Transmission of Extensively Drug-Resistant Tuberculosis in South Africa

Tuberculosis (TB) remains a significant cause of morbidity and mortality globally and a leading cause of death among people with human immunodeficiency virus (HIV). Although TB is a treatable and curable condition, cases of multidrug-resistant tuberculosis (MDR-TB) have emerged because of inadequate treatment, poor treatment adherence, or subtherapeutic drug levels. In the last decade, cases of extensively drug-resistant tuberculosis (XDR-TB) have emerged due to acquired resistance against second-line therapy for MDR-TB and transmission of these resistant strains. The burden of XDR-TB has been increasing in South Africa in part due to the high rate of coinfection with HIV-1. In order to identify an effective intervention to reduce the development of XDR-TB, we need to determine whether inadequate treatment or transmission of resistant strains is driving the XDR-TB epidemic.    

In this week’s issue of NEJM, Shah and colleagues prospectively examined XDR-TB transmission in patients in KwaZulu-Natal, South Africa. Patients were considered to have developed XDR-TB by acquired resistance if they had a diagnosis of MDR-TB (self-reported or in the medical record) before the diagnosis of XDR-TB, they received documented treatment with second-line therapy, or drug-susceptibility tests demonstrated resistance to isoniazid and rifampicin. Patients who did not meet any of these criteria were considered to have developed XDR-TB through transmission of resistant strains. Targeted gene sequencing of Mycobacterium tuberculosis isolates were used to characterize a genotypic cluster due to transmission. Further, social network and geospatial data were used to analyze person-to-person links and hospital and community locations of transmission.

Of the 1027 patients who were diagnosed with XDR-TB between 2011 and 2014, 404 were enrolled in the study; 77% were coinfected with HIV-1 (median CD4 count, 340 cells/mm³), and 50% of these patients had an undetectable HIV-1 viral load. Overall, 31% of study participants had been treated for MDR-TB before the XDR-TB diagnosis; these cases were presumed to have been caused by acquired resistance (most had documentation of treatment failure). The remaining 69% of participants who had not previously been treated for MDR-TB were presumed to have acquired XDR-TB through transmission of resistant strains; 61% of these patients were in a genotypic cluster, implying that XDR-TB was due to transmission rather than acquired resistance.

Social network analysis identified person-to-person or hospital-based epidemiological links for 30% of study participants. Most participants (75%) had been hospitalized in the 5 years prior to study enrollment and 39% of these patients were admitted before they received the XDR-TB diagnosis. Of those admitted before the XDR-TB diagnosis, 61% had a hospital-based link with another study participant. Participants spent substantial time in community areas such as churches, bars, and restaurants.    

The authors note that case definitions of acquired resistance versus transmission of resistant strains can be subject to misclassification. Nonetheless, the study provides interesting insights about how the global health community needs to approach XDR-TB control. Ensuring that patients with TB are treated appropriately is a crucial aspect to reducing acquired resistance. Learning how to interrupt transmission in communities is also important to consider, although difficult and less well-defined.

Browse more From Pages to Practice »

Ramya Ramaswami is a 2016-2017 NEJM editorial fellow. She is a medical oncologist within the National Health Services of the United Kingdom. Ramya received her medical degree, postgraduate medical and oncology training from Imperial College London, and a masters in public health from Columbia University, Mailman School of Public Health. Her clinical and research interests include cancer prevention, viral driven cancers, as well as disparities and access issues in global oncology.