Вы находитесь на странице: 1из 13


medicines have fewer or no side-effects, and

Complementary medicines in psychiatry many with chronic anxiety and depression
understandably feel disillusioned by the ap-
Review of effectiveness and safety parent ineffectivenessof conventional treat-
ment. The aim of this review is to acquaint
psychiatrists with the complementary med-
icines routinely encountered in clinical
practice, to review the evidence base for
'. their purported effectivenessand to discuss
potential adverse effects and interactions.

Background The use of Complementary medicines are either used
as an alternative or in addition to conven- We searched the Medline and Cochrane
complementary medicines inthose with
tional medicine (Zimmerman & Thomp- databases for evidence of the effectiveness
mental health problems iswell
son, 2002). Their use by those with of complementary medicines for the treat-
documented. However, their effectiveness
chronic disorders such as cancers, with ment of psychiatric conditions. We divided
is often not established and they may be their associated physical and psychological the substances into different categories:
less harmless than commonly assumed. problems, is well documented (Eisenberg cognitive enhancers, sedatives and anxioly-
et ai, 1993; Ernst & Cassileth, 1999). In tics, antidepressants, antipsychotics and
Aims To review the complementary psychiatric patients, estimates of their use remedies for movement disorders, and
medicines routinely encountered in range from 8 to 57%, with the most fre- anti-addictives. Search terms included the
psychiatric practice, their effectiveness, quent use being in depression and anxiety. identified substances in each category and
A population-based study from the USA EFFECTIVENESS or SIDE-EFFECTS or
potential adverse effects and interactions. found that 9% of respondents had anxiety ADVERSE DRUG REACTION or
Method Electronic and manual attacks, 57% of whom used complemen- INTERACTION. All recovered papers
tary medicines; 7% of respondents reported were reviewed for further relevant
literature search on the effectiveness and references. All evidence was collated and
severe depression, with 54% of these using
safety of psychotropic complementary complementary medicines (Kessler et ai, ranked as available. We also accessed
medicines. 2001). Another survey from the USA re- web-based resources, such as the Natural
ported mental disorders in 14% of respon- Medicines Comprehensive Database
Results Potentially useful substances dents, 21% of whom used complementary (http://www.naturaldatabase.com). and for
include ginkgo and hydergine as cognitive medicines (Unutzer et ai, 2000). Usage mularies, such as the PDR (Physicians'
enhancers, passion flower and valerian as was highest (32%) in respondents with Desk Reference for Herbal Medicines;
panic disorder. In studies restricted to those Medical Economics, 2000) for further in-
sedatives, St John's wort and s- formation on the identified substances.
with psychiatric disorders, usage ranged
adenosylmethionine as antidepressants, from 13 to 54% (Knaudt et ai, 2001; Wang Where available, we used reviews summar-
and selenium and folate to complement et aI, 2001; Alderman & Kiepfer, 2003; ising the proposed mechanism of action and
antidepressants. The evidence is less Matthews et aI, 2003). Complementary effectiveness, since presenting all the evi-
medicines are also used by those seen by dence in detail was beyond the scope of this
conclusive for the use of omega-3 fatty
liaison psychiatrists and a recent study of paper. Whenever possible, we gave priority
acids as augmentation treatment in cancer patients showed that 25% took to meta-analyses, systematic reviews and
schizophrenia, melatonin for tardive substances with psychoactive properties double-blind randomised controlled trials
dyskinesia and 18-methoxycoronaridine, (Werneke et aI, 2004a). (RCTs), but we also included other evi-
an ibogaine derivative, for the treatment Complementary medicines can be dence such as open trials and case reports
of cocaine and heroin addiction. grouped into herbal remedies, food supple- when the findings were relevant to our
ments, including vitamin preparations, and review. Where standardised comparative
Conclusions Systematic clinicaltrials other organic and inorganic substances, in- measures such as the Hamilton Rating
cluding omega-3 fatty acids. Some people Scale for Depression (HRSD; Hamilton,
are needed to test promising substances. take food supplements and vitamin pre- 1967) were available for meta-analyses,
Meanwhile, those wishing to take parations in high doses, often outside the we reported the relevant risk ratios. Owing
psychotropic complementary medicines safety margins recommended by the Food to the heterogeneity of the data, no attempt
require appropriate advice. Standards Agency (Food Standards Agency, at meta-analysis of other trials was made.
2003). People with mental health problems We included only studies applicable to gen-
Declaration of interest None. may take complementary medicines to treat eral adult psychiatry and psychiatry of old-
anxiety and depression or to counter side- er adults. Other special patient groups and
effects of conventional treatments, for ex- healthy volunteers were excluded, as were
ample tardive dyskinesia and weight gain. studies on a combination of substances
Some seek a more holistic approach to with evidence available for the single
treatment, others hope that complementary substance (Fig. 1).


Cognitive enhancers 296
Two thousand and seven studies were identi- Anxiolyticsand sedatives 227
Papers and documents possibly eligible 2007
Antidepressants and augmentation 1165
fied for the 20 remedies under review for the Antipsychotics,augmentation and
five categories of mental health problems. treatments for tardive dyskinesia 100
Excluded 1963
Anti-addictives 43
The literature ranged from case reports and
narrative reviews to systematic reviews in-
cluding meta-analyses. For four categories,
Exclusion criteria:
higher ranking evidence
the evidence with regard to efficacy could combinations of substances;
be limited to RCTs, systematic revie;.ys and special patient groups;
healthy volunteer swdies
meta-analyses. For anti-addictive substances
we also considered open trials, since there
was very lirtle evidence available (Fig. 1).
Studies included 44

Cognitive enhancers
Cognitive enhancers are either used in the
treatment of dementia to enhance mental
performance or to prevent cognitive decline
in healthy people. This can be achieved by Cognitive Anxiolytics and Antidepressants and Antipsychotics, Anti-addictives 4
enhancers 8 sedatives 8 augmentation 15 augmentation
increasing choline availability in the brain, RCTs 2
and treatments
e.g. by inhibiting acetylcholinesterase. SRs/MAs 4 SRs/MAs 3 SRs/MAs 8 for tardive OTRs 2
Alternative non-cholinergic neuroprotec- RCTs 4 RCTs 5 RCTs 7 dyskinesia 7
tive strategies have been postulated; these
SRslMAs 2
include antioxidants scavenging free radi- RCTs 5
cals, thereby reducing neurotoxicity, and
anticoagulants increasing cerebral blood
Fig. I Selection of efficacy studies. SRs. systematic reviews; MAs. meta-analyses; RCTs. randomised
flow (Spinella, 2001). Suggested herbal
controlled trials; OTRs. open trials.
cognitive enhancers include ginkgo, gin-
seng, hydergine, which is an ergot (Clavi-
ceps purpurea) derivative, and solanceous Anxiolytics and sedatives Kava has an anxiolytic effect that has been
plants, including potatoes, tomatoes and established in several RCTs (pittler &
aubergines (Table 1). Anxiolytics and sedatives essentially have Ernst, 2003). Kava has been associated
Although in some individuals with the same underlying mechanisms of action. with several cases of liver toxicity (Natural
Alzheimer's disease ginkgo biloba has been The stronger the agent the greater the seda- Medicines Database, 2004a), which has led
reported to improve cognitive performance tive effect, leading to coma in extreme to its voluntary withdrawal from the UK
(Birks et ai, 2002; Kanowski & Hoerr, cases. Four mechanisms of action have been market. Valerian (Valeriana officinalis or
2003), another trial did not show any ben- implicated (Spinella, 2001): (a) binding to Valeriana edulis) is a sedative believed to
efit in elderly people with Alzheimer's dis- gamma-aminobutyric acid (GABA) recep- have been known to Galen and Dioscorides,
ease of vascular type or age-associated tors leading to hyperpolarisation of the cell which has maintained its importance
cognitive impairment (van Dongen et ai, membrane through increased influx of throughout the centuries (Spinella, 2001).
2003). Whether the effect in those with Alz- cWorine anions; (b) inhibition of excitatory In 1845, Coffin described it as 'an excellent
heimer's disease is equivalent to that of syn- amino acids, thereby also impairing the sedative. . .predisposing the mind to quiet-
thetic cholinesterase inhibitors is debatable ability to form new memories; (c) sodium ness and the body to sleep'. Valerian may
(hil et ai, 1998; Wettstein, 2000; Schreiter- channel blockade, decreasing depolarisa- have comparable efficacy to oxazepam
Gasser & Gasser, 2001). Hydergine tion of the cell membrane; and (d) calcium (Dorn, 2000; Ziegler et ai, 2002). However,
was reported to lead to significant improve- channel blockade, decreasing the release a systematic review on the effectivenessof
ment of cognitive impairment in dementia, of neurotransmitters into the synaptic valerian in insomnia produced inconclusive
but most studies were performed before cleft. Most complementary medicines results (Stevinson& Ernst, 2000).
standardised dementia criteria were agreed prescribed for anxiolysis/sedation (e.g. kava Passion flower (Passiflora incamata)
(Olin et ai, 2001). The results for panax kava, valerian, passion flower and contains chrysin, a partial agonist to benzo-
ginseng and vitamin E were inconclusive chamomile) are GABAergic, though for diazepine receptors. One study comparing
(Sano et ai, 1997; Vogeler et ai, 1999). some such as hops the mechanism of action passion flower with oxazepam found both
Solanaceous plants may exercise strong remains unknown. As expected, all rem- to be equally effective (Akhondzadeh et
cholinergic effects by inhibiting not only edies can lead to drowsiness when taken ai, 2001a); more trials are needed to
acetyl- but also butyrylcholinesterase. in high doses and can potentiate the effect confirm the effect. No data are available
However, no clinical studies have been con- of synthetic sedatives (Table 2). on chamomile and hops. Chamomile
ducted to determine their effects on cogni- The most researched substance is kava (Chamaemelum nobile or Matricaria
tion. They may augment cocaine toxicity kava (Pipermethysticum), which originated recutita) contains apigenin which binds to
via the same mechanism (Krasowski et ai, from Polynesia and was traditionally used benzodiazepine receptors (Viola et ai,
1997). for religious rituals (Chevallier, 1996). 1995). The mechanism of action for hops

