Table 3

Trials testing kava


Study Design

Sample Population



Length of Treatment


Direction of Evidence

Reported Adverse Events

Volz (1997) [42]

Randomized; Double-blind; Parallel Group

101 outpatients with anxiety of non-psychotic origin1

Kava-kava extract WS 1490 (90- 110 mg dry extract = 70 mg kl per capsule)


24 weeks

Significant reduction in anxiety (HAMA, CGI, SCL-90-R, AMS) in favour of kava-kava treatment.


Excellent tolerability, similar to placebo; no clinically relevant changes in laboratory results.

Stomach upset.

Scherer (1998)* [48]

Open-label; Uncontrolled Observational study

52 outpatients with nonpsychotic anxiety

Kava preparation (no dose reported in abstract)


Not reported in abstract

42 patients (80.8%) rated kava treatment as "very good" or "good".



Malsch (2001) [45]

Randomized; Double-blind; Parallel group

40 adult outpatients with non-psychotic nervous anxiety, tension and restlessness, impairing work performance, normal social activities and relationships2

Pre-treatment with benodiazepines (tapered off over two weeks) followed by capsules of 50 mg/day of dry extract standardized to 35 mg kava lactone for three weeks

Pre-treatment with benodiazepines (tapered off over two weeks) followed by placebo for three weeks

5 weeks

Significant reduction in anxiety (HAMA, Bf-S, EAAS, CGI) in kava-treated group.


No serious adverse events

Watkins (2001) [44]

Randomized; Double-blind; Parallel Group

13 patients with GAD

Kava 280 mg/day

(standardized to 30% kavalactones)


4 weeks

Significant improvement in baroreflex control of heart rate in kava-treated group;

respiratory sinus arrhythmia did not respond to kava treatment.


Not reported

Connor (2002) [52]

Randomized; Double-blind; Parallel Group

38 adults with DSM-IV GAD3

Kava (standardized to 70 mg kavalactones [kl]).

Treatment initiated at 149 mg kl/day and increased to 280 mg kl/day for the next 3 weeks.


4 weeks

No significant difference to placebo4


Well tolerated.

No evidence of withdrawal or sexual side effects.

Boerner (2003) [43]

Randomized; Double-blind; Parallel Group

129 outpatients diagnosed with GAD (GAD; ICD-10: F41.1)

400 mg/day Kava extract LI 150 (standardized to 30% kavapyrones, extraction solvent 96% ethanol in water, drug-extract ratio 13-20:1)

(1) 10 mg/day Buspirone or (2) 100 mg/day Opipramol

8 weeks

Kava was shown to be as effective as reference treatments; 75% of patients responded (50% reduction of HAMA score).


1 treatment-related adverse event.

No systematic difference between treatments.

No liver toxicity reported5.

Cagnacci (2003) [46]



Parallel Groups (3)

80 peri-menopausal women

Calcium (1 g/day) plus:

(1) Kava-Kava,100 mg/day (55% of kavaina; Natural Bradel, Milano, Italy)

(2) Kava-Kava, 200 mg/day

Calcium (1 g/day)

3 months

Significant reduction in STAI scores in favour of combination treatment.




Gastric pain.

No liver toxicity.

Gastpar (2003) [50]

Randomized; Double-blind;

Parallel Group

141 adult outpatients diagnosed with neurotic anxiety6

150 mg/day kava special extract WS 1490 (standardized to 35 mg kl)


4 weeks

Pronounced decrease in ASI score for the kava group; however not statistically significant overall; however an exploratory analysis of variance across the differences between treatment end and baseline, with center as a second factor, showed superiority of kava over placebo.


Increased tiredness.

No liver toxicity

Jacobs (2005) [53]

Randomized; Double-blind; Parallel Group (3)

391 healthy volunteers with anxiety7 and insomnia

(1) 100 mg kl/day kava (30% total kavalactones in extract) with valerian placebo

(2) 6.4 mg/day valerian (1% valerenic acid in extract) with kava placebo

Double placebo

28 days

Greater reductions in placebo group, but not statistically significant (STAI-State substest).


Similar frequency between treatments and placebo.

No reports of liver toxicity

Sarris (2009) [47]


Double-blind; Crossover

41 adult participants with 1 month or more of elevated generalized anxiety

Kava tablets (250 mg/day kavalactones)


3 weeks

Highly significant reduction in anxiety (HAMA, BAI, MADRS) in kava-treated group.


No serious adverse events.

Mild dizziness, nausea.

No liver toxicity.

Sarris (2009) [51]

Randomized; Double-blind; Crossover

28 adults with MDD and co-occurring anxiety

Hypericum perforatum8

(1 × 1.8 g tablet, three times/day); Kava rhizome aqueous extract9

(1 × 2.66 g tablet, 3 times/day)


4 weeks

Combination treatment had no significant effects on anxiety (BDI-II).


No serious adverse events.

Mild gastrointestinal upset.

No liver toxicity

HAMA: Hamilton Anxiety Scale; CGI: Clinical Global Impressions; SCL-90-R-ANX: Self-Report Symptom Inventory-90 Items revised, subscore somatic anxiety; AMS: Adjective Mood Scale; kl: kavalactones (kl); Bf-s: Befindlichkeitsskala [subjective well-being score]; EAAS: Erlanger Anxiety, Tension and Aggression Scale; DSM-IV: Diagnostic and Statistical Manual of Mental Disorders, fourth edition; GAD: generalized anxiety disorder; BAI: Beck Anxiety Inventory; BDI-II: Beck Depression Inventory-II; NADRS: Montgomery-Asberg Depression Rating Scale; MDD: major depressive disorder; SARA: Self-Assessment of Resilience and Anxiety; HADS: Hospital Anxiety and Depression Scale.

* No full text available.

1. DSM-III-R criteria: agoraphobia, specific phobia, generalized anxiety disorder and adjustment disorder with anxiety.

2. Diagnosis of agoraphobia (300.22), simple (300.29) or social phobia (300.23), generalized anxiety disorders (300.02) or adaptation disturbances (309.24) according to DSM-III-R.

3. According to the Results section, "Thirty-eight subjects were randomized, including 31 female (82%) and 32 Caucasian participants (97%)...Three subjects withdrew their consent following the baseline visit...and did not return for further assessment, leaving 35 subjects in the evaluable sample;" however, the Abstract states: "Thirty-seven adults with DSM-IV GAD were randomly assigned to...treatment."

4. Post-hoc analyses: kava was superior in low anxiety (SARA) and placebo was superior in high anxiety (HADS; SARA)

5. Slight increases in transaminase levels to above the upper limit of normal were reported in all three groups.

6. DSM-III-R diagnoses 300.02, 300.22, 300.23, 300.29, or 309.24.

7. Scores of at least 0.5 standard deviations above the mean on STAI-State.

8. Standardized to 990 μg of hypericin, and 1500 μg of flavone glycosides.

9. Standardized to 50 mg of kavalactones.

Lakhan and Vieira Nutrition Journal 2010 9:42   doi:10.1186/1475-2891-9-42

Open Data