Background

Mental disorders plague millions of people around the world. Depression and anxiety are two of the most common mental disorders, affecting nearly 55 million people in the United States alone [1]. The complexities of the central nervous system make diagnoses, treatment, and amelioration of these debilitating illnesses exceptionally difficult. Advancement in these areas would be invaluable contributions in the effort to reduce the global impact of anxiety-based conditions. The universality of herbal remedies in many cultures makes them an appropriate treatment to explore.

According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR), anxiety is characterized by a feeling of persistent worry that hinders an individual's ability to relax [2]. This can range from the transient anxiety levels a person feels before surgery or a menstrual cycle to the pervasive feeling of nervousness that characterizes an anxiety disorder (e.g. generalized anxiety disorder, obsessive-compulsive disorder, panic disorder and social phobia). The impact of the anxiety is not limited to consistent stress, which is associated with higher risk of cardiovascular and cerebrovascular disease [3]. Anxiety also has debilitating physical manifestations as headaches, uncontrolled trembling and sweating, muscle tension and aches, among others.

To date, the biological explanations for many types of anxiety disorders remain inadequate. Postulations have implicated a dysregulation of specific neurotransmitters such as serotonin, dopamine and gamma-aminobutyric acid (GABA) as potential causes for both depression and anxiety disorders [4–6]. These hypotheses are based on the results of pharmacological treatments, but there are no definitive clinical trials that demonstrate the dysregulation of these neurotransmitters as causative factors of anxiety, potentially explaining why the treatment of anxiety with antidepressants is often ineffective. Thus far, cognitive behavioural therapy (CBT) has proven to be the most effective, long-term treatment for anxiety-related disorders [7].

With the lifetime prevalence of anxiety disorders reaching 16.6% worldwide [8], great strides have been made with ongoing research into its causes and treatments. In addition to antidepressants, serotonin-specific reuptake inhibitors (SSRIs) and benzodiazepines have also been prescribed to patients suffering from GAD [9, 10]. However, while often effective, both of these classes of drugs come with many unwanted side effects such as suicidal ideation, decreased alertness, sexual dysfunction and dependency [11–16]. Additionally, the costs of these medications pose problems to patients who must take them on a daily, long-term basis

As a result, there has been increased interest in the use of complementary and alternative medicines (CAM) as a natural method for treating numerous types of anxiety. Herbs such as passionflower, kava, St. John's wort and valerian root, as well as the amino acid lysine and the cation magnesium, have been used for centuries in folk and traditional medicine to calm the mind and positively enhance mood. However, the efficacy and safety of utilizing CAMs to treat anxiety, both as a symptom and as a disorder, has only just begun to be rigorously tested in clinical trials within the last 10 to 15 years [17–19].

A number of reviews of the clinical effectiveness of herbal and nutrient treatments for depression, anxiety disorders, and sleep disturbance have been published over the past decade [19–25]. These have reviewed data associated with a number of treatments, including St. John's Wort, S-adenosyl-methionine (SAM-e), B vitamins, inositol, choline, kava, omega-3 fatty acids/fish extracts, valerian, lavender, melatonin, passionflower, skullcap, hops, lemon balm, black cohosh, ginkgo biloba, extracts of Magnolia and Phellondendron bark, gamma-aminobutyric acid (GABA), theanine, tryptophan and 5-hydroxytryptophan (5-HTP). However, none of these studies has been conducted in a systematic way.

The objective of this paper is to systematically review and summarize the available literature on herbal remedies and dietary supplements for treating anxiety and related symptoms in order to aid mental health practitioners in advising their patients and provide insight for future research in this field.

Methods

Search strategy

MEDLINE/PubMed and EBSCO databases were searched without regard for date of publication, using the search terms "alternative therapies," "herbal supplement" and individual herb and supplement names from popular sources, each crossed with the term "anxiety." In addition, key publications were hand-searched for references. [See Additional file 1 for a Quality of Reporting of Meta-analyses (QUOROM) statement checklist.]