Table I Cognitive enhancers

Substance Postulated mechanism of action Effectiveness Side-effects Potential drug


Ginkgo Antioxidant: destroying free radicals Possible improvement of cognitive function, t Bleeding time. Case reports of Antithrombolytic agents (Medical
implicated in cell damage (Oken et aI, 1998; activities of daily living and mood in those intracerebral haemorrhage Economics, 2000)
Tabet et 01,2000); t cerebral blood flow with Alzheimer's disease or other cognitive (Matthews, 1998; Benjamin et 01, 2001);
through platelet activation factor inhibition decline, but recent results inconsistent (Birks possibly adverse effects on male and
and nitric oxide pathways (Maclennan et aI, etal, 2002; Kanowski & Hoerr, 2003; van female fertility (Ondrizek et al. 1999)
2002; Ahlemeyer & Krieglstein, 2003); Dongen etal, 2003)
cholinergic effects (Tang etal. 2002)
Panax ginseng Interference with platelet aggregation and Improvement of mental arithmetic and Insomnia, mania, hyper- and hypo- Insulin and oral hyperglycaemics, anti-
coagulation (Medical Economics, 2000); abstraction; age-delaying properties tension, t vaginal bleeding (Medical thrombombolytic agents, MAOls
neuroprotection through nicotinic activity unproven (Vogeler et al. 1999); one Economics, 2000) (phenelzine). loop diuretics (Medical
(Lewis et aI, 1999). antioxidant effects (Lee recent trial reporting marginal improvement Economics. 2000)
et aI, 1998) in Mini-Mental State Examination and

improvement in memory tests (Tian etal.

Hydergine t Cholinergic activity (Le Poncin-Lafitte Significant improvement in dementia patients; Cholinergic and monoaminergic Serotonergic antidepressants,
(ergoloid) et aI, 1985); reversal of age-related decline of hydergine was more effective at higher doses side-effects possible; risk of cholinesterase inhibitors

choline acetyltransferase (Dravid, 1983) and and in younger patients (Olin et 01,200 I) psychotic recurrence
muscarinic receptors (Amenta et aI, 1989);
modulation of all monoaminergic neuro-
transmitter systems (Markstein, 1985)
Vitamin E Antioxidant (Tabet et aI, 2000) Behavioural but no cognitive improvement High doses: t risk of bleeding due to Anticoagulants including aspirin, clodipro-
with combination of vitamin E with selegiline; antagonism of vitamin K-dependent gel, warfarin (Corrigan, 1982; Liedeetal, n
possibly delay of residential care using vitamin E clotting factors (Liede et aI, 1998) ~
1998) and herbal anticoagulants such as ."
(Sanoetal,1997) ginkgo, garlic and ginseng; possible preven- m
tion of nitrate tolerance (Watanabe et aI, Z
1997); possible t effect of sildenafil and »
related phosphodiesterase-5 inhibitors
Solanaceous plants Inhibit acetylcholinesterase and No study available Cholinergic poisoning through Prolonged action of pancuronium, other m
butyrylcholinesterase (Krasowski dietary intake possible myorelaxants and cocaine through t n
et aI, 1997) (Krasowski et al. 1997) inhibition of butyrylcholinesterase m
(Krasowski etal, 1997) Z
MAOls, monoamine oxidase inhibitors. -<

(Humulus lupulus) remains unclear. Bach However, four of these meta-analyses have psychosis.Rauwolfia (Rauwolfia serpentina)
flower remedies are a combination of 38 demonstrated equivalence to standard anti- extracts were traditionally used before
flowers and seem to have no effect (Ernst, depressants (Linde et aI, 1996; Whiskey et synthetic antipsychotics became widely
2002). Melatonin extracted from the pineal aI, 2001; Roder et aI, 2004; Linde et aI, available, several alkaloid derivatives, in-
gland may improve sleep in those with 2005). One recent trial using high doses cluding reserpine, being introduced in the
delayed sleep phase disorder, but no benefit (up to 1800 mg) of St John's wort reported 1950s (Malamud et aI, 1957). Rauwolfia
has been shown in the treatment of primary equivalence to paroxetine in those with originates from India and was mentioned
sleep disorder (MacMahon et aI, 2005). It moderate or severe depression (Szegedi et in Ayurvedic medicine around 700 BC(Che-
may also improve initial sleep quality in aI, 2005). Hyperforin, which inhibits the vallier, 1996). It blocks vesicular storage of
older adults with insomnia (Olde Rikkert reuptake of monoamines, is thought to monoamines, allowing them to be more
& Rigaud, 2001), but its role as a treatment be the most likely active component easily degraded by monoamine oxidases in
for insomnia in those with Alzheimer's dis- (Chatterjee et aI, 1998; Miiller et aI, the cytoplasm. As a consequence, the
ease remains disputed (Cardinali et aI, 1998). Thus, the use of extracts with amount of neurotransmitter available on
2002; Singer et aI, 2003). maximum hyperforin content should be depolarisation of the cell membrane is re-
examined more systematically (Werneke duced (Spinella, 2001), which may lead to
et aI, 2004b). a reduction in dopamine and the resolution
Antidepressants Folate and S-adenosylmethionine are in of psychotic symptoms. However, serotonin
and augmentation therapy the same biochemical pathway, with folate and noradrenaline will also be less avail-
All known synthetic antidepressants act via being required to synthesise methionine, the able, which explains why reserpine readily
the enhancement of serotonergic and nora- direct precursor of S-adenosylmethionine, precipitates depression. An alternative
drenergic neurotransmission. Most comple- from homocysteine. S-adenosylmethionine strategy is the augmentation of anti-
mentary antidepressants are thought to facilitates many methylation reactions psychotic treatment with omega-3 fatty
work through the same pathways (Tables required for the synthesis of many neuro- acids, but the results of clinical trials remain
3 and 4). The mechanism of action for sele- transmitters (Bottiglieri et aI, 2000; Morris inconclusive (Joy et aI, 2003; Table 5).
nium is not clear but does seem to be differ- et aI, 2003). Thus, those with high levels Attempts have been made to treat tard-
ent. Its antioxidant qualities may reduce of homocysteine may be more likely to ive dyskinesia with vitamin E. This treat-
nerve cell damage (Benton, 2002), and it become depressed, or possibly less likely ment strategy relies on the assumption
is also known to facilitate conversion from to respond to antidepressant treatment. that tardive dyskinesia not only results
thyroxine (T4) to thyronine (T3); T3 substi- Interestingly, hypothyroidism can lead to from dopamine receptor supersensitivity
tution is one possible means of augmenta- an increase in homocysteine levels (Roberts but is also related to the oxidative tissue
tion of antidepressants in conventional & Ladenson, 2004) and this may contri- damage of antipsychotic drugs (Shamir et
psychiatry. There are no clinical studies bute to the associated depression. In clinical aI, 2001; Lohr et aI, 2003). Meta-analysis
but low selenium levels have been asso- studies, folate has been reported to be effec- of ten small trials has indicated that vitamin
ciated with depression, anxiety and hosti- tive only when added to antidepressant E protects against deterioration of tardive
lity (Hawkes & Hornbostel, 1996), and therapy (Taylor et aI, 2004). dyskinesia (Soares & McGrath, 2001);
high dietary intake or supplementation Parenteral S-adenosylmethionine has one recent trial has reported improvement
has been associated with mood improve- been reported to be superior to placebo (Zhang et aI, 2004). A far more powerful
ment. The apparent therapeutic effect may (Bressa, 1994), and equivalence to imipra- antioxidant than vitamin E is melatonin,
be dose-dependent (Benton & Cook, 1991; mine has been demonstrated in two RCTs which attenuates the dopaminergic activity
Benton, 2002). Like lithium, there may be (Delle Chiaie et aI, 2002; Pancheri et aI, in the striatum as well as the release of
a narrow therapeutic index. A recent report 2002). The onset of action may be more dopamine from the hypothalamus (Shamir
by the Food Standards Agency (2003) re- rapid (Fava et aI, 1995). Oral S-adenosyl- et aI, 2001; Lohr et aI, 2003). However,
duced the recommended limits of safe daily methionine may require doses four times as with omega-3 fatty acids, clinical trials
intake. Trials may be most promising in as high as the parenteral formulation (Delle have been inconclusive (Shamir et aI,
those with a low baseline selenium level. Chiaie et aI, 2002). Finally, omega-3 fatty 2000, 2001), and larger trials are required
The most robust clinical data are acids are known to stabilise membranes to test its therapeutic effectiveness(Table 5).
available for St John's wort (Hypericum and to facilitate monoaminergic, serotoner-
perforatum). These have been extensively gic and cholinergic neurotransmission
reviewed in meta-analyses (Linde et aI, (Haag, 2003) but their antidepressant effect Although there are many addictive plants,
1996; Williams et aI, 2000; Whiskey et has not been convincingly demonstrated in few have been identified as having the
aI, 2001; Roder et aI, 2004; Werneke et clinical studies (Marangell et aI, 2003; Su potential to counter addiction (Table 6).
aI, 2004b; Linde et aI, 2005; Table 3) which et aI, 2003). Omega-3 fatty acids are poss- Such may be ibogaine, which is derived
have found a decrease in effect size over ibly effective when added to lithium in the from the West African shrub Tabemanthe
time when tested against placebo. The more treatment of bipolar affective disorder iboga. It has hallucinogenic properties and
recent meta-analyses mostly suggest that (Bowden, 2001; Table 4). is traditionally used in religious ceremonies
the effectivenessof St John's wort is limited and initiation rites, but has also been
to mild depression, and more homogenous Antipsychotics, augmentation claimed to counter nicotine, cocaine and
studies targeting patients with mild and treatment of tardive dyskinesia opiate addiction, via blockade of dopamine
depression are required (Roder et aI, 2004; Only two complementary medicines have release in the nucleus accumbens (and the
Werneke et aI, 2004b; Linde et aI, 2005). been suggested for the treatment of dopamine response in general) in chronic

