Short-term impacts of a change in ART initiation threshold for patients co-infected with TB in Johannesburg, South Africa

Abstract

Introduction: Recent studies showed earlier antiretroviral therapy (ART) initiation in patients co-infected with tuberculosis (TB) leads to reduced mortality. On April 1st, 2010 South Africa changed its HIV treatment guidelines to initiate patients with TB at CD4 counts <350 cells/mm3, up from <200 cells/mm3.  We evaluated short-term impacts of the guideline change by comparing treatment outcomes for co-infected patients before and after the policy change at the Themba Lethu Clinic, in Johannesburg, South Africa. Methods: We compared treatment outcomes among 373 co-infected patients initiated on ART between April 2009-March 2010 (pre-period) to 410 patients initiated between April 2010-March 2011 (post-period). TB-HIV co-infection was defined as any episode of pulmonary TB 3 months prior to 30 days after ART initiation. We compared rates of death or loss to follow-up (LTF) over the first six months on ART. We then adjusted for age, sex and ART regimen using logistic regression. LTF was defined as three months late for a scheduled visit. Our primary outcomes is death or LTF combined as deaths are more likely to be distinguished from LTF in the pre-period. Results: Of the 783 TB/HIV co-infected patients 48.0% were female. They had a median age of 36.3 (31.4-42.8) years at ART initiation and a median follow-up of 11.1 person-months. Due to changes in the 2010 treatment guidelines, 85% of patients initiated d4T-3TC-EFV as their first ART regimen in the pre-period, while 74% initiated TDF-3TC-EFV in the post-period. The mean CD4 count at ART initiation was 82.4 cells/mm3 in the pre-period and 101.5 cells/mm3 in the post-period (mean difference 19.1 cells/mm3; 95% CI: 5.3-32.8). Of all ART patients, 12.2% patients were co-infected with TB in the pre-period compared to 12.5% in the post-period. 210 patients died or were LTF after a median of 4.9 (IQR: 3.9-8.6) person-months. After adjustment for age, sex and ART regimen, the odds of death or LTF within 6 months of ART initiation was unchanged in the post-period compared to the pre-period (OR=0.98; 95%CI: 0.56-1.72). When limited to just the 391 patients initiated onto d4T-3TC-EFV, the odds of death or LTF within 6 months was also unchanged (OR=1.06; 95% CI: 0.52-2.19). Conclusions: While early outcomes for TB patients were similar pre- and post- policy change, our data suggest that initiating CD4 counts have begun to increase. If this trend continues it could lead to better outcomes for co-infected patients.

Conference: CROI 2012, Seattle, USA