Selection criteria

The search was restricted to herbs and supplements that acted as anxiolytic agents and whose effects were measured either through quantitative rating scales or self-reports. Studies also had to be published in English, conducted with human subjects, have a sample size greater than 10, use a whole extract of the plant (if applicable) and detail data clearly. Case studies, review articles, meta-analyses, safety trials and studies that attempted to link vitamin and mineral deficiencies to the presence or absence of anxiety symptoms were excluded, as were trials in animals. Studies of anxiety parameters in healthy volunteers were also examined to provide supporting evidence.

Data abstraction and synthesis of results

Study results were abstracted into data tables (Tables 1, 2, 3. 4, 5, 6). Because of the heterogeneous nature of the patients, preparations and outcome measures, data pooling was not possible. Therefore, the data was summarized qualitatively. The most common outcome measures encountered in these trials included: Hamilton Anxiety Scale (HAMA), State Trait Anxiety Inventory (STAI), Erlanger Anxiety, Tension and Aggression Scale (EAAS), Bf-S self-rating scale of well-being, Anxiety Sensitivity Index (ASI), and Clinical Global Impressions (CGI) scale. Some studies used measurements of anxiety biomarkers such as adrenocorticotropic hormone, cortisol, adrenaline, noradrenaline and chromogranin-A levels; skin conductance; heart rate; and blood pressure. A significant positive change in at least one of the primary outcome measures was required to categorize the trial as positive.

Table 1 Participant characteristics
Full size table
Table 2 Trials testing passionflower
Full size table
Table 3 Trials testing kava
Full size table
Table 4 Trials testing St. John's wort
Full size table
Table 5 Trials testing lysine
Full size table
Table 6 Trials testing magnesium
Full size table

Results and discussion

Flow of included studies

Electronic searches found 106 papers that were potentially relevant to the present systematic review. Of these, 24 met the inclusion/exclusion criteria (see Figure 1 for a flow diagram). Of the 82 that did not meet the criteria, 21 were excluded from the main review because they were not original research (e.g. reviews or meta-analyses) or were case studies, 14 did not investigate the supplement as a treatment (e.g. safety analysis, pharmacological evaluations, study of nutritional deficiencies), 32 did not use human subjects, and 15 were published in a language other than English. Some of the excluded papers listed as reviews are cited in the background and discussion sections of this manuscript. Papers that were mainly discussions of philosophical and ethical issues were not reviewed at all.

Figure 1
figure 1

Flow diagram of included studies.

Full size image

Study characteristics

A total of 24 studies were found that met the aforementioned requirements. These studies examined the effectiveness of five monotherapies (passionflower, lysine, magnesium, kava and St John's wort) and eight combination treatments (a herbal combination, multivitamin, L-lysine + L-arginine, magnesium + vitamin B6, herbal combination + magnesium, calcium + kava, St John's wort + kava, St John's wort + valerian). Of these studies, 13 were randomized controlled trials in outpatients with a DSM-IV-diagnosed disorder, and three were randomized controlled trials in patients with other types of anxiety (perimenopausal, menstrual, and pre-surgery). Five trials were done in healthy volunteers, three of which recruited healthy volunteers with high-normal anxiety levels. In addition, there were three uncontrolled observational studies.

Overall, 2619 participants between the ages of 18 and 82 took part in these studies. Twenty-eight percent were male, 63% were female and 9% did not have their gender reported (Table 1). Ethnicity/race, although an important demographic factor, was not reported in 72% of patients. As a result, it is difficult to draw any overarching conclusions from the results because these factors can significantly affect the potential for herbs to treat anxiety illnesses.

Some cultures have a greater preference for natural medicine than modern medicine, and therefore will likely exhibit positive results towards it. Because culture, gender, and age are potential confounding variables, efforts should be made to control for them in future studies.

This review presents the available evidence for passionflower, lysine, magnesium, kava and St John's wort, either alone or in combination. Methodological details and results of these trials are summarized in Tables 2, 3, 4, 5, 6. These tables are divided according to the treatment studied and include the reference, study design, sample population, intervention, control, outcomes, direction of evidence, and reported adverse events.

Herbal Medicines

Passionflower

Passionflower or Passiflora incarnata Linn. has a long history of use as an anxiolytic agent in folklore and has been used by people all over the world to treat anxiety [26]. More importantly, several studies involving the biochemical makeup of passionflower have been conducted [27–29]. Between the 1970 s and 1990 s, passionflower was listed as an official plant drug by the pharmacopoeias of America, Britain, Germany, France, Switzerland, Egypt and India; its wide use has made it an acceptable treatment for restlessness and nervousness [30].