Table 2 Anxiolytics and sedatives

Substance Postulated mechanism of action Effectiveness Side-effects Potential drug


Valerian GABAergic effects (Houghton. 1999) Inconclusive evidence. May improve sleep Cognitive impairment and drowsiness; case i Effect of sedatives. CYPA4 inhibitor (see
latency and slow wave sleep (Stevinson & reports of hepatotoxicity (Klepser & Table 3)
Ernst. 2000); no more effective than placebo Klepser. 1999)
in patients with acute sleep problems (Diaper
& Hindmarch. 2004); comparable efficacy to
oxazepam in patients with non-organic
insomnia (Dorn. 2000; Ziegler et al. 2002)
Passion flower Partial agonist to benzodiazepine Comparable efficacy to oxazepam in treat- Case report of severe nausea. vomiting. Anticoagulants. i effect of sedatives. CYP3A4
receptors (Wolfman et al. 1994) ment of general anxiety disorder drowsiness. prolonged QTc and episodes of inhibitor (see Table 3)
(Akhondzadeh et al. 2001a) non-sustained ventricular tachycardia (Fisher
et al. 2000); may contain cyanogenic glyco-
sides (Jaroszewski et al. 2002)

Kava GABAergic effects (Jussofie et al. 1994; Meta-analysis of nine studies showed At least 68 cases of liver toxicity (NMCD. CYPIA2. 2C9. 2C19. 2D6. 3A4 and 4A9/11
Dinh et al. 200 I). D2 antagonist significant reduction of HRSA score 20040); one case report of movement inhibition (Mathews et al. 2002)
(Schelosky et al. 1995) compared with placebo (weighted mean disorder (Meseguer et al. 2002)
difference=5.0. 95 CILI-8.8; P=O.OI)
(Pittler & Ernst. 2003); one trial report
of equivalence to buspirone and opipramole
(Boerner et al. 2003); one trial report of
improvement of sleep in patients with
anxiety disorder (Lehrl. 2004)

Chamomile Birds to benzodiazepine receptors (Viola No study available Contains coumarins: increase of bleeding Anticoagulants. i effect of sedatives. CYP3A4
et al. 1995) time inhibitor (see Table 3)

Hops Possibly oestrogenic mechanism (Medical Comparable efficacy of a hops-valerian None reported i Effect of sedatives
Economics. 2000) preparation to bromazepam in the n
treatment of sleep disturbance (Schmitz & J:
Jackel. 1998); no studies on hops alone ...
available J:
Bach flower Unclear Unclear ...
38 herbs with postulated differential Not effective (Ernst. 2002) »
remedies effects. 5 herbs in rescue remedies
Melatonin Regulation of the body's circadian rhythm. Possibly effective in delayed sleep phase Daytime sleepiness (Herxheimer & Petrie. Anticoagulants. i effect of sedatives C
(N-acetyl-5- endocrine secretions and sleep pattern; i disorder. no clear-cut effect in primary 2003); i depressive mood (Carman et al. (Herxheimer & Petrie. 2003) z
methoxy- GABA binding (Munoz-Hoyos et al. 1998) insomnia (MacMahon et al. 2005); inconclusive 1976) m
tryptamine) evidence for effectiveness as insomnia treat-
ment in Alzheimer's disease (Cardinali et al. ."
2002; Singer et al. 2003); may improve initial -<
sleep quality in older adults with insomnia :I:
(Olde Rikkert & Rigaud. 2001) j;
w -<
GABA, gamma-aminobutyric acid; CYP3A4. cytochrome P3A4; HRSA. Hamilton Rating Scale for Anxiety; NMCD. Natural Medicines Comprehensive Database.
". m
Table 3 Antidepressants and augmentation: St John's wort

Postulated mechanism Effectiveness' Side-effects Potential drug interactions

of action

MAOI inhibition and GABAergic activity Six meta-analyses v. placebo using HRSD scores with Similar to SSRls, photosensitivity Serotonergic antidepressants; CYP3A4, IA2 and
(Cott, 1997), monoamine reuptake (Perovic trend toward reduced effect size Linde etal, 1996: (Whiskey etal, 2001) 2C9 induction: HIV protease inhibitors, HIV non-
& Muller, 1995; Neary & Bu, 1999), OR=2.67 (1.78-4.01); Williams et ai, 2000: RR=1.9 nucleoside reverse transcriptase inhibitors,
upregulation of 5HT'A and 5HT 2Areceptors (1.2-2.8); Whiskey etal, 200 I: RR=1.98 (1.49-2.62); warfarin, cyclosporin, oral contraceptives, anti-
(Teufel-Mayer & Gleitz, 1997); modulation Werneke et ai, 2004b: RR=I.73 (1.40-2.14); Roder et ai, 2004: convulsants, digoxin and theophylline (Committee
of cytokine production (Thiele et ai, 1994) RR= 1.5I (1.28-1.75)2; Linde et ai, 2005: major depression - on Safety of Medicines & Medicines Control
RR=2.06 (1.65-2.59) (smaller trials), 1.15 (1.02-1.29) (larger Agency, 2000)
trials), not restricted to major depression - RR=6.13 (3.63-
10.38) (smaller trials), 1.71 (1.40-2.09) (larger trials), Linde
et ai, 2005: all studies,' RR=1.71 (1.40-2.09)
Four meta-analyses v. standard antidepressants Linde et ai,
1996: OR=I.IO (0.93-1.31) compared with TCAs or
maprotiline; Whiskey et ai, 200 I: RR= 1.00 (0.93-1.09)
compared with TCAs, maprotiline or SSRls; Roder et ai, 2004:
RR=0.96 (0.85-1.08) compared with TCAs, maprotiline or
SSRls and RR=0.85 (0.75-0.97) in mild or moderate
depression; Linde et ai, 2005: RR=I.OI (0.93-1.10) for all trials,
RR= 1.03 (0.93-1.14) compared with TCAs or maprotiline,
RR=0.98 (0.85-1.12) compared with SSRls

RCT, Szegedi et ai, 2005: RR=1.I8 (0.98-1.42) compared with


I. 95% Cl in parentheses.
2. Calculated by changing the baseline to enable comparison with the other meta-analyses: original RR=0.66 (0.57-0.88).
3. Calculated from all studies (random effect size) and not included in original paper; subgroup analysis reports fixed effect size.
4. Calculated from the reported figures of treatment responses, not included in original paper.
MAOI, monoamine oxidase inhibitors; GABA, gamma-aminobutyric acid: 5HT, 5-hydroxytryptamine; HRSD, Hamilton Rating Scale for Depression: OR, odds ratio: RR, risk ratio; TCAs, tricyclic antidepressants; SSRls, selective serotonin
reuptake inhibitors: RCT, randomised controlled trial; CYP3A4, cytochrome P3A4.
Table 4 Antidepressants and augmentation: supplements

Substance Postulated mechanism of action Effectiveness Side-effects Potential drug


