Tea Tree Oil and Vaginal Infections
November 1999; Volume 1: 89-91
By Adriane Fugh-Berman, MD
Melaleuca alternifolia (family myrtaceae) grows only in Australia. Captain Cook gave melaleuca its misleading name—tea tree—after observing aborigines brewing the leaves into a medicinal tea. Melaleuca is no relation to the Camellia varieties used to brew regular tea; its camphor smell would discourage anyone from consuming it for enjoyment as it’s one of the most medicinal smelling herbs. To add further confusion, several different species of plants are called tea trees.
During World War I, melaleuca oil was included in the first-aid kits of Australian soldiers for use on burns, bites, and infections. Today, the essential oil of tea tree is widely used as an antibacterial and antifungal topical medication, and is also used in several brands of toothpaste.
Melaleuca contains a complex mixture of hydrocarbons and terpenes. In Australia, commercial tea tree oil must meet the Australian standard, which stipulates that the terpinen-4-ol (thought to be the most active antimicrobial constituent) content must exceed 30% and the 1,8 cineole (a skin irritant) content must be less than 15%.1
Perhaps the most remarkable aspect of tea tree’s antimicrobial properties is its specificity for pathogens on skin and mucosal surfaces. In an in vitro study of skin flora susceptibility, S. aureus and most of the gram-negative bacteria tested (including Acinetobacter baumannii, Klebsiella pneumoniae, and Serratia marsescens) were more susceptible to tea tree oil than the coagulase-negative staphylococci and micrococci.1 Thus, tea tree oil may preferentially remove transient skin flora while preserving resident flora. The authors of this study suggest that because tea tree oil-containing preparations readily penetrate the outer layers of skin, they may be particularly useful in hand disinfection products for health care workers.
In vitro, tea tree oil appears very promising against organisms associated with bacterial vaginosis (the beneficial lactobacilli are relatively resistant).2 In an agar dilution assay, tea tree oil minimum inhibitory concentrations (MIC) for five Gardnerella vaginalis isolates were all 0.06%, and for three of four Mobiluncus isolates the MICs were 0.03% (it was 0.06% for the fourth). The MIC90 (minimum inhibitory concentration for 90% of organisms) was 0.5% for Bacteroides (12 isolates); 0.25% for Prevotella (24 isolates), Fusobacterium (10 isolates), and Peptostreptococcus anaerobius (12 isolates); and 0.12% for other gram-positive anaerobic cocci (12 isolates). By broth macrodilution, MICs for all six Bacteroides species were 0.06%; the minimal bactericidal concentrations (MBCs) for five isolates were 0.06% while the MBC for the last isolate was 0.12%. In contrast, MICs for the lactobacilli ranged from 0.12% to 2.0% and the MBC90 was 2.0%. Tea tree oil also inhibits the growth of many fungi, including 32 strains of Candida.3
Evidence in Humans
There are no controlled trials of tea tree oil in treating vaginitis or bacterial vaginosis. A case series published in 1961 found that tea tree oil was effective in trichomonal vaginitis.4 However, this required quite a bit of effort. The vagina was first washed with pHisoHex, then washed with 0.4% tea tree oil, then a tampon saturated in an emulsified 40% solution of tea tree oil and alcohol was inserted and left in place. Additionally, subjects administered daily vaginal douches containing 0.4% tea tree oil. Clinical cure was achieved after an average of six office visits and 42 vaginal douches. (One wonders whether such repeated ablutions with tap water wouldn’t have had the same effect.)
There is also one case report of a 40-year-old woman with bacterial vaginosis in whom a five-day course of tea tree oil suppositories (containing 200 mg tea tree oil in a vegetable oil base) was successful treatment.5
Although not a vaginitis trial, an open trial of melaleuca for refractory oral candidiasis in AIDS patients did show some effect.6 Thirteen patients with oral candidiasis clinically refractory to fluconazole (in vitro resistance was also demonstrated by a MIC of >20 µg/ml) were given 15 ml melaleuca-containing commercial mouthwash (Breath-Away, Melaleuca, Inc., Idaho Falls, Idaho) qid to swish and expel. The concentration of tea tree oil in this preparation is not specified.
