabetes and Genetic Epidemiology Unit, National Public Health
Institute, Mannerheimintie 166, 00300 Helsinki, Finland
([email protected])

Vijan S, Kent DM, Hayward RA. Are randomized controlled trials

sufficient evidence to guide clinical practice in type II (non-insulindependent) diabetes mellitus? Diabetologia 2000:43:125-30.
2 Collins R, Peto R, MacMahon S, Herbert P, Fiebach NH, Eberlein KA, et
al. Blood pressure, stroke, and coronary heart disease. 2. Short-term
reductions in blood pressure: overview of randomised drug trials in their
epidemiological context. Lancet 1994;335:827-39.
3 Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, et
al. Prevention of coronary heart disease with pravastatin in men with
hypercholesterolemia. N Engl J Med 195;333:1301-6.
4 Systolic Hypertension in the Elderly Program Cooperative Research
Group. Prevention of stroke by antihypertensive drug treatment in older
subjects with isolated systolic hypertension. JAMA 1991;265:3255-64.
5 Staessen J, Fagard RH, Thijs L, Celis H, Arabidze G, Birkenhger W, et al.
Randomised double-blind comparison of placebo and active treatment
for older patients with isolated systolic hypertension. Lancet
1997;350:757-64.
6 Bucher HC, Griffith LE, Guyatt GH. Effect of HMCoA reductase
inhibitors on stroke: a meta-analysis of randomised, controlled studies.
Ann Intern Med 1998;128:89-95.
7 Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS36): prospective observational study. BMJ 2000;321:412-9.
8 Tuomilehto J, Rastenyte D, Birkenhger WH, Thijs L, Antikainen R, Bulpitt CJ, et al. Effect of calcium-channel blockade in older patients with
diabetes and systolic hypertension. N Engl J Med 340:677-84.
9 Alberti KGMM, Gries FA, Jervell J, Krans HMJ for the European NIDDM
Policy Group. A desktop guide for the management of non-insulindependent diabetes mellitus (NIDDM): an update. Diabet Med
1994;11:899-909.
10 Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS35): prospective observational study.
BMJ 2000;321:405-12.
11 World Health Organization, International Society of Hypertension
Guidelines Subcommittee. Guidelines for the management of hypertension. J Hypertens 1999;17:151-83.
12 DECODE Study Group. Glucose tolerance and mortality: comparison of
WHO and American Diabetes Association diagnostic criteria. Lancet
1999;354:617-21.

Herbal medicines: where is the evidence?

Growing evidence of effectiveness is counterbalanced by inadequate regulation

A full list of
systematic reviews
of trials of herbal
medicines appears
on the BMJs
website
BMJ 2000;321:3956

BMJ VOLUME 321

ales of herbal medicines are booming. This is

particularly true in the United States, where the
market for herbal supplements is now approaching $4bn a year. The fastest growth has been recorded
for St Johns wort, a herbal antidepressant whose sales
increased in one year by 2800%.1 Faced with such
figures doctors are inclined to ask where the evidence
is. Are there rigorous trials to show that herbal
treatments work?
Single studies are unlikely to convince sceptics, but
an increasing body of evidence is now emerging from
systematic reviews and meta-analyses of randomised
clinical trials. These suggest that some herbal medicines
are efficacious. The increased demand for St Johns wort,
for instance, was triggered by press reports of a
meta-analysis of 23 randomised trials of 1757 patients
with mild or moderate depression. The authors
concluded that extracts of hypericum were significantly
more effective than placebo (odds ratio 2.67; 95% confidence interval 1.78 to 4.01) and as effective as
12 AUGUST 2000

bmj.com

conventional antidepressants (odds ratio 1.10; 93 to

1.31) in alleviating the symptoms of mild to moderate
depression.2 Since this article was published, at least nine
further randomised trials have appeared, all of which
confirm the efficacy of this herbal antidepressant.3
Systematic analyses of other herbal medicines
followed and drew similarly positive conclusions. A
review of all nine placebo controlled, double blind
randomised trials of ginkgo biloba for dementia,
covering 1497 patients, showed that ginkgo was more
effective than placebo in delaying the clinical course of
dementia.4 A meta-analysis of 18 randomised controlled
trials (2939 patients) of saw palmetto as a symptomatic
treatment for benign prostate hyperplasia showed that it
improved urological symptoms and flow measures
significantly more than placebo.5 Saw palmetto was as
effective as finasteride and had fewer adverse effects. A
systematic review of horse chestnut seed extracts for
chronic venous insufficiency included eight placebo
controlled and five comparative randomised trials with a
395