The anxiolytic effects of passionflower are well documented in mice [30, 31]. However, one of the problems with herbal supplements is that plant material contains thousands of phytochemicals, making it challenging to pinpoint the specific biochemicals responsible for the anxiolytic properties. In other words, although herbal remedies often produce positive results, identifying the active ingredients can be difficult. Therefore, users of herbal remedies may be consuming ineffective or possibly toxic substances in addition to the active, anxiolytic ingredients. To date, three human trials have documented the efficacy of passionflower as a treatment for anxiety-related disorders [32–34].

One double-blind, placebo-controlled study analyzed the difference in efficacy between oxazepam, a prescription benzodiazepine used to treat chronic anxiety symptoms, and passionflower in patients (n = 36) who met the criteria for GAD [32]. The results showed no difference between the two anxiolytics with regard to the treatment of GAD, suggesting that passionflower is as effective as benzodiazepines in eliminating anxiety symptoms. Subjects from the passionflower group also reported lower job impairment performance than those in the benzodiazepine group; however, subjects in the benzodiazepine group reported a faster onset of symptom relief.

This anxiolytic effect was also seen in two other subsets of patients: those undergoing surgery (n = 60) who were treated with passionflower monotherapy [33], and those diagnosed with adjustment disorder with anxious mood (n = 182) who were treated with passionflower in combination with crataegus oxyacantha, ballota foetida, valeriana officinalis, cola nitida and paullinia cupana [34].

Mild adverse events were reported in only one study, including dizziness, drowsiness and confusion [32]. This preliminary evidence suggests that passionflower may have a role in the treatment of anxiety and warrants future research.

Kava

Kava is a drink that is prepared from the plant Piper methysticum. It has been consumed in many cultures because it is known to relieve anxiety, restlessness and insomnia for centuries [35, 36]. Several studies in animals have also demonstrated the kava plant's abilities as an anxiolytic agent [37, 38]. The attractiveness of kava is that it is anxiolytic but not sedative or mentally impairing, which are typical side effects caused by benzodiazepines [32]. The biochemical mechanism of kava's anxiolytic activity has been postulated to occur through enhanced ligand binding to GABA type A receptors, blockage of violated-gated sodium channels and calcium ion channels, norepinephrine and dopamine reuptake inhibition, and reversible inhibition of monoamine oxidase (MAO) B [see [39] for a review]. To note, the binding of kava extracts to several neurotransmitters such as GABAA1, dopamine D2 and the opiates (μ and δ), were demonstrated in vitro and in the rat brain [40, 41].

The first randomized, placebo-controlled, double-blind study of kava for the treatment of patients who were diagnosed with anxiety disorder was conducted in 1997 [42]. The subjects (n = 101) were given either an extract of kava or a placebo for 25 weeks. The participants who were given the kava extract showed improvement in their primary and secondary anxiety symptoms based on the HAMA -- a method of patient self-reporting -- and a CGI, which was determined by physicians. Primary anxiety is described as the inability to regulate stress and anxiety since early childhood. Secondary anxiety, which develops later in life, can be caused by clinical disorders. The researchers concluded that when kava is used an anxiolytic alternative to benzodiazepines or tricyclic antidepressants, individuals typically suffer from less side effects.

These results were later supported by five other RCTs [43–47] and one uncontrolled, observational study [48]. These studies showed kava's therapeutic potential both as a monotherapy for patients with anxiety disorder [48], GAD [43, 44, 49], elevated generalized anxiety [47] and those being tapered off of benzodiazepines [45], as well as in combination with calcium for perimenopausal women [46].

However, four RCTs showed that kava alone or in combination with St John's wort is no more effective than placebo in reducing symptoms of anxiety [50–53]. Two of these studies showed no significant difference between kava treatment and placebo [51, 53], while one found that placebo was actually better at reducing anxiety symptoms in patients with higher baseline anxiety scores [52]. According to the researchers, this could have been partly due to the study population. In this trial, patients were actively looking for alternative treatments for their GAD and were, therefore, highly motivated for kava treatment to produce an effect. This in turn could have influenced their response to treatment and led to an increased placebo effect. It is important to note that the sample size of this study was very small.