Selenium Antioxidant; brain had a preferential No study available Acute toxicity: nausea causes vomiting, nail Potential! effect of
affinity for selenium (Benton, 2002). changes, irritability and weight loss; chronic simvastatin/niacin
Selenium facilitates the conversion ofT4 toxicity: resembles arsenic combination used with a selenium/beta-caro-
intoTI (Sher, 2001) toxicity (Werneke, 2003) tene/vitamins C and E supplement (Brown et aI,
2001). Other statins may also be affected
Folicacid Cofactor in neurotransmitter synthesis: Two pooled studies using HRSD and Caution in pernicious anaemia: not to be ! Effect of methotrexate, primidone, pheno-
methylation homocysteine to methionine, addition of folic acid to antidepressants given without vitamin B,,; large doses can barbital and pyrimethamine (NMCD, 2004(1)
the immediate precursor of S-adenosyl- RR=0.47 (0.24-0.92); effect possibly lead to agitation, insomnia, confusion and
methionine (Bottigleri et 01, 2000; Morris dose-dependent; folic acid alone not increased seizure frequency
et 01, 2003) effective (Taylor et 01,2004)
S-adenosyl- As above Parenteral S-adenosylmethionine Induction of mania in patients with bipolar Serotonergic antidepressants
methionine superior to placebo (Bressa, 1994); affective disorder (Friedel et aI, 1989;
comparable efficacy to imipramine in Mischoulon & Fava, 2002); significantly better
two RCTs (Delle Chiaie et 01, 2002; tolerated than TCAs (Delle Chiaie et 01,2002;
Pancheri et 01, 2002); oral S-adenosyl- Mischoulon & Fava, 2002; Pancheri etal, 2002);
methionine requires high dose more rapid onset of effect (Fava et aI, 1995)
(1600 mg; Delle Chiaie et aI, 2002)
Omega-3 fatty Influences catecholaminergic, serotonergic Results of two small trials with short r INR with high or changing doses r Effect of warfarin, aspirin and non-steroidal
acids and cholinergic neurotransmission, end-points inconclusive (Marangell (Fugh-Bergman, 2000) anti-inflammatory drugs (Fugh-Bergman, n
modulation of signal transmission et 01,2003; Su et 01,2003); successful 2000); serotonergic antidepressants? :I
mechanisms in neuronal membranes, augmentation of antidepressant r-
modulation of prostaglandins and ion treatment reported (Nemets et aI, m
channels (Haag, 2003) 2002); r of remission period when added -I
to lithium in bipolar affective disorder -<
(Stoll et aI, 1999) m
T4, thyroxine; n, thyronine; HRSD, Hamilton Rating Scale for Depression; RR, risk ratio; RCTs, randomised controlled trials; TCA, tricyclic antidepressants; INR, international normalised ratio; NMCD, Natural Medicines Comprehensive n
Database. Z
III -<
0. '"

Table 5 Antipsychotics, augmentation and treatment of tardive dyskinesia

Substance Postulated mechanism of aaion Effeaiveness Side-effeas Potential drug interaaions

Rauwolfia ! Dopamine availability: blocks vesicular More effective than placebo in patients with Depression, seizures, extrapyramidal r Effea of antipsychotics and barbiturates,
(includes storage of monoamines, which can then be chronic schizophrenia with regard to general reaaions, blood pressure changes and severe bradycardia with digitalis glycosides,
reserpine) degraded by monoamine oxidases (Spinella, mental state and behavioural disturbance heart rate (NMCD, 2004e) ! effect of levodopa, hypertension in
2001) (Malamud et aI, 1957); adjunaive treatment to combination with sympathomimetics
antipsychotics (Wolkowitz, 1993) (Medical Economics, 2000)
Omega-3 fatty See Table 4 Meta-analysis of five small studies inconclu- See Table 4 See Table 4
acids sive; the use of omega-3 fatty acids remains
experimental until larger trials are conducted
(Joy et aI, 2003)
Melatonin Antioxidant, also attenuates dopaminergic Two small RCTs led to inconclusive results; See Table 2 See Table 2
aaivity in the striatum and dopamine duration of tardive dyskinesia and melatonin
release from the hypothalamus (lohr et aI, dosage may influence treatment outcome
2003; Shamir et aI, 2001) (Shamir et aI, 2000, 2001)
Vitamin E Antioxidant Meta-analysis of ten small trials of uncertain See Table I See Table I
quality indicate that vitamin E proteas against
deterioration of tardive dyskinesia but there is
no evidence that vitamin E improves symp-
toms (Soares & McGrath, 2001). One recent
trial reports significant improvement of ab-
normal involuntary movements (Zhang et aI,

RCTs, randomised controlled trials; NMCD, Natural Medicines Comprehensive Database.

Table 6 Anti-addictives

Substance Postulated mechanism of action Effectiveness Side effects Potential drug interactions

Ibogaine and its derivative Blocks sensitised dopamine response One small open trial only: tested in Cholinesterase inhibitor: can lead to Cholinergic and anticholinergic drugs
18-MC for nicotine, cocaine to chronic morphine and cocaine seven with opiate misuse: three cholinergic toxicity (NMCD, 20040; (NMCD, 20040
and opiate addiction administration (Maisonneuve & Glick, remained abstinent at 14 weeks bradycardia (Maisonneuve & Glick,
2003); may alter morphine induced (Sheppard, 1994); otherwise evidence limited 2003); ibogaine but not 18-MC:
dopamine release in nucleus accumbens to case studies (Spinella, 200 I) significant cerebellar toxicity
(Maisonneuve & Glick, 2003); binds to the 18-MC only tested in animal experiments; (Maisonneuve & Glick, 2003)
cocaine site of the serotonin transporter ! morphine, cocaine and alcohol intake in rats

(Staley et ai, 1996); 18-MC binds to the (Rezvani et ai, 1997; Glick et ai, 2000)

NMDA receptor (Mash etal, 1995)

Valerian for benzodiazepine See Table 2 May improve sleep in patients withdrawing See Table 2 See Table 2
addiction from benzodiazepines (Poyares et ai, 2002);
no controlled trials available
Passion flower for cannabis, Cannabis: unclear, benzoflavones may Opiates: effective as adjuvant therapy See Table 2 See Table 2
benzodiazepine, nicotine and decrease tolerance and withdrawal to clonidine demonstrated in one small
opiate addiction symptoms (Dhawan et ai, 2002a); opiates: RCT (Akhondzadeh et ai, 200 Ib)
use of anxiolytic effect, see Table 2
St John's wort for alcohol Not fully understood, and possibly related ! Alcohol craving demonstrated in See Table 3 See Table 3
addiction to reducing serotonin, dopamine, animal experiments only (De Vry et ai, 1999;
noradrenaline and GABA reuptake Rezvani et ai, 1999; Overstreet et ai, o
(Rezvanietal,1999,2003) 2003a,b); effect may be dose-dependent "V
Kudzu for alcohol m
Ingredient puerarin, counteracts the Only one small trial in humans available None reported (NMCD, 2004g) Theoretically with anticoagulants, 3:
addiction anxiogenic effects associated with alcohol showing no difference to placebo (Shebek & aspirin, cardiovascular agents and Z
withdrawal; may also have flumazenil-like Rindone, 2000) hypoglycaemic drugs (NMCD, »
properties (Overstreet et ai, 2003b) 2004g) :I
18-MC, 18-methoxycoronaridine; NMDA, N-methyl-D-aspartate; GABA, gamma-aminobutyric acid; RCT, randomised controlled trial; NMCD, Natural Medicines Comprehensive Database.

