Loss to follow-up and mortality among HIV-infected people co-infected with TB at ART initiation in Durban, South Africa

Abstract

Objective: To quantify the impact of tuberculosis (TB) co-infection on death and loss to follow-up (LTFU) 12 months after entry into an ART program.

Design: Prospective intervention study.

Methods: From May 2007 to 2008, patients undergoing pre-ART training in Durban, South Africa, were screened for pulmonary TB using mycobacterial culture. Subjects missing appointments for >3 months were phoned. Patients who could not be reached were considered LTFU. Deaths were ascertained by report from family members. We used the Kaplan-Meier method to estimate time to LTFU or death for 3 groups at enrollment as follows: (1) newly diagnosed with TB by sputum culture; (2) on TB treatment (ie, previously diagnosed); and (3) TB free. We evaluated the role of TB on mortality and LTFU using Cox proportional hazards models.

Results: Nine hundred fifty-one HIV-infected subjects were enrolled; 59% were female, and median baseline CD4 count was 90 cells per microliter (IQR: 41-148 cells/μL). One hundred forty-four (15%) were newly diagnosed with TB by sputum culture; an additional 199 (21%) were already on TB treatment. By 12 months, 26% newly diagnosed with TB at enrollment died or were LTFU, compared with 19% already on TB treatment, and 14% who were TB free (P = 0.001). Controlling for age, sex, smoking, CD4, and opportunistic infection history, subjects newly diagnosed with pulmonary TB were 76% more likely to die or be LTFU (hazard ratio: 1.76, 95% confidence interval: 1.20 to 2.60) than those without TB.

Conclusions: HIV/TB co-infected individuals are more likely to die or be LTFU within 12 months of ART clinic entry in South Africa. These patients require intensive follow-up during ART initiation.

Figure 1. Flow chart of study population, Sinikithemba HIV clinic, McCord Hospital, Durban, South Africa, 2007-2008

A schematic of HIV-infected participants undergoing ART literacy training at the HIV clinic who were screened, enrolled, eligible to initiate ART, and diagnosed with pulmonary TB by sputum culture. ART: antiretroviral therapy; TB: Tuberculosis

Figure 2. Kaplan-Meier curve of the combined endpoints of death or LTFU

The 3 groups of patients are classified based on TB status at enrollment: 1) newly-diagnosed with TB by sputum culture, 2) already taking TB treatment, and 3) TB free. LTFU: Lost to follow-up

Figure 3. Sensitivity analysis showing Cox proportional hazards model for the combined outcome of death or LTFU at 12 months based on TB status and stratified by baseline CD4 count among HIV-infected adults in Durban, South Africa

The three TB groups: 1) newly-diagnosed with TB by sputum culture, 2) already taking TB treatment for known/presumed active disease prior to enrollment, and 3) TB free, meaning a negative enrollment sputum culture and not on TB treatment, are further stratified based on CD4 count ≥100/μl and CD4 <100/μl. Analysis is adjusted for smoking status, age, opportunistic infection history, and gender. Vertical lines represent 95% CI around hazard ratios.