The last negative trial [50] was classified on the basis that it failed to meet its primary endpoint -- a significant reduction in anxiety based on the Zung Anxiety Scale from memory. However, an exploratory analysis of variance across the differences between treatment end and baseline, with the treatment center as a second factor, showed superiority of kava over placebo. According to the researchers, this variance between centers did not endanger the validity of their findings; however, it did reinforce the importance of standardizing ratings across participating centers in multi-center studies.

All of these trials also revealed that taking doses less than 400 mg/day does not cause serious side effects. This is important to note, especially since the U.S. Food and Drug Administration (FDA) published a consumer advisory warning in 2002 about the potential for severe liver damage from kava-containing supplements [54]. This potential, yet rare, risk of hepatitis, cirrhosis and liver failure led nations such as Canada and the United Kingdom to ban kava supplements. However, Teschke et al. reported in 2008, that owing to the fact that kava was considered to be well tolerated until 1998 when the first reported cause of hepatotoxicity occurred, these rare, but serious side effects may have occurred due to poor quality kava, as well as other risk factors such as overdose, prolonged therapy and co-medication [55].

Of the 435 clinical trial participants taking kava supplements in our review, some at high doses, no liver issues were reported. Therefore, the current review supports the conclusion that liver toxicity is indeed a rare side effect.

St John's wort

Hypericum perforatum, or St John's wort (SJW), is derived from the flowering tops of a perennial shrub. It has been used in traditional medicine for centuries to treat a wide range of disorders and is licensed in Germany to treat anxiety, depression and sleep disorders [56]. There are numerous hypotheses for its anxiolytic effects based on the binding affinity of at least 10 different extracts, including naphthodianthrones like hypericins, flavonoids, xanthones, and bioflavonoids, for adenosine, GABAA, GABAB and glutamine receptors, as well as the inhibition of monoamine oxidase-A and -B activity and synaptosomal uptake of serotonin, dopamine and noradrenaline (norepinephrine) [57]. Of these active ingredients, hypericin has been studied the most, and the amount present is generally used to standardize extracts.

SJW is probably most recognized for its use in depression. A meta-analysis published in 1996, showed that SJW was more effective than placebo in treating mild to moderate clinical depression [56]. Based on the author's recommendations, researchers began comparing the efficacy and safety profile of SJW against other routinely prescribed antidepressants. One trial conducted in Germany concluded that SJW was as effective as imipramine in treating mild to moderate depression (n = 324) [58].

Depression has been linked to anxiety, with many symptoms, panic attacks for example, overlapping between the two disorders. Little is known about the specific reasons for the link in the conditions; however, there may be as high as an 85% overlap with the diagnoses and many conventional treatment options are prescribed for both disorders. There has been little study of the effectiveness of SJW in treating anxiety disorders specifically, with only four RCTs [51, 59–61] and two uncontrolled observational studies [62, 63].

These published studies presented contradictory results. A small 12-week observational study (n = 13) of patients with OCD showed that SJW caused significant improvements, with results comparable to those seen in clinical trials with SSRIs [63]. However, a larger 12-week RCT (n = 60) showed no significant difference between patients treated with SJW (at doses higher than the observational study) or those treated with placebo [59]. Based on previous studies, OCD has one of the lowest placebo response rates of all of the anxiety disorders [64]. For this reason, these negative results were probably due to lack of response to SJW treatment rather than the high placebo response rates noted in the negative kava trials [59].

A second set of RCTs investigated the use of SJW combination treatments for depression with co-morbid anxiety. A combination of SJW and valerian was found to significantly reduce anxiety disorder symptoms; however, greater reductions were seen with higher doses of valerian (SJW doses remained constant between treatment groups), suggesting that valerian has more of an effect on symptoms [62]. A combination of SJW and kava was shown to have no significant effects on anxiety [51].

Finally, a RCT of 149 patients with depression with co-morbid anxiety, OCD and somatization disorder demonstrated that six weeks of treatment with SJW significantly reduced anxiety [61]. However, a RCT of 40 patients diagnosed with generalized social anxiety disorder found that SJW was no more effective than placebo in reducing anxiety symptoms [60]. In the discussion of this study, the researchers stated that a negative trial was conducted but speculated that minimum severity levels may be necessary for SJW to be effective in this patient population [60].