cocaine and opiate users (Maisonneuve & RCTs are required to assess other those from a wide variety of cultural
Glick, 2003). Ibogaine also binds to the promising agents such as selenium and backgrounds. This may lead to higher
cocaine site of the serotonin transporter S-adenosylmethionine for the treatment acceptance and adherence compared with
(Staley et ai, 1996), but its therapeutic of depression, ideally in individuals conventional drugs, making it important
value is limited as it is highly neurotoxic showing the corresponding deficiencies to be 'willing and prepared to work in part-
and can cause irreversible cerebellar at baseline. This may lead to new thera- nership with patients' beliefs and prefer-
damage (Maisonneuve & Glick, 2003); as peutic approaches for treatment-resistant ences - provided their actions are safe'
a result, further clinical stUdies have been depression. Valerian and passion flower (Brugha et ai, 2004). Also, we do not know
abandoned. A synthetic derivative 18- should be tested as anxiolytics and seda- whether the agreed use of complementary
methoxycoronaridine has similar reported tives; their potential value in the treat- medicines could in itself improve insight
effects but no cerebellar toxicity or specific ment of addiction also requires further and subsequently lead to greater adherence
effects on the serotonin transporter clarification. The role of omega-3 fatty to conventional treatment regimens. This
(Maisonneuve & Glick, 2003). To date acids as an adjunct to antipsychotics emphasises the importance of further re-
18-methoxycoronaridine has only been and melatonin as a treatment or prophy- search on complementarymedicinesfocusing
tested in animal experiments where it has lactic agent for tardive dyskinesia remain on promising agents such as passion flower,
been shown to reduce cocaine, morphine ambiguous, both requiring trials with valerian and S-adenosylmethionine, which
and alcohol intake in rats (Rezvani et ai, sound methodology. appear to be obvious candidates for further
1997; Glick et ai, 2000). We have outlined only a limited range RCTs. In addition, it might be important to
Passion flower has also been used to of complementary medicines used for the consider patients' attitUdes and preferences
ameliorate the effects of opiate, cannabis, treatment of common psychiatric problems. in future studies, possibly targeting those
benzodiazepine and nicotine addiction, Clearly there are many more remedies that demanding complementary medicines.
but clinical data are limited (Dhawan et may be taken to improve general health or Finally, clinicians need to be aware of
ai, 2002a,b, 2003; Akhondzadeh et ai, to counter the side-effects of conventional side-effects associated with complementary
2001b). Likewise, valerian has been tried treatments. Clinicians need to be aware of medicines and any interactions with other
in benzodiazepine withdrawal (Poyares et and enquire about such forms of self- treatments. They should be able to identify
ai, 2002) and StJohn's wort has beenused medication, since all remedies may interact hazards, advising patients accordingly and
for the treatment of alcohol dependence with prescribed medication or have avoiding uncritical encouragement of po-
(De Vry et ai, 1999; Rezvani et ai, 1999; associated side-effects in their own tentially harmful use. Ignorance in this
Overstreet et ai, 2003b), but effectiveness right. For instance, patients may take area, given the independent usage of com-
has not been established. Kudzu, Japanese phyto-oestrogens, such as black colosh plementary medicines, may lead to criticism
arrowroot (Pueraria lobata), has tradition- (Actaea racemosa), wild yam (Dioscorea and possibly litigation (Cohen & Eisenberg,
ally been used for the treatment of alcoholic composita) or dong quai (Angelica sinensis) 2002). Equally, patients should be encour-
hangover. The active ingredient, purerarin, to counter sexual side-effects, and this aged to disclose information about
counteracts the anxiogenic effects asso- might pose a problem in patients with complementary medicines to healthcare
ciated with alcohol withdrawal (Overstreet oestrogen receptor-positive breast cancer professionals. These discussions need to be
et ai, 2003a). Kudzu also contains two po- (Werneke et ai, 2004a). For the same conducted sensitively in order to avoid alie-
tent, reversible inhibitors of human alcohol reason, patients may also try evening prim- nating patients who may feel that they have
dehydrogenase isozymes (Keung, 1993), rose oil (Oenothera biennis), which could not been taken seriously or have been criti-
but an effect has only been demonstrated decrease the effect of sodium valproate cised for using complementary medicines.
in vitro (Lin & Li, 1998). One small trial (Miller, 1989). Kelp (Laminaria digitata Such discussions can be complex and may
among those with chronic alcohol misuse or Fucus vesiculosus) may be taken to demand more time than is available in
has not shown any difference from placebo counter weight gain, but can contain sub- routine clinics. Service models need to be
(Shebek& Rindone, 2000). Further trials stantial amounts of iodine and can interfere designed to meet this challenge, with con-
are required to test its genuine therapeutic with treatment for thyroid function sideration being given to specialist clinics
potential, perhaps using more standardised disorders. Iodine taken together with providing regular updated advice to both
formulations of the active ingredient. lithium may have additive hypothyroid clinicians and patients.
effects (Natural Medicines Comprehensive
Database,2004b). ACKNOWLEDGEMENT
Given the complex pattern of potential We thank Mr Oded Horn for his assistancewith
Our review demonstrates that the evidence interactions, clinicians should not be afraid interpretation of the meta-analyses.
base for the use of psychotropic comple- to discuss complementary medicines with
mentary medicines is extremely limited. their patients. Although some patients REFERENCES
The best evidence is available for St may choose to use complementary medi-
John's wort and kava kava, both of cines as alternatives to conventional treat- Ahlemeyer, B. & Krieglstein, J. (2003)
Neuroprotectiveeffectsof Ginkgobilobaextract.
which are used extensively in various ment, many may decide to use them in CellularandMolecularLifeSciences,60, 1779-1792.
cultures. However, trials of St John's addition to prescribed medications. Com-
wort need improved definition of inclu- plementary medicines have - rightly or Akhondzadeh, S., Naghavi, H. R., Vazirian, M., et 01

sion criteria (Werneke et ai, 2004b), wrongly - a very positive 'natural' reputa- (20010) Passionflower in the treatment of generalized
anxiety: a pilot double-blind randomized controlled trial
and kava kava has been withdrawn due tion among significant sections of the popu- with oxazepam. journal of Clinical Pharmacy and
to concerns about hepatotoxicity. Further lation, and therefore can be popular with Therapeutics, 26, 363-367.


