Long-term effectiveness of isoniazid prophylaxis on TB incidence in a cohort of IDUs

Jonathan Golub and co-workers from the AIDS Linked to Intravenous Experience (ALIVE), cohort in Baltimore USA assessed long-term effectiveness of an 8-year tuberculin skin testing (TST)/isoniazid (IPT) programme among a cohort of HIV-positive and HIV-negative IDUs.

This cohort includes >2000 IDUs in Baltimore, 35% of whom were HIV-positive at baseline. TST and IPT were offered to all ALIVE participants from 1990 to 1998. TB incidence was measured in 3 periods:  pre-purified protein derivative (PPD) era (1988-1990), PPD era (1990-1998), and post-PPD era (1998-2004). Incidence rate ratios compared TB incidences among eras.

The investigators found out of a group of 753 HIV-positive participants, 651 (86%) had a TST ; 103 (16%) had a positive result (>5 mm); 65 (60%) started IPT; and 40 (62%) completed 6 months. Of the1264 HIV-negative participants, 1105 (87%) had a TST; 435 (39%) had a positive result (>10 mm); 246 (56%) started IPT; and 133 (54%) completed 6 months. In total, 32% of those with a positive TST completed 6 months of IPT.

In this study 30 TB cases were diagnosed over 28,750 person-years:  IR = 1.04/1000 person-years in HIV-negative; IR = 2.66/1000 person-years among the HIV-positive population. The investigators reported TB incidence in the post-PPD-era for the overall cohort was half that seen in the PPD-era (IRR = 0.44, 95%CI 0.19 to 1.04), but they found no significant difference between eras in the HIV-positive population (2.04 vs 3.14/1000 person-years; IR = 0.64, 95%CI 0.27 to 1.58).

Both overall and amongst the HIV-positive participants, TB incidence among those who never received IPT was greater than those who started IPT; no cases were detected for those who received 6 months of IPT. Among the HIV-positive participants the investigators found body mass index <21 (RH = 3.1, p <0.01) and CD4 <200 (RH = 9.6, p <0.01) to be most predictive of TB. ART use had no association with risk of TB.

The investigators reported that a significant long-term reduction in TB incidence was observed in a cohort of IDU with a high HIV prevalence after an 8-year strategy of TST/IPT, but no change was seen in the HIV-positive subset. They noted that IPT was highly effective for those who completed it, but only 32% of TST-positive patients completed. “Broader use of IPT in HIVIDUs could substantially decrease TB incidence” they wrote.


Jonathan Golub, J Astemborski, M Ahmed et al. Long-term effectiveness of isoniazid preventive therapy on TB Incidence in a cohort of injection drug users. 14th CROI, 2007. Los Angeles. Abstract 851.

Links to other websites are current at date of posting but not maintained.