RÉSUMÉ
BACKGROUND: Tuberculosis remains a public health threat responsible as recently as 2018 for more than one million deaths. Chemoprophylaxis with isoniazid is one of the strategies implemented to control the disease. Although it is not yet widely prescribed, its utilization raises additional questions in the "test and treat" era of for anti-retroviral therapy. The objective of this study is to review the different randomized controlled trials of antitubercular Isoniazid Preventive Therapy (IPT). We have distinguished (a) "efficacy trials" (ET) comparing IPT to a placebo or the absence of chemoprophylaxis and (b) "IPT regimen trials" (RT) comparing IPT to one or several other regimens. METHODS: Literature search (keywords from published articles found in the Medline and Scopus data bases: "tuberculosis", "prophylaxis", "HIV", "randomized controlled trial") and standardized reading of selected articles reporting results from randomized trials of IPT in HIV-infected people. RESULTS: Eighteen selected trials (11 ET and 7 RT), including 19,725 participants. The regimens studied were 3H, 6H, 9H, 12H, 12H, 36H/2RZ, 3RH, 3RZ, 3RHZ, and 3HP [H: Isoniazid, R: Rifampicin, Z: Pyrazinamide, P: Rifapentine]. LOCATIONS: Ten in Africa, three in Haiti, one in India, one in the USA, one in the Americas and two multi-continental trials. In ET with or without antiretrovirals (ART), IPT significantly reduces the risk of tuberculosis, by 32 to 71%. In ET prior to ART, IPT does not appear to reduce mortality. In ET in patients receiving ART, on the other hand, IPT reduces mortality. As regards RT, there seems to be no reason to prefer other regimens to IPT. Tolerance is good. Importantly, IPT may reduce (rather than worsen) the risk of multidrug-resistant bacilli selection by decreasing the number of TB episodes and, consequently, the number of curative tuberculosis treatments. CONCLUSION: Far from becoming obsolete due to ARV treatment, IPT has remained a timely and relevant intervention.