Akhondndeh, S., Kashani, L., Mobaseri, M., et 01 Cott, J. M. (1997) In vitro receptor binding and enzyme Fugh.Bennan, A. (2000) Herb-drug interactions.
(200Ib) Passionflower in the treatment of opiates inhibition by hypericum perforatum extract. Lancet. 355, 134-138.
withdrawal: a double-blind randomized controlled trial. Pharmacopsychiatry. 30 (suppl.).108-112.
Glick, S. D., Maisonneuve, I. M. & Dickinson, H. A.
journalof Clinical Pharmacy and Therapeutics. 26, 369-373. De Vry, J., Maurel, S., Schreiber, R., et 01 (1999) (2000) 18-MC reduces methamphetamine and nicotine
Aldennan, C. P. & Kiepfer, B. (2003) Complementary Comparison of hypericum extracts with imipramine and self-administration in rats. Neuroreport, 11,2013-2015.
medicine use by psychiatry patients of an Australian fluoxetine in animal models of depression and alcoholism.
Haag, M. (2003) Essential fatty acids and the brain.
hospital. Annals of Pharmacotherapy. 37, 1779-1784. European Neuropsychopharmacology. 9.61-68.
Canadian journal of Psychiatry.48. 195-203.
Amenta, F., Cavallotti, C., Franch, F., et 01 (1989) Delle Chiaie, R., Pancheri, P. & Scapicchio, P. (2002)
Muscarinic cholinergic receptors in the hippocampus of Efficacy and tolerability of oral and intramuscular 5- Hamilton, M. (1967) Development of a rating scale for
the aged rat: effects of long-term hydergine adenosyl-L-methionine 1.4-butanedisulfonate (SAMe) in primary depressive illness. British journal of Social and
administration. Archives Intemationales de the treatment of major depression: comparison with Clinical Psychology.6, 278-296.
Pharmacodynamie et de Therapie, 297, 225-234. imipramine in 2 multicenter studies. Americanjournalof Hawkes, WoC. & Hornbostel, L. (1996) Effects of
Benjamin, J., Muir, T., Briggs, K., et 01 (2001) A case ClinicalNutrition.76 (suppl.).11725-11765. dietary selenium on mood in healthy men living in a
of cerebral haemorrhage - can Ginkgo biloba be Dhawan, K., Kumar, S. & Sharma, A. (20020) metabolic research unit. BiologicalPsychiatry.39, 121-128.
implicated? Postgraduate Medical journal. 77. 112-113. Reversal of cannabinoids (delta9-THC) by the Herxheimer, A. & Petrie, K. J. (2003) Melatonin for
benzoflavone moiety from methanol extract of
Benton, D. (2002) Selenium intake. mood and other the prevention and treatment of jet lag.CochraneLibrary.
Passiflora incarnata Linnaeus in mice: a possibletherapy
aspects of psychological functioning. Nutrition and issue 4. Oxford: Update Software.
Neurasciences. 5. 363-374. for cannabinoid addiction.journalof Pharmacyand
Pharmacology. 54. 875-881. Houghton, P.J. (1999) The scientific basis for the
Benton, D. & Cook, R. (1991) Selenium reputed activity of Valerian.journalof Pharmacy and
Dhawan, K., Kumar, S. & Sharma, A. (2002b)
supplementation improves mood in a double-blind Pharmacology. 51.505-512.
Nicotine reversal effects ofthe benzoflavone moiety
crossover trial. Biological Psychiatry. 29. 1092-1098.
from Passiflora incarnata Linneaus in mice.Addiction Itil, T. M., Ahmed, I., Kunitz, A., et 01(1998) The
Birks, J., Grimley Evans J. & Van Dongen, M. (2002) Biology.7,435-441. pharmacologicaleffects of ginkgo biloba, a plant extract.
Gingko biloba for cognitive impairment and dementia. on the brain of dementia patients in comparison with
Cochrane Library. Issue 2. Oxford: Update Software.
Dhawan, K., Dhawan, S. & Chhabra, S. (2003)
Attenuation of benzodiazepine dependence in mice by a tacrine. Psychopharmacology Bulletin.34,391-397.
Boerner, R. J., Sommer, H., Berger, Wo, et 01 (2003) tri-substituted benzoflavone moiety of Passiflora Jaroszewski, J. Wo,Olafsdottir, E. S., Wellendorph, P.,
Kava-Kava extract LlI50 is as effective as opipramol and incarnata Linneaus: a non-habit forming anxiolytic. et al (2002) Cyanohydrin glycosides of Passiflora:
buspirone in generalised anxiety disorder - an 8-week journalofPharmacy andPharmacological Science.6, distribution pattern, a saturated cyclopentane derivative
randomized. double-blind multi-centre clinical trial in 215-222. from P. guatemalensis. and formation of
129 out-patients. Phytomedicine. 10 (suppl.), 38-49.
Diaper, A. & Hindmarch, I. (2004) A double-blind. pseudocyanogenic alpha-hydroxyamides as isolation
Bottiglieri, T., Laundy, M., Crellin, R., et 01 (2000) placebo-controlled investigation of the effects of two artifacts. Phytochemistry. 59. 501-511.
Homocysteine, folate. methylation. and monoamine doses of a valerian preparation on the sleep. cognitive Joy, C. B., Mumby-Croft, R. & Joy, L. A. (2003)
metabolism in depression. journal of Neuralogy. and psychomotor function of sleep-disturbed older Polyunsaturated fatty acid supplementation for
Neurosurgery and Psychiatry. 69. 228-232. adults. Phytotherapy Research. 18,831-836. schizophrenia. Cochrane Library. issue 4. Oxford: Update
Bowden, C. L. (2001) Novel treatments for bipolar Dinh, L. D., Simmen, U., Bueter, K. B., et 01 (2001) Software.
disorder. Expert Opinionin InvestigatingDrugs. 10.661-671. Interaction of various Piper methysticum cultivars with
Jussofie, A., Schmiz, A. & Hiemke, C. (1994)
Bressa, G. M. (1994) S-adenosyl-I-methionine (SAMe) CNS receptors in vitro. Planta Medica. 67. 306-311.
Kavapyrone enriched extract from Piper methysticum
as antidepressant: meta-analysis of clinical studies. Acto Dorn, M. (2000) Efficacy and tolerability of Baldrian as modulator of the GABAbindingsite in different
Neurologica Scandinavica. 154 (suppl.).7-14.
versus oxazepam in non-organic and non-psychiatric regions of rat brain. Psychopharmacology.116,469-474.
Brown, B. G., Zhao, X. Q., Chait, A., et 01 (2001) insomniacs: a randomised. double blind, clinical
Simvastatin and niacin, antioxidant vitamins, or the comparative study. ForschendeKomplementarmedizin und Kanowski, S. & Hoerr, R. (2003) Ginkgo biloba
Klassische Naturheilkunde. 7. 79-84. extract EGb761 in dementia: intent-to-treat analyses of
combination for the prevention of coronary disease.
a 24-week, multi-center. double-blind, placebo-
New England journal of Medicine. 29. 1583-1592. Dravid, A. R. (1983) Deficits in cholinergic enzymes controlled,randomizedtrial.Pharmacopsychiatry. 36,
Brugha, T., Rampes, H. & Jenkins, R. (2004) Surely and muscarinic receptors in the hippocampus and 297-303.
you take complementary and alternative medicines? striatum of senescent rats: effect of chronic hydergine
Psychiatric Bulletin. 28. 36-39. treatment. Archives Internationales Pharmacodynamie et Kessler, R. C., Soukup, J., Davis, R. B., et al (2001)
de Therapie. 264. 195-202. The use of complementary and alternative therapies to
Cardinali, D. P., Gvozdenovich, E., Kaplan, M. R., et 01
treat anxiety and depression in the United States.
(2002) A double blind-placebo controlled study on Eisenberg, D. M., Kessler, R. C., Foster, C., et 01
American journal of Psychiatry. 158. 289-294.
melatonin efficacy to reduce anxiolytic benzodiazepine (1993) Unconventional medicine in the United States.
use in the elderly. Neuroendocrinology Letters, 23. 55-60. Prevalence, costs and pattern of use. New England Keung, WoM. (1993) Biochemical studies of a new class
journal of Medicine. 328. 246-252. of alcohol dehydrogenase inhibitors from Radix puerariae.
Carman, J. S., Post, R. M., Buswell, R., et 01 (1976)
Alcoholism. Clinicaland ExperimentalResearch.17,1254-1260.
Negative effects of melatonin on depression. American Ernst, E. (2002) "Flower remedies": a systematic review
journal of Psychiatry. 133. 1181-1186. ofthe clinicalevidence. Wiener Klinische Wochenschrift, Klepser, T. B. & Klepser, M. E. (1999)Unsafe and
Chatterjee, S. S., Nolder, M., Koch, E., et 01 (1998) 30.963-966. potentially safe herbal therapies. American journal of
Antidepressant activity of hypericum perforatum and Ernst, E. & Cassileth, B. R. (1999) How useful are Health-System Pharmacy. 56, 125-138.
hyperforin: the neglected possibility. Pharmacopsychiatry. unconventional cancer treatments? European journal of Knaudt, P. R., Cannor, K. M., Weisler, R. H., et al
31 (suppl.),7-15. Cancer. 35. 1608-1613. (2001) Alternative therapy use by psychiatric outpatients.
Chevallier, A. (1996) The Encyclopedia of Medicinal Fava, M., Giannelli, A., Rapisarda, V., et 01(1995) journal of Nervous and Mental Disease.187.692-695.
Plants. London: Dorling Kindersley. Rapidity of onset of the antidepressant effect of Krasawski, M. D., McGehee, D. S. & Moss, J. (1997)
Coffin, A. I. (1845) BotanicGuide to Health and Natural parenteralS-adenosyl-L-methionine. Psychiatry Natural inhibitors of cholinesterases: implications for
Pathology of Diseases.London:J.Caudwell.
Research. 56. 295-297. adverse drug reactions. Canadian journal of Anaesthesia,
Fisher, A. A., Purcell, P. & Le Couteur, D. G. (2000) 44. 525-534.
Cohen, M. H. & Eisenberg, D. M. (2002) Potential
physician malpractice liability associated with Toxicityof Passiflora incarnata L.journalof Clinical Lee, B. M., Lee, S. K. & Kim, H. S. (1998) Inhibition of
complementary and integrative medicinal therapies. Toxicology.
38, 63-66. oxidative DNA damage, 8-0HdG, and carbonyl
Annals of Internal Medicine. 136. 596-603. Food Standards Agency. Expert Group on Vitamins contents in smokers treated with antioxidants (vitamin
Committee on Safety of Medicines & Medicines and Minerals (2003) SateUpperLevelsfor Vitaminsand E. vitamin C, beta-carotene and red ginseng).Cancer
Minerals.http://www.foodstandards.gov.uk Letters. 132. 219-227.
Control Agency (2000) Reminder: St John's wort
(Hypericum perforatum) interactions. CurrentProblems Friedel, H. A., Goa, K. L. & Benfield, P. (1989) Lehrl, S. (2004) Clinical efficacy of kava extract WS
Pharmacovigilance.26. 6-7. S-adenosyl-L-methionine.A review of its 1490 in sleep disturbances associated with anxiety
Corrigan, J. J. Jr (1982) The effect of vitamin E on pharmacological properties and therapeutic potential in disorders. Results of a multicenter. randomized.
warfarin-induced vitamin K deficiency. Annals of the liver dysfunction and affective disorders in relation to its placebo-controlled, double-blind clinical trial. journal of
New York Academy of Sciences.393. 361-368. physiologicalrole in cellmetabolism. Drugs.38. 389-416. Affective Disorders. 78. 101-110.