Editorials
total of 1083 patients.6 All placebo controlled trials were
positive and all comparative trials indicated equivalence
with other active therapies. The effects included a reduction in leg volume and leg circumference as well as
symptomatic improvements. In all, about 40 systematic
reviews or meta-analyses of herbal drugs are available
today (a full list provided by the author is available on
the BMJ s website).
With many of these herbal medicines we do not
fully understand how they work. Nor do we always
know which component is pharmacologically active.
For example, hypericin was originally thought to be
the active ingredient in St Johns wort, but evidence is
now accumulating that hyperforin may be equally
important.7 Similarly, we assumed until recently that its
mode of action was that of a monoamine oxidase
inhibitor, but its actions may be due, at least partly, to
serotonin uptake inhibition.3
Even though herbal remedies may be effective, do
their benefits outweigh the risks? Most herbal remedies
in the United Kingdom and United States are sold as
food supplements. Thus they evade regulation of their
quality and safety. The UKs minister for public health
recently pointed out that the regime for unlicensed
medicines does not give systematic protection to the
public against low quality and unsafe unlicensed herbal
remedies.8 Two recent British cases of severe
nephropathy caused by Chinese herbal tea administered to treat eczema9 illustrate this. Huge variations
exist in the quality of herbal medicinal preparations.
When, for example, German commercial products of
devils claw were tested, an unacceptable variability of
quality was noted.10 Yet Germany is often praised for
the exemplary standard of quality control of herbal
medicines. A recent study of herbal creams in the
United Kingdom showed that 8 of 11 preparations
contained undeclared dexamethasone at a mean
concentration of 456 mg/g.11
The possibility of herb-drug interactions is a
further importantand under-researchedissue. On
its own, for instance, ginseng has few serious adverse
effects. When combined with warfarin, its antiplatelet
activity might cause overanticoagulation.12 Many other
interactions between herbal remedies and synthetic
drugs are conceivable, even likely.13 This issue is
destined to play an increasingly important part in the
debate about the safety of phytomedicines.
With rationing looming in virtually all healthcare
systems, the question whether herbal medicines can
save money is important. Not all plant based medicines
are cheap. A standard daily dose of St Johns wort, for
instance, will cost more than that of a tricyclic

antidepressant. However, such comparisons are oversimplistic, particularly in view of the fact that St Johns
wort is associated with only about half the incidence of
adverse effects of a conventional antidepressant.2
As more and more herbal medicines are being
used by more and more people, doctors should
consider changing their often negative attitude towards
them. Doctors, pharmacists, and other healthcare professionals need to be knowledgeable to advise their
patients responsibly, and there is an unquestionable
need for reliable information on herbal medicines, a
demand that must be met adequately by undergraduate and postgraduate education. Doctors also have to
realise that detailed questions about use of herbal
drugs form an essential part of taking a medical
history. Finally, doctors should monitor the perceived
benefits and adverse effects of self prescribed herbal
treatments consumed by their patients and bear in
mind the possibility of herb-drug interactions. The
minister for public health has emphasised the need for
better protection and information for the public on
herbal medicines,8 and doctors should take an active
part in this process.
E Ernst professor
Department of Complementary Medicine, School of Postgraduate
Medicine and Health Sciences, University of Exeter, Exeter EX2 4NT
([email protected] )

1
2

3
4
5

8
9
10
11

12
13

Brevoort P. The booming US botanical market. A new overview.

Herbalgram 1998;44:33-46.
Linde K, Ramirez G, Mulrow CD, Pauls A, Weidenhammer W, Melchart D.
St Johns wort for depression: an overview and meta-analysis of
randomised clinical trials. BMJ 1996;313:253-8.
Stevinson C, Ernst E. Hypericum for depression: an update of the clinical
evidence. Eur Neuropharmacol (in press).
Ernst E, Pittler MH. Ginkgo biloba for dementia: a systematic review of
double-blind, placebo-controlled trials. Clin Drug Invest 1999;17:301-8.
Wilt TJ, Iskani A, Stark G, McDonald R, Lan J, Murlow C. Saw palmetto
extracts for treatment of benign prostatic hyperplasia. JAMA 1998;280:
1604-9.
Pittler MH, Ernst E. Horse-chestnut seed extract for chronic venous
insufficiency: a criteria-based systematic review. Arch Dermatol 1998;134:
1356-60.
Mller WE, Singer A, Wonnemann M, Hafner U, Schfer C. Hyperforin
represents the neurotransmitter uptake inhibiting constituent of hypericum extract. Pharmacopsychiat 1998;31(suppl):16-21.
Jowell T. Herbal medicines. House of Commons official report (Hansard) 1999
March 26;426-27.
Lord GM, Tagore R, Cook T, Gower P, Pusey CD. Nephropathy caused by
Chinese herbs in the UK. Lancet 1999;354:481-2.
Chrubasik S, Sporner F, Wink M. Zum Wirkstoffgehalt in Arzneimitteln
aus Hapargophytum procumbens. Forsch Komplementrmed 1996;3:57-63.
Keane FM, Munn SE, du Vivier AWP, Taylor NF, Higgins EM. Analysis of
Chinese herbal creams prescribed for dermatological conditions. BMJ
1999;318:563-4.
Janetzky K, Morreale AP. Probable interactions between warfarin and
ginseng. Am J Health Syst Pharm 1997;54:692-3.
Ernst E. Possible interactions between synthetic and herbal medicinal
products. Part 1: a systematic review of the indirect evidence. Perfusion
(in press).

The increasing use of peripheral bone densitometry

Better at assessing fracture risk than diagnosing osteoporosis

steoporotic fractures are a major cause of

excess mortality, morbidity, and expenditure
worldwide. There is a strong inverse relation
between bone mineral density and the risk of fracture,
with a doubling in fracture incidence for each standard
396

deviation reduction in bone mineral density.1 The

World Health Organization has defined osteoporosis
as a bone mineral density of more than 2.5 standard
deviations (T score < 2.5) below the mean value for
young adults.2 This definition was made for epidemioBMJ VOLUME 321

BMJ 2000;321:3968

12 AUGUST 2000

bmj.com