More research needs to be done using SJW in all the indications presented in this review in order to determine its effectiveness. However, the results point to a potential anxiolytic agent with a side effect profile similar to placebo. All of the side effects reported in the reviewed trials were mild to moderate and were most often cases of gastrointestinal upset, dizziness, sleep disturbances, and headaches.

Nutritional Supplements

Lysine

It has long been postulated that the dysregulation of neurotransmitters may be a cause for anxiety. These neurotransmitters include GABA, serotonin, dopamine and norepinephrine [4–6]. Amino acids such as L-tyrosine and L-tryptophan are known precursors for specific neurotransmitters. Recent studies in animals have identified two other amino acids, L-lysine and L-arginine [65, 66], which may influence neurotransmitters involved in stress and anxiety. L-lysine has been shown to act as a partial serotonin receptor 4 (5-HT4) antagonist, decreasing the brain-gut response to stress as well as decreasing blood cortisol levels [65]. Based on the results from animal studies, two placebo-controlled studies were conducted to analyze the effects of L-lysine-containing supplements in humans [67, 68].

The first of these clinical trials was conducted in healthy male volunteers who were suffering from high-trait anxiety based on a STAI questionnaire [67]. Results from this study showed that L-lysine and L-arginine combination supplements improved participants' ability to handle induced stress through an increase in cortisol, while placebo had no reported improvement of anxiety symptoms. In the discussion, the researchers attributed the increase in cortisol to a previous stress hormone regulation deficiency. A previous report indicated that during moments of induced stress, an increase in cortisol levels, which is the typical reaction in healthy persons, does not increase in people with high-trait anxiety [69]. This dysregulation of cortisol may lead to augmented feelings of anxiousness when stress is induced.

The second RCT recruited 108 healthy Japanese individuals [68]. After one week of treatment with an oral L-lysine and L-arginine supplement, basal levels of salivary cortisol decreased in male subjects (n = 54) but not in females, presumably because these participants were not selected based on high-trait anxiety. Supplementation also resulted in significant reductions in state anxiety (a temporary condition characterized by apprehension, tension and fear about a specific situation or activity) and trait anxiety (a pre-set level of anxiety or a tendency to be anxious) in both males and females.

For the two available RCTs, it seems that the L-lysine + L-arginine combination effectively reduces anxiety scores with no reported side effects. Amino acid supplements may also help in balancing cortisol levels triggered by stress in both healthy individuals and those with high trait anxiety. However, more research needs to be conducted on both lysine combinations and monotherapy to confirm these results.

Magnesium

Magnesium is a positively charged ion, a cation, that is involved in many important molecular functions in the body and has been linked to anxiety-related disorders [70–74]. To date, three human trials have been conducted that test the anti-anxiety effects of increased magnesium intake in combination therapies [75–77], and all showed a positive direction of evidence.

In the first study, 28-day treatments with a multivitamin that contained large amounts of magnesium, zinc and calcium dramatically decreased psychological distress (according to the GHQ-28) compared to placebo, which worsened symptoms [75]. Results from the HADS also showed a decrease in anxiety for the treatment group. The effects became more pronounced as the multivitamin treatment progressed but could not be linked solely to magnesium supplementation.

A second study published in 2000 looked at the effects of magnesium and vitamin B6 supplementation on premenstruation-related anxiety [76]. The women were given 1) magnesium, 2) B6, 3) magnesium + B6, and 4) a placebo over four menstrual cycles, respectively. The average magnesium intake for this study was approximately 300 mg daily. The women were asked to keep a log of their symptoms and categorize them into six groups: anxiety, craving, depression, hydration, other, and total. The results showed that the combination of magnesium and B6 created a synergistic affect that provided women with the greatest relief from premenstrual anxiety. However, magnesium monotherapy was shown to provide results similar to placebo.

