Infection with tuberculosis (TB) remains the leading cause of mortality among HIV-infected people worldwide . South Africa has the third highest TB incidence in the world, with ~1,000 cases per 100,000 people per year and the most HIV-infected citizens of any country , . Even for people successfully diagnosed with TB in South Africa, TB treatment completion remains a challenge. In 2005, KwaZulu-Natal, the South African province with the highest HIV prevalence, reported the greatest number of smear-positive pulmonary TB cases in the country and had the lowest cure rate (45%) .
The emergence of drug-resistant TB, which is challenging to diagnose and treat, threatens to undo the progress made in HIV treatment and care within an overburdened public health infrastructure in South Africa , .
In 2005, a substantial burden of multi-drug (MDR) and extensively-drug resistant (XDR) TB was identified among HIV-infected patients in the rural Msinga District of KwaZulu-Natal. Enhanced surveillance there revealed 39% MDR TB and 6% XDR TB among HIV-infected patients with culture-confirmed TB, exceeding previous surveys . However, province-wide surveillance in KwaZulu-Natal has largely reflected sputum cultures performed on request based on clinical suspicion of treatment failure. This practice has varied widely among districts, with an estimated 80% of reported cases lacking culture confirmation and drug susceptibility testing , . In addition, individual-level data, such as prior TB treatment exposure and HIV status, have not been available in population-based surveys of TB drug resistance.
In the setting of an intensive TB screening program using mycobacterial culture and drug susceptibility testing for all patients, our objective was to assess the prevalence of drug-resistant TB and to describe the resistance patterns in TB culture positive patients commencing ART in an urban HIV clinic in Durban, South Africa.
To estimate the prevalence of drug-resistant tuberculosis (TB) and describe the resistance patterns in patients commencing antiretroviral therapy (ART) in an HIV clinic in Durban, South Africa.
Cross-sectional cohort study.
Consecutive HIV-infected adults (≥18y/o) initiating HIV care were enrolled from May 2007–May 2008, regardless of signs or symptoms of active TB. Prior TB history and current TB treatment status were self-reported. Subjects expectorated sputum for culture (MGIT liquid and 7H11 solid medium). Positive cultures were tested for susceptibility to first- and second-line anti-tuberculous drugs. The prevalence of drug-resistant TB, stratified by prior TB history and current TB treatment status, was assessed.
1,035 subjects had complete culture results. Median CD4 count was 92/µl (IQR 42–150/µl). 267 subjects (26%) reported a prior history of TB and 210 (20%) were receiving TB treatment at enrollment; 191 (18%) subjects had positive sputum cultures, among whom the estimated prevalence of resistance to any antituberculous drug was 7.4% (95% CI 4.0–12.4). Among those with prior TB, the prevalence of resistance was 15.4% (95% CI 5.9–30.5) compared to 5.2% (95% CI 2.1–8.9) among those with no prior TB. 5.1% (95% CI 2.4–9.5) had rifampin or rifampin plus INH resistance.
The prevalence of TB resistance to at least one drug was 7.4% among adults with positive TB cultures initiating ART in Durban, South Africa, with 5.1% having rifampin or rifampin plus INH resistance. Improved tools for diagnosing TB and drug resistance are urgently needed in areas of high HIV/TB prevalence.