Le PoncinoLafitte, M., Rapin, J. R., Duterte, D., et 01 Miller, L. G. (1989) Herbal medicinals: selected clinical Pancheri, P., Scapicchio, P. & Chiaie, R. D. (2002) A
(1985) Learning and cholinergic neurotransmission in old considerations focusing on known or potential drug- double-blind, randomized parallel-group, efficacy and
animals: the effect of Hydergine. Pharmacology. 16 herb interactions. Archives of Internal Medicine, 158, safety study of intramuscular S-adenosyl-L-methionine
(suppl.). 57-63. 2200-2211. 1.4-butanedisulphonate (SAMe) versus imipramine in
patients with major depressive disorder. International
Lewis, R., Wake, G., Court, G., et 01 (1999) Non- Mischoulon, D. & Fava, M. (2002) Role of S-adenosyl-
journal of Neuropsychopharmacology.5, 287-294.
ginsenoside nicotinic activity in ginseng species. L-methionine in the treatment of depression: a review of
Phytotherapy Research, 13, 59-64. the evidence. American journal of Clinical Nutrition, 76 Perovic, S. & Miiller, W. E. (1995) Pharmacological
(suppl.), 11585-11615. profile of hypericin extract. Effect on serotonin reuptake
Liede, K. E., Haukka, J. K., Saxen, L. M., et 01 (1998)
by postsynaptic receptors. Arzneimittelforschung, 45,
Increased tendency towards gingival bleeding caused by Morris, M., Fava, M. & Jaques, P. F. (2003) Depression 1145-1148.
joint effect of alpha-tocopherol supplementation and and folate status in the US population. Psychotherapie
acetylsalicylic acid. Annals of Medicine, 30. 542-546. Psychosomatik, 72, 80-87. Pittler, M. H. & Ernst, E. (2003) Kavaextract for
treating anxiety.CochraneLibrary.issue 4. Oxford:
Un, R. C. & U, T. K. (1998) Effects of isoflavones on Miiller, W. E., Singer, A., Wonnemann, M., et 01 (1998) Update Software.
alcohol pharmacokinetics and alcohol-drinking behavior Hyperforin represents the neurotransmitter reuptake
in rats. American journal of Clinical Nutrition. 68 (supplJ, inhibiting constituent of hypericum extract. Poyares, D. R., Guilleminault, C., Ohayon, M. M., et
15125-15155. 01 (2002) Can valerian improve the sleep of insomniacs
Pharmacopsychiatry. 31 (suppl.),16-21.
after benzodiazepine withdrawal? Progressin
Linde, K., Ramirez, G., Mulrow, C. D., et 01(1996) St Muno%-Hoyos, A., Sanche%oForte, M., Molina-
John's wort for depression -an overview and meta- Carballo, A., et 01(1998) Melatonin'srole as an
Neuropsychopharmacology and Biological Psychiatry. 26,
analysis of the randomised clinicaltrials. BMj, 313, 253-258. anticonvulsant and neuronal protector: experimental
and clinicalevidence.journalof ChildNeurology.13, Rezvani, A. H.,Overstreet, D. H., Yang,Y., et 01
Linde, K., Berner, M., Egger, M., et 01 (2005) St john's
wort for depression. Meta-analysis of randomised 501-509. (1997) Attenuation of alcohol consumption by a novel
nontoxic ibogaineanalogue (18-methoxycoronaridine) in
controlled trials. British journal of Psychiatry. 186,99-107.
Naturai Medicines Comprehensive Database (20040) alcohol-preferringrats. Pharmacology.
Biochemistry and
Lohr, J. B., Kuc%enski, R. & Niculescu, A. B. (2003) http: ffwww.naturaldatabase.com. Product search; Kava. Behavior.58, 615-619.
Oxidative mechanisms and tardive dyskinesia. CNS
Natural Medicines Comprehensive Database Rezvani, A. H., Overstreet, D. H., Yang, Y., et 01
Drugs. 17,47-62.
(2004&) http://www.naturaldatabase.com. Product (1999) Attenuation of alcohol intake by extract of
Mac/ennan, K. M., Darlington, C. L. & Smith, P. F. search: Iodine. Hypericum perforatum (St. John's Wort) in two different
(2002) The CNS effects of Ginkgo biloba extracts and strains of alcohol-preferring rats. Alcohol and Alcoholism,
Natural Medicines Comprehensive Database 34,699-705.
ginkgolide B.Progressin Neurobiology.67. 235-257.
(2004c) http://www.naturaldatabase.com. Product
MacMahon, K. M. A., Broomfield, N. M. & Espie, search: Selenium. Rezvani, A. H.,Overstreet, D. H., Perfumi, M., et of
C. A. (2005) A systematic review of the effectiveness of (2003) Plant derivatives in the treatment of alcohol
oral melatonin for adults (18 to 65 years) with delayed
Natural Med:cines Comprehensive Database dependency. Pharmacology.Biochemistry and Behavior,75,
sleep phase syndrome and adults (18 to 65 years) with (2004d) http: ffwww.naturaldatabase.com. Product 593-606.
search: Folicacid.
primary insomnia. Current Psychology Reviews, 1,103-113.
Roberts, C. G. & Ladenson, P.W. (2004)
Maisonneuve, I. M. & Glick, S. D. (2003) Anti- Natural Medicines Comprehensive Database Hypothyroidism. Lancet. 363, 793-803.
addictive actions of an iboga alkaloid congener: a novel (20040) http;ffwww.naturaldatabase.com. Product
search: Indiansnakeroot. Roder, C., Schaefer, M. & Leucht, S. (2004) Meta-
mechanism for a novel treatment. Pharmacology.
analysis of effectiveness and tolerability oftreatment of
Biochemistry and Behavior. 75, 607-618.
Naturai Medicines Comprehensive Database (20040 mild to moderate depression with St. John's Wort.
Malamud, W., Barton, W. E., Flemlng, A. M., et oJ http: ffwww.naturaldatabase.com. Product search: Iboga. Fortschritte der Neurologie Psychiatrie, 72, 330-343.
(1957) The evaluation of the effects of derivatives of
Naturai Medicines Comprehensive Database (2004g) Sano, M., Ernesto, C., Thomas, R. G., et 01 (1997) A
Rauwolfia inthe treatment of schizophrenia. American
http://www.naturaldatabase.com.Productsearch:Kudzu. controlled trial of selegeline. alpha-tocopherol or both as
journal of Psychiatry.114,193-200.
treatment for Alzheimer's disease. New England journal
Neary, J. T. & Bu, Y. (1999) Hyperforin inhibits uptake
Marangell, L. B., Martlne%, J. M., Zboyan, H. A., et of Medicine, 336, 1216-1222.
of serotonin and norepinephrine in astrocytes. Brain
01(2003) A double-blind, placebo-controlled study of Research, 816, 358-363. Schelosky, L., Raffauf, C. & Jendroska, K. (1995) Kava
the omega-3 fatty acid docosahexaenoic acid in the
and dopamine antagonism. journal of Neurology.
treatment of major depression. Americanjournalof Nemets, B., Stahl, Z. & Belmaker, R. H. (2002)
160,996-998. Neurosurgery and Psychiatry. 58, 639-640.
Addition of omega-3 fatty acid to maintenance
medication treatment for recurrent unipolar depressive Schmit%, M. & Jackel, M. (1998) Comparative study
Markstein, R. (1985) Hydergine: interaction with the
disorder. American journal of Psychiatry. 159,477-479. for assessing quality of life of patients with exogenous
neurotransmitter systems in the central nervous system.
sleep disorders (temporary sleep onset and sleep
journal of Pharmacology.16 (suppl.), 1-17. Oken, B. S., Stor%bach, D. M. & Kaye, J. A. (1998)
interruption disorders) treated with a hops-valerian
Mash, D. C., Staley, J. K., Pablo, J. P., et 01 (1995) The efficacy of ginkgo biloba on cognitive function in
preparation and a benzodiazepine drug. Wiener
Properties of ibogaine and its principal metabolite (12- Alzheimer disease. Archivesof Neurology. 55. 1409-1415.
Medizinische Wochenschri/t, 148.291-298.
hydroxyibogamine) at the MK-801 binding site of the Olde Rikkert, M. G. & Rigaud, A. S. (2001) Melatonin
Schreiter-Gasser, U. & Gasser, T. (2001) A
NMDAreceptor complex. Neuroscience Letters, 192, in elderly patients with insomnia. A systematic review.
53-56. comparison of cholinesterase inhibitors and ginkgo
Gerontology and Geriatrics, 34,491-497. extract in treatment of Alzheimer dementia. Fortschritte
Mathews, J. M., Etheridge, A. S. & Black, S.R. der Medizin Originalien, 119,135-136.
Olin, J., Schneider, L., Novit, A., et 01(2001)
(2002) Inhibition of human cytochrome P450 activities
Hyderginefor dementia.CochraneLibrary.issue4. Shamir, E., Barak, Y., Plopsky, I., et 01 (2000) Is
by kava extract and kavalactones. Drug Metabolism and Oxford; Update Software.
Disposition, 3D, 1153-1157. melatonin treatment effective for tardive dyskinesia?
Ondri%ek, R. R., Chan, P. J., Patton, W. C., et of (1999) journal of Clinical Psychiatry. 61. 556-558.
Matthews, M. K. Jr (1998) Association of Ginkgo biloba
Inhibitionof humansperm motility by specific herbs Shamir, E., Barak, Y. & Shaiman, I. (2001) Melatonin
with intracerebralhemorrhage. Neurology.50, 1933-1934.
used in alternative medicine. journal of Assisted treatment for tardive dyskinesia: double-blind. placebo-
Matthews, s. C., Camacho, A., Lawson, K., et 01 Reproduction and Genetics, 16,87-91. controlled, crossover study. Archives of General
(2003) Use of herbal medications among 200 Psychiatry. 58, 1049-1052.
Overstreet, D. H., Keung, W. M., Rezvani, A. H., et at
psychiatric outpatients: prevalence, patterns of use, and
potential dangers. General Hospital Psychiatry. 25, 24-26.
(20030) Herbal remedies for alcoholism:promises and Shebek, J. & Rindone, J. P. (2000) A pilot study
ClinicalandExperimental exploring the effect of kudzu root on the drinking habits
Medical Economics (2000) PDR(Physicians'Desk Research,27,177-185. of patients with chronic alcoholism. journal of Alternative
Reference)for Herbal Medicines (2nd edn). Montvale,NJ: and Complementary Medicine, 6, 45-48.
Overstreet, D. H., Kraiic, J. E., Morrow, A. L., et 01
Medical Economics Company.
(2003&)NPI-03IG (puerarin) reduces anxiogenic effects Sheppard, S. G. (1994) A preliminary investigation of
Meseguer, E., Taboada, R., Sanche%, V., et 01 (2002) of alcohol
withdrawal or benzodiazepine inverse or 5- ibogaine: case reports and recommendations for
Life-threatening parkinsonism induced by kava-kava. HT(2C) agonists. Pharmacology.Biochemistry and further study. journal of SubstanceAbuse and Treatment,
Movement Disorders, 17,195-196. Behavior.75, 619-625. 11,379-385.