The third clinical study was conducted in 2004 and investigated the effects of three compounds in combination, including magnesium, versus placebo in patients diagnosed with GAD (n = 264) [77]. The researchers found that both the magnesium-containing supplement and the placebo drastically decreased anxiety systems based on HAMA, a personal assessment, and a physician's evaluation, hinting at a potential placebo effect for this treatment. Also, due to the fact that one of the herbal extracts contained in the preparation is closely related to the opium poppy, these effects may not have been due to the action of the magnesium.

Although the exact mechanism has yet to be determined, it appears magnesium supplementation is effective at treating anxiety and anxiety-related disorders when used in combination with other vitamins, minerals and herbal extracts. However, more research of magnesium monotherapy and its pharmacology is needed to determine whether magnesium itself possesses anxiolytic characteristics. Overall, available literature shows that magnesium-containing supplements are generally well-tolerated with very few reported side effects.

Conclusions

Anxiety disorders are one of many common psychological ailments. Natural remedies have been used for centuries in many cultures to alleviate anxiety and its symptoms with surprising effectiveness. In Western cultures, however, research that proves the usefulness of medicinal herbs and natural substances has only begun to gain momentum over the past few decades. In addition, the absence of proper guidelines governing the production and use of vitamins, minerals, amino acids and herbs for medicinal purposes is also causing the clinical prescription of these natural treatments to lag behind in the United States.

Of the RCTs reviewed in this report, 71% (15 out of 21) showed a positive direction of evidence, and any reported side effects were mild to moderate. Based on this data, it appears that nutritional and herbal supplements are effective methods for treating anxiety and anxiety-related conditions without the risk of serious side effects. However, the effectiveness of each of the reviewed combinations and monotherapies has not been substantiated to the same degree.

Passionflower has been studied in three different RCTs, twice as a monotherapy and once as part of an herbal combination. All three of these studies showed a positive benefit for treatment with passionflower, providing good evidence of its effectiveness as an anxiolytic agent. However, since each of these studies was conducted in a different patient type, more research is needed to prove its efficacy in each indication.

Kava is the most researched supplement in this review with 11 different studies (10 RCTs and one observational). Of the RCTs of kava monotherapy, 63% (5/8) showed treatment significantly reduced anxiety symptoms in a variety of patient types. This provides good evidence for the use of kava in patients with GAD, non-psychotic anxiety and other anxiety-related disorders.

The evidence for St John's wort was mixed, with 50% (3/6) of the studies having positive results. However, the fact that only 1 out of the 4 RCTs had a positive direction of evidence and that the active treatment in this trial was a combination of SJW and valerian suggests that SJW monotherapy should not be recommended to patients suffering from anxiety disorders or other anxiety-related conditions.

For all three of the reviewed herbal supplements, more research needs to be done to establish the most effective dosage and to determine whether this varies between different types of anxiety or anxiety-related disorders. Furthermore, as 3 of the 4 herbal combinations showed positive results, future research should focus on determining whether herbal combinations are similarly or more effective than monotherapy as well as refining the type of herbs and dosages contained in combination supplements.

Combination nutritional supplements containing lysine or magnesium also appear to hold promise as treatments for anxiety symptoms and disorders. Both RCTs of L-lysine and L-arginine combinations demonstrated positive results, providing good but limited evidence of its usefulness as a treatment for anxiety.

The evidence for magnesium is mixed. Even though all three RCTs of magnesium-containing supplements had positive results, magnesium monotherapy was shown to be no different than placebo [76], raising the question of whether magnesium provides any anxiolytic benefits in combination or whether the results were based on the actions of the other nutrients/herbal extracts. However, this study was conducted in women with premenstrual anxiety rather than an anxiety disorder. Future research should focus on elucidating magnesium's mode of action in order to determine if it has anxiolytic properties and provides any synergistic effects when combined with other natural anxiolytic agents.

Herbal medicines hold an important place in the history of medicine, as most of our current remedies, and the majority of those to be discovered in the future, will contain phytochemicals derived from plants. While locating the active ingredients in herbal substances is pivotal to being able to produce effective supplements, understanding the quantity needed and potency of different ways of extracting and preparing the phytochemicals is vital to creating a standard measure of their effectiveness. In addition, the dangers of overconsumption and interactions with prescription medications and over-the-counter medications need to be further analyzed. This understanding of the standards for effective preparation further minimizes the chance of side effects from herbal medicines and helps to create an undisputable body of evidence for their effectiveness.