Current issues of ACP Journal Club are published in Annals of Internal Medicine


Isoniazid prophylaxis reduced the incidence of tuberculosis in HIV infection and increased time to progression of symptomatic HIV disease and AIDS

ACP J Club. 1994 Mar-April;120:38. doi:10.7326/ACPJC-1994-120-2-038

Source Citation

Pape JW, Jean SS, Ho JL, Hafner A, Johnson WD Jr. Effect of isoniazid prophylaxis on incidence of active tuberculosis and progression of HIV infection. Lancet. 1993 Jul 31;342:268-72.



To determine whether isoniazid prevents the development of active tuberculosis (TB) in symptom-free patients who are seropositive for human immunodeficiency virus (HIV).


Randomized controlled trial starting in 1986. After an interim analysis in 1990, the study was open labeled.


A national referral center for the acquired immunodeficiency syndrome (AIDS) in Haiti.


118 symptom-free adults (mean age 31 y, 77% women) with newly diagnosed HIV seropositivity. The skin test reaction to purified protein derivative of tuberculin (PPD) was positive (> 5 mm induration) in 63 patients. Exclusion criteria were a history of TB or of abnormal chest radiographs or liver function tests. Mean follow-up was 34 months in the treatment group and 40 months in the control group.


58 patients were allocated to receive isoniazid, 300 mg/d, and vitamin B6, 50 mg/d; 60 patients were allocated to receive vitamin B6 only. Medication was given for 1 year. After an interim analysis, patients taking only vitamin B6 could choose to take isoniazid (21 of 60 accepted).

Main outcome measures

Active culture-proven TB, symptomatic HIV disease, AIDS, and death. Patients were examined every 3 months.

Main results

For the whole study population, compared with patients taking isoniazid, patients taking vitamin B6 alone had a higher risk for developing TB (11 [18%] vs 4 [7%] patients, relative risk [RR] 3.4, 95% CI 1.1 to 10.6). A similar increase was seen in the subset of patients who were PPD positive (RR 5.8, CI 1.2 to 27.8) but not in the patients who were PPD negative. Excluding patients who developed TB, vitamin B6 alone compared with isoniazid increased the incidence of symptomatic HIV disease (up to the year 1990, 30 [50%] vs 14 [24%] patients, RR 2.1, CI 1.2 to 3.5) and of AIDS (12 [20%] vs 2 [3%] patients, RR 5.8, CI 1.4 to 24.8). Isoniazid delayed the time to HIV disease (32 vs 19 mo, P < 0.05) and to AIDS (38 vs 30 mo, P < 0.05). The subset of patients who were PPD positive, but not those patients who were PPD negative, had a similar increase in time to HIV disease and AIDS. The study groups did not differ for mortality.


Isoniazid reduced the incidence of active tuberculosis and increased the time to development of symptomatic HIV disease and AIDS for the entire study group. The results were similar for patients who were initially positive for PPD, but there were no differences for patients who were PPD negative.

Source of funding: Public Health Service.

For article reprint: Dr. W.D. Johnson, Jr., Division of International Medicine, Cornell University Medical College, A-431, 1300 York Avenue, New York, NY 10021, USA. FAX 212-746-8675.


Since this study by Pape and colleagues showing efficacy in the use of isoniazid chemoprophylaxis in dual HIV-tuberculous infected patients was published, there has been a substantial increase in our understanding of the pathogenesis of HIV-related TB. The recently published meta-analysis evaluating the role of preventive therapy in HIV infected patients has extended and further defined the role of chemoprophylaxis (1). Wilkinson and colleagues identified 7 trials, 5 of which met their eligibility criteria. 3 of the studies were located in developing countries and 1 in the United States. All studies evaluated ≥ 6 months of isoniazid and, in 1 study, a 6-month regimen was compared with isoniazid and rifampin for 3 months and with isoniazid, pyrazinamide, and rifampin for 3 months (2). The studies used self-administered therapy. Various mechanisms were used for evaluating adherence, including self reporting, attendance at scheduled clinic appointments, and urine testing. Overall, the risk for TB was reduced with an RR of 0.57 (95% CI 0.41 to 0.76) but mortality was not significantly reduced (RR 0.93, CI 0.83 to 1.05). When the pooled risk was evaluated, the TB risk was reduced by 0.32 (CI 0.19 to 0.51) and mortality risk by 0.73 (CI 0.57 to 0.95). The differences were not significant for those patients who had a negative tuberculin reaction. Adverse reactions were more common in the treatment group, but not to a signficant extent (RR 1.45, CI 0.98 to 2.14). The follow-up varied between 15 and 33 months. Only the study by Pape and colleagues looked at the effect of isoniazid on the development of symptomatic HIV disease.

Chemoprophylaxis undoubtedly has an important role to play in the management of tuberculosis and, in particular, HIV-related TB. Although chemoprophylaxis is to be encouraged, until such time as adequate chemotherapy is available and delivered systematically in developing countries, its overall role in TB control needs to be considered carefully (3). These data suggest that widespread use of chemoprophylaxis among PPD-negative patients, even in settings with a high prevalence of dual HIV and tuberculous infection, cannot be recommended. The potential for ultra short-course chemoprophylaxis with rifampin and pyrazinamide needs to be studied in a twice weekly, supervised format. With evidence from molecular epidemiological studies showing that there is a risk for reinfection, particularly in the setting of HIV infection, trials addressing the duration of chemoprophylaxis need to carefully differentiate reinfection from relapse.

J. Mark FitzGerald, MB
University of British ColumbiaVancouver, British Columbia, Canada


1. Wilkinson D, Squire SB, Garner P. Effect of preventive treatment for tuberculosis in adults infected with HIV: systematic review of randomised placebo controlled trials. BMJ. 1998;317:625-9.

2. Whalen CC, Johnson JL, Okwera A, et al. A trial of three regimens to prevent tuberculosis in Ugandan adults infected with the human immunodeficiency virus. Uganda-Case Western Reserve University Research Collaboration. N Engl J Med. 1997;337:801-8.

3. FitzGerald JM. The downside of isoniazid chemoprophylaxis. Lancet. 1995;345:404.