The efficacy of intermittent directly observed isoniazid in preventing tuberculosis in HIV-infected adults with advanced disease

[Introduction] Meta-analysis of the treatment of latent tuberculosis infection (LTBI) in HIV-infected adults has shown significant reduction in the incidence of tuberculosis in participants with a positive tuberculin skin test (TST), but not in those with a negative TST. However, there are insufficient data on patients with advanced HIV disease from high tuberculosis incidence areas. It is important to exclude tuberculosis prior to such preventive therapy, but this can be difficult in patients with symptomatic HIV disease. A tuberculosis screening instrument is thus needed to ensure that patients placed on preventive therapy do not have tuberculosis. Furthermore, to ensure adherence and avoid drug resistance optimal supervision of the treatment administrations is required. [ Methods ] Patients with clinically advanced HIV disease were screened for active tuberculosis using a symptom questionnaire, measured weight loss, chest radiography, sputum microscopy and culture prior to receiving tuberculosis preventive therapy. Once tuberculosis was excluded, a randomized doubleblind trial was conducted comparing INH with placebo among TST negative status participants with WHO Stage 3 or 4 HIV disease. INH/placebo was administered for 12 months by patient-nominated supervisors. TST-positive participants were given open-label INH. Participants who did not have access to ART were followed up for 24 months with 6-monthly sputum culture and chest radiography. All those enrolled for the trial were required to visit a clinic on a monthly basis for 12 months during the period of weekly intermittent supervised administration of INH/placebo to assess for tuberculosis and adherence. [ Results ] A total of 118 participants were enrolled: TST was negative in 98. Tuberculosis was diagnosed in 11 of 129 patients screened. A simple screening instrument of two or more of the symptoms cough, night sweats or fever, (plus measured weight loss) had a sensitivity of 100% and specificity of 88.1% (against the gold standard of sputum culture) and positive and negative predictive values of 44% and 100%, respectively. In the randomized trial arms, the incidence of tuberculosis was 18/100 person-years (py) in the INH arm and 11.6/100 py in the placebo arm [hazard ratio 1.59, 95% confidence interval (CI) 0.57-49)]. There was no significant difference in mortality, hospitalization rate or CD4+ lymphocyte decline. Patient adherence for INH/placebo was 8 5% and was significantly higher among participants with work-based treatment supervisors than among those who were supervised by home-based or community-based treatment supervisors. The daily self-administered treatment (SAT) of cotrimoxazole (CTX) showed a good adherence especially among the TST positive participants, where a greater benefit in terms of survival among participants with good cotrimoxazole adherence was observed.

Grantor dc:publisher.institution

Department of Public Health and Family Medicine

Year dc:date.issued

2008

Author dc:creator

• Mohammed, Ashraf Allie

Advisors dc:contributor.advisor

• Maartens, Gary
• Ehrlich, Rodney

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/9426

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/9426