Sher, L. (2001) Role of thyroid hormones in the effects

of selenium on mood, behavior, and cognitive function.
Medical Hypotheses,57, 480-483. CLINICAL IMPLICATIONS

Singer, C., Tractenberg, R. E., Kaye, J., et of (2003)

Alzheimer's Disease Cooperative Study. A multicenter,
placebo-controlled trial of melatonin for sleep
. Depending on the inclusion criteria chosen, between 8 and 57% of psychiatric
patients (most commonly those with depression and anxiety) use complementary
disturbance in Alzheimer's disease, Sleep, 26, 893-901.
Soares, K. V. S. & McGrath, J. J. (2001) Vitamin E for
neuroleptic-induced tardive dyskinesia. Cochrone Library.
issue 4. Oxford: Update Software.
. Cliniciansmust be prepared to discussthe use of complementary medicine with
patients who prefer a holistic approach to treatment.
Spinella, M. (2001) Psychotherapeutic herbs. In The
Psychophormacology of Herbal Medicine, p.278.
. Cliniciansneed to be aware of side-effects associated with complementary
Cambridge, MA: MIT Press.
medicines and their interactions with conventional treatments.
Staley, J. K., Ouyang, Q. & Pablo, J. (1996)
Pharmacological screen for activities of 12-
hydroxyibogamine: a primary metabolite of the indole
alkaloid ibogaine. Psychopharmacology. 127, 10-18.

Stevinson, C. & Ernst, E. (2000) Valerian for insomnia:

. The evidence base for the use of psychotropic complementary medicines is
a systematic review of randomized clinical trials. Sleep extremely limited and randomised controlled trials of promising agents are urgently
Medicine, 1,91-99. needed.
Stoll, A. L., Severus, W. E., Freeman, M. P., et 0'
(1999) Omega 3 fatty acids in bipolar disorder: a . Often, the active ingredient in herbal formulations has not been identified, which
preliminary double-blind, placebo-controlled trial.
leads to problems with standardisation of extracts and dose recommendations.
Archives of Generol Psychiatry. 56,407-412,

Su, K. P., Huang, S. Y., Chlu, C. C., et 0' (2003) . Discussionsof complementary medicine use maydemand more time than is
Omega-3 fatty acids in major depressive disorder. A availablein routine clinics.
preliminary double-blind, placebo-controlled trial.
European Neuropsychopharmacology. 13,267-271.

Szegedi, A., Kohnen, R., Dienel, A., et 0' (2005)

Acute treatment of moderate to severe depression with
hypericum extract WS5S70 (St John's wort): URSULA WERNEKE,MRCPsych,Division of Health Service Research,Institute of Psychiatry,and Division of
randomised controlled double blind non-inferiority trial Psychiatry.Homerton University Hospital,TREVORTURNER, FRCPsych,Division of Psychiatry,Homerton
against paroxetine. BMj, 330, 503-506. University Hospital,STEFANPRIEBE,FRCPsych,Unit for Socialand Community Psychiatry.Barts and The
Tabet, N., Birks, J. & Grimley Evans J. (2000) Vitamin London, NHS Trust, Queen Mary Schoolof Medicine and Dentistry. London,UK
E for Alzheimer's disease. Cochrone Librory. issue 4.
Oxford: Update Software. Correspondence: Or Ursula Werneke, Division of Psychiatry, Homerton University Hospital, East Wing,
Tang, F., Nag,S., Shiu, S. Y., et 0' (2002) The effects of Homerton Row, London E9 6SR,UK. E-mail: UrsulaWerneke@elcmht.nhs.uk
melatonin and Ginkgo biloba extract on memory loss
and choline acetyltransferase activities in the brain of (First received 2 April 2004, final revision 17March 2005, accepted3 May 2005)
rats infused intracerebroventricularly with beta-
amyloid. Life Sciences, 71, 2625-2631.
Vogeler, B. K., Pittler, M. H. & Ernst, E. (1999) The depression. International Clinical Psychopharmacology.16,
Taylor, M. J., Carney, 5., Geddes, J., et 0' (2004) Folate efficacy of ginseng. A systematic review of randomized 239-252.
for depressive disorders. Cochrane Librory. issue I. controlled trials. European journal of Clinical
Oxford: Update Software. Pharmacology.55, 567-575. Williams, J. W., Mulrow, C. D. & Chiquette, E. (2000)
A systematic review of newer pharmacotherapies for
Teufel-Mayer, R. & Gleltz, J. (1997) Effect of long term Wang, J. L., Patten, S. B. & Russell, M. L. (2001) depression in adults. Annalsof InternalMedicine,132,
administration of hypericin extracts on the affinity and Alternative medicine use by individuals with major 743-756.
density of the central serotonergic 5HTIA and 5HT 2A depression. Canadian journal of Psychiatry.46, 528-533.
receptors. Pharmacopsychiatry. 30 (suppl. 2), 113-116. Wolfman, C., Viola, H., Paladini, A., et 0' (1994)
Watanabe, H., Kakihana, M., Ohtsuka, 5., et 0' Possible anxiolytic effects of chrysin, a central
Thiele, B., Brink, I. & Ploch, M. (1994) Modulation of (1997) Randomized, double-blind,placebo-controlled benzodiazepine receptor ligand isolated from Passiflora
cytokine expression by hypericin extract. journal of studyof supplementalvitamin Eon attenuationof the coerulea. Pharmacology.Biochemistryand Behavior,47, 1-4,
Geriatric Psychiatry and Neurology. 7 (suppl. I),560-562. developmentof nitrate tolerance.Circulation,96,
2545-2550. Wolkowltz, O. M. (1993) Rational polypharmacy in
Tian, J. Z., Zhu, A. H. & Zhong, J. (2003) A follow-up
schizophrenia. Annals of Clinical Psychiatry. 5, 79-90,
study on a randomized, single-blind control of King's Werneke, U. (2003) Alternative nutrition therapies in
Brain pills in treatment of memory disorder in elderly cancer - the evidence. Clinical Nutrition Update, 8, 6-8. Zhang, X. Y., Zhou, D. F., Cao, L. Y., et 0' (2004) The
people with MC! in a Beijing community. Zhongguo effect of vitamin E treatment on tardive dyskinesia and
Zhong YaoZo Zhi, 28,987-991. Werneke, U., Earl, J., Seydel, C., et 0' (20040) Potential
blood superoxide dismutase: a double-blind placebo-
health risks of complementary alternative medicines in controlled trial. Clinical Psychopharmacology.24, 83-86.
Unutzer, J., Klap, R., Sturm, R., et 0' (2000) Mental cancer patients.Britishjournalof Concer, 90,408-413.
disorders and the use of alternative medicine: results
Ziegler, G., Ploch, M. & Miettinen-Baumann, A.
from a national survey. Americanjournalof Psychiatry. Werneke, U., Horn, O. & Taylor, D. (2004b) How
(2002) Efficacy and tolerability of valerian extract LI 156
157,1851-1857. effective isSt John'swort? The evidence revisited.journal compared with oxazepam in the treatment of non-
of Clinical Psychiatry.65, 611-617.
van Dongen, M., van Rossum, E., Kessels, A., et Q' organic insomnia - a randomized, double-blind,
(2003)Ginkgoforelderlypeoplewithdementiaand Wettstein, A. (2000) Cholinesterase inhibitors and comparative clinicalstudy. European journal of Medical
age-associated memory impairment: a randomized gingko extracts - are they comparable inthe treatment Research,25, 480-486.
clinicaltrial. ClinicalEpidemiology.
56, 367-376. of dementia? Comparison of publishedplacebo-
Zimmerman, R. A. & Thomspon, I. M. Jr (2002)
controlled efficacy studies of at least six months.
Viola, H., Wasowski, C. & Levi de Stein, M. (1995) Prevalence of complementary medicine in urologic
Phytomedicine, 6, 393-401.
Apigenin, a component of Matricaria recutita flowers, is practice. A review of recent studies with emphasis on
a central benzodiazepine receptors-ligand with Whiskey, E.,Werneke, U. & Taylor, D. (2001) A use among prostate cancer patients. Urology Clinicsof
anxiolytic effects. PlantaMedica,61,213-216. systematic review of hypericum perforatum in North America, 29, 1-9.