Skip navigation

Complementary AIDs therapies

Clinical bottom line: Acupuncture, herbal remedies, massage, stress management and vitamins/other supplements do not appear to have any beneficial effect on immune function or behavioural factors in patients with HIV/AIDs.

Alternative therapies are widely used by HIV/AIDs patients, even in the advanced stages of illness, to help boost immunity, to prevent weight loss, nausea and diarrhoea and to relieve stress or depression. Herbal therapies, vitamins, massage, acupuncture and imagery are the most commonly used treatments.

Systematic review:

Özsoy M, Ernst E. How effective are complementary AIDs therapies for HIV and AIDS? A systematic review. International Journal of STD and AIDS 1999;10:629-635.

Date review completed: 1997

Number of trials included: 14

Number of patients: (see below)

Control groups: placebo, sham acupuncture therapy, or no treatment.

Main outcomes: viral count; CD4+ count; effect on mood (e.g. anxiety); pain.

Inclusion criteria were: randomised trial; assessment of alternative therapies for HIV infection or AIDs; treatment groups which included placebo, sham therapy, standard treatment or no treatment. Medline, Embase, CISCOM, the Cochrane Library and the authors' database were searched (all years to December 1997) for reports of studies which assessed the effectiveness of alternative therapies for HIV infection or AIDs. Reference lists of retrieved reports were checked for additional citations and experts were contacted. No language restrictions were made. Data were extracted in a standardised, predefined manner. A descriptive analysis was conducted.


Fourteen randomised trials of variable modality, sample size and methodological rigour were included. Various alternative therapies were assessed. The reviewers' commented on the methodological flaws of the studies, particularly their small sample size. The presence or absence of blinding was not mentioned; unblinded studies are open to observer bias. Additionally, some studies where of short duration or follow-up. Numbers of patients improved with the different treatments were not provided.

Herbal remedies:

Two randomised trials were assessed. One showed a beneficial effect of boxwood extract 990 mg/day compared with placebo on CD4 count, but was aborted due to ethical reasons. The number of patients was not provided. The other trial showed no difference between a mixture of 31 Chinese herbal treatments (650 mg/day for 12 weeks) and placebo on CD4 count and mood, but was reported to have reduced neurological, gastrointestinal and respiratory symptoms and disruption to sleep. The reviewers stated that this study was methodologically flawed: small sample size (30 patients), short duration etc.

Vitamins and other supplements:

Five studies assessed either vitamins (A, B6, C, E), beta-carotene, selenium or other nutritional supplements in 217 HIV-infected patients. Three studies used a placebo control. No significant difference in CD4 count or viral load was shown between treatment and control with any of these vitamins/supplements.

Stress management:

Five studies (178 patients) assessed stress management compared with no treatment on behaviour and/or immune functioning in mainly HIV-positive men. One study (10 patients) showed significantly lower stress scores and T-cell count with treatment after one month. Two studies (104 patients) showed no difference in immune function, one (45 patients) showed a decline in HIV-related intrusive thinking, and one (19 patients) showed significantly better scores for anger in the group which received relaxation training than in the groups which received psychotherapy or no treatment.


One randomised trial assessed a standardised acupuncture regimen versus sham acupuncture (239 patients), and amitriptyline versus placebo (136 patients) over 14 weeks for the relief of pain due to HIV-related peripheral neuropathy. After 14 weeks there was no significant difference between acupuncture and sham acupuncture. Amitriptyline was more effective than placebo after six weeks, but no different after 14 weeks.


One randomised trial assessed the effect of 3-15 minutes of massage therapy daily for 10 days in 28 newborns who had been exposed to HIV. The control group received no treatment. The trial reported improved performance and weight gain in the massage group. It was not stated if this difference was significant.

Adverse effects

The reviewers' stated that the common belief that complementary therapies do more good than harm is not upheld by randomised controlled trials. No further details were provided.



Many different treatments are assessed in this review; all were based on few data from studies which were small and of variable quality. Different outcomes were used in each of the trials which assessed stress management. This meant that studies could not be compared and the effectiveness of this intervention on HIV/AIDs is unknown. None of these treatments worked except, perhaps, massage, but it was not stated whether significant improvements occurred with massage treatment for weight gain.

Further reading

1. Ernst E. Complementary AIDS therapies: the good, the bad and the ugly (editorial review). Int J STD AIDs 1997; 8: 271-285.

2. Ernst E, White AR. Acupuncture may be associated with serious adverse events (letter). BMJ 2000;320;513.

Related topics