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


Review: 6- and 12-month courses of isoniazid prevent active tuberculosis in HIV-negative persons

ACP J Club. 1999 July-Aug;131:16. doi:10.7326/ACPJC-1999-131-1-016

Source Citation

Smieja MJ, Marchetti CA, Cook DJ, Smaill FM. Isoniazid for preventing tuberculosis in non-HIV infected persons. Cochrane Review, latest version 21 Nov 1998. In: The Cochrane Library. Oxford: Update Software.



What is the effectiveness of a 6- or 12-month course of isoniazid (INH) for preventing active pulmonary tuberculosis (TB) in at-risk persons who are HIV-negative?

Data sources

Studies were identified by searching MEDLINE (1966 to 1998), EMBASE/Excerpta Medica (1974 to 1998), and the Cochrane Library using combinations of the terms tuberculosis, isoniazid, prophylaxis, prevention and control, and clinical trials; hand searching Science Citation Index (1955 to 1993), Current List of Medical Literature (1955 to 1959), and the Cumulated Index Medicus (1960 to 1970); and reviewing the bibliographies.

Study selection

Studies were selected if they were published randomized controlled trials that compared a ≥ 6-month course of INH alone (≥ 200 mg/d or 5 mg/kg of body weight) with placebo and had at least 2 years of follow-up. Patients were at risk for TB but did not have active TB at study enrollment. Studies of children with primary TB were excluded.

Data extraction

Data were extracted on study methods, participants, interventions, and outcomes. Methodologic quality (method of randomization, blinding, verification of outcomes, and length of follow-up) of individual studies was assessed.

Main results

11 studies (n = 73 375) met the selection criteria and were included in the analysis. All studies were double-blinded and had > 80% follow-up. Baseline risk for TB in all patients randomized to placebo (n = 33 113) was 1.7% over 5 years. A random effects model was used for all analyses. Patients who received INH were less likely to develop active TB (11 studies) (Table). INH also reduced extrapulmonary TB (4 studies). The groups did not differ for total mortality (5 studies) or TB deaths (2 studies). In 1 study (n = 20 874), patients receiving INH had higher rates of hepatitis than those receiving placebo.


6- and 12-month courses of isoniazid reduce the risk for active pulmonary tuberculosis in at-risk persons who are HIV-negative.

Sources of funding: European Commission, Belgium, and Department for International Development, UK.

For correspondence: Dr. M. Smieja, Laboratory Medicine, McMaster University Medical Centre, Microbiology HSC 2N29, 1200 Main Street West, Hamilton, Ontario L8N 3Z5, Canada. FAX 905-521-5099.

Table. Isoniazid vs placebo for tuberculosis prevention in at-risk persons

Outcome at ≥ 2y Isoniazid Placebo RRR (95% CI) NNT (CI)
Active tuberculosis 0.28% 1.68% 60% (48 to 69) 72 (50 to 143)

*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.


When INH was first studied in large clinical trials of > 50 000 persons, no deaths were reported from INH hepatotoxicity (1). After the introduction of INH into general practice, a mortality rate of ≤ 0.5% was reported for persons ≥ 35 years of age (2). These deaths and the ensuing controversy over whether INH should be given to so-called low-risk reactors (persons with positive tuberculin tests but normal chest radiographs and no risk factors for reactivation) were generally taken to mean that INH was no longer indicated for any clinical situation. As a result, INH is underutilized in North America.

The ideal duration of INH therapy for latent TB infection is controversial. In this meta-analysis by Smieja and colleagues, only 1 study directly addressed this question and found that 12 months of therapy was more effective than 6 months (93% vs 75% risk reduction) in patients who took > 80% of the medication each month (3). However, compliance was lower in patients assigned to longer treatment, so overall effectiveness was not greater (75% vs 65%) (3). A recent reanalysis of earlier studies suggests that 9 months may be the optimal duration of therapy (4). Evidence also exists that longer duration of therapy is more beneficial for patients with a higher risk for disease (3). Close follow-up of patients included in this meta-analysis resulted in early detection of TB, which averted death, and explains why groups did not differ for total or tuberculosis mortality.

INH treatment of latent TB infection has a long track record. When patients comply with therapy, INH is highly efficacious in reducing risk for TB. If patients are carefully monitored, the risk for fatal hepatotoxicity appears to be low (5). The major challenge now is to convince physicians to prescribe INH for patients at risk for TB and for physicians to convince their patients to take the therapy regularly for the required period.

Richard Menzies, MD
McGill UniversityMontreal, Quebec, Canada


1. Ferebee SH. Controlled chemoprophylaxis trials in tuberculosis. Adv Tuberc Res. 1969;17:28-106.

2. Kopanoff DE, Snider DE Jr, Caras GJ. Isoniazid-related hepatitis: a U.S. Public Health Service cooperative surveillance study. Am Rev Respir Dis. 1978;117:991-1001.

3. International Union Against Tuberculosis Committee on Prophylaxis. Efficacy of various durations of isoniazid preventive therapy for tuberculosis: five years of follow-up in the IUAT trial. Bull WHO. 1982;60:555-64.

4. Comstock GM. How much isoniazid is needed for prevention of tuberculosis in immunocompetent adults. Int J Tuberc Lung Dis. 1999. [In press].

5. Nolan CM, Goldberg SV, Buskin SE. Hepatotoxicity associated with isoniazid preventive therapy: a 7-year survey from a public health tuberculosis clinic. JAMA. 1999;281:1014-8.