Weekly examinations consisted of evaluation of signs and symptoms as well as quantitative yeast cultures. At four weeks, of 12 evaluable patients, two were cured, six improved, four had no response, and one was worse. Mycological response was seen in seven of 12 patients. Follow-up two to four weeks after therapy was discontinued showed no clinical relapse in the two patients who were cured. Eight of 12 patients noted mild-to-moderate oral mucosal burning on contact with the solution, primarily during the first week of therapy. This symptom improved with resolution of the candidiasis.
This is inadequate clinical evidence to recommend tea tree oil for vaginitis, but there is certainly sufficient preliminary evidence to justify performing a trial of tea tree oil for recurrent or resistant vaginal Candida infections. Obviously, the genital mucosa is more sensitive than oral mucosa, and vaginal or vulvar burning would quickly disqualify a product.
Terpenes can produce central nervous system depression and gastrointestinal and dermal irritation. Topical use of tea tree oil has been associated with dermatitis; interestingly, it appears to be the degradation products of tea tree oil that are most sensitizing.7
There are a few case reports of poisonings in humans, primarily in children. In one case, a 23-month-old boy ingested 10 ml of melaleuca oil. Initially confused and unable to walk, the child was treated with activated charcoal with sorbitol and became asymptomatic within five hours of ingestion.8 In another case, a 17-month-old boy who ingested less than 10 ml tea tree oil also developed ataxia and drowsiness.9 Animals appear to be especially susceptible to CNS effects. High doses of topically applied tea tree oil have caused depression, weakness, incoordination, and muscle tremors in pets; supportive care usually results in recovery within two to three days.10
Essential oils should not be taken orally. There are rare exceptions, and these few should be taken orally only under the supervision of a practitioner experienced in their use. Topical tea tree oil may be sensitizing; patients should keep tea tree oil products refrigerated and tightly capped to limit degradation.
Tea tree oil is effective against a number of pathogens that cause vaginal infections. While the current level of clinical evidence is not strong enough to recommend tea tree oil suppositories to patients with bacterial vaginosis or candidiasis, this is a promising area for further research.
1. Hammer KA, et al. Susceptibility of transient and commensal skin flora to the essential oil of Melaleuca alternifolia (tea tree oil). Am J Infect Control 1996;24:186-189.
2. Hammer KA, et al. In vitro susceptibilities of lactobacilli and organisms associated with bacterial vaginosis to Melaleuca alternifolia (tea tree) oil. Antimicrob Agents Chemother 1999;43:196.
3. Nenoff P, et al. Antifungal activity of the essential oil of Melaleuca alternifolia (tea tree oil) against pathogenic fungi in vitro. Skin Pharmacol 1996;9:388-394.
4. Pena EF. Melaleuca alternifolia oil. Ob-Gyn 1962;19:793-794.
5. Blackwell AL. Tea tree oil and anaerobic (bacterial) vaginosis. Lancet 1991;337:300.
6. Jandourek A, et al. Efficacy of melaleuca oral solution for the treatment of fluconazole refractory oral candidiasis in AIDS patients. AIDS 1998;12:1033-1037.
7. Hausen BM, et al. Degradation products of monoterpenes are the sensitizing agents in tea tree oil. Am J Contact Dermat 1999;10:68-77.
8. Jacobs MR, Hornfeldt CS. Melaleuca oil poisoning. J Toxicol Clin Toxicol 1994;32:461-464.
9. Del Beccaro MA. Melaleuca oil poisoning in a 17-month-old. Vet Hum Toxicol 1995;37:557-558.
10. Villar D, et al. Toxicity of melaleuca oil and related essential oils applied topically on dogs and cats. Vet Hum Toxicol 1994;36:139-142.