University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Effects of duration of HIV infection and secondary tuberculosis transmission on tuberculosis incidence in the South African gold mines
Global Health Sciences Literature Digest
Published March 10, 2009
Journal Article

Glynn J, Murray J, Bester A, Nelson G, Shearer S, Sonnenberg P. Effects of duration of HIV infection and secondary tuberculosis transmission on tuberculosis incidence in the South African gold mines. AIDS 2008;22:1859-67.


To estimate the increase in the incidence of tuberculosis (TB) in HIV directly through immunosuppression and indirectly through onward transmission of Mycobacterium tuberculosis

Study Design

A retrospective cohort study


South African gold mines, utilizing medical and occupational records, including results of HIV testing that began in 1991


Male South African gold miners working in four mines in Gauteng Province, South Africa from 1991 through 2004. Records of men with negative HIV test results (n=5814) were compared to men who seroconverted after HIV testing began (n=1950).



Primary Outcomes

The primary outcome was TB incidence rates. Aside from overall trends, analyses were restricted to the first episode of pulmonary TB in someone with no known previous TB. Analyses were repeated restricted to culture-confirmed cases


TB incidence increased both with age and with later calendar period among both HIV-infected and HIV-uninfected men and by time since seroconversion among HIV-infected men. By 11 years after seroconversion, nearly half of the HIV-infected miners had been diagnosed with TB. Rates of pulmonary TB increased shortly after seroconversion to 1.4/100 person-years and rose to 10/100 person-years at 10 or more years after seroconversion. In men with seroconversion intervals of less than one year (58%), the incidence of TB was 1.9/100 person-years in the first year and 1.6/100 person-years in the second year. Among HIV-uninfected miners, the TB rate rose from 0.48/100 person-years to 1.2/100 person-years over the period of the study. The incidence in the HIV-uninfected group doubled over the period of the study after adjusting for age.

Rate ratios of TB in HIV-infected and HIV-uninfected men, after adjusting for age and calendar period, showed a nine-fold increase in TB incidence by 10 years post-seroconversion. In an age-adjusted model estimating the increase in rates due to the increasing duration of HIV alone, the incidence at 10 or more years from seroconversion is 7.1/100 person-years. A model adjusting the rates to age 30-35 years and to the earliest time period (1991-1993) gave an estimate of the proportion of the increase due to increasing duration of HIV infection (i.e., increasing immunosuppression) alone. The incidence in the age-adjusted model at 10 or more years was 7.1/100 person-years, reduced to 3.8/100 person-years when adjusted for calendar period.

Because many HIV-uninfected miners lacked serial HIV test results, a sensitivity analysis was conducted by censoring the HIV-uninfected men one year after the last negative test, with results similar to the original analysis. When analysis was restricted to culture-confirmed cases (85%), the incidence among HIV-infected men was 3.1/100 person-years and 0.63/100 person-years among HIV-uninfected men. Similar estimates of rate ratios at two and 10 years were found.


The authors conclude that the increase in TB risk by time since seroconversion is a function of both the direct effects of HIV increasing susceptibility and the indirect effects of onward transmission of TB. It points out the need for measures to reduce TB transmission.

In Context

The evidence relating to the increased TB risk in the context of HIV requires cohorts with known dates of seroconversion to be followed for long periods of time. Thus, there have been few studies able to examine the contributory causes of increasing TB incidence in HIV-infected compared with HIV-uninfected populations,(1,2,3) including the potential indirect effect of HIV on TB incidence through onward transmission.(4) This study shows the impact of both direct and indirect effects of HIV on TB incidence and the importance of case finding and TB treatment in HIV-infected and HIV-uninfected populations.

Programmatic Implications

Although the risk of TB is reduced with antiretroviral therapy (ART), it remains elevated compared with the risk in HIV-uninfected persons, and with longer survival times, the case load and onward transmission will likely increase in all populations. The authors recommend a combination of widespread ART; rapid, active case finding; and isoniazid prophylaxis as part of a sustained TB program to control the epidemic.


  1. van Asten L, Langendam M, Zangerle R, et al. Tuberculosis risk varies with the duration of HIV infection: a prospective study of European drug users with known date of HIV seroconversion. AIDS 2003;17:1201-8.
  2. Minga A, Danel C, Abo Y, et al. Progression to WHO criteria for antiretroviral therapy in a 7-year cohort of adult HIV-1 seroconverters in Abidjan, Cote d'Ivoire. Bull World Health Organ 2007;85:116-23.
  3. Deschamps MM, Fitzgerald DW, Pape JW, JohnsonWD Jr. HIV infection in Haiti: natural history and disease progression. AIDS 2000;14:2515-21.
  4. Sonnenberg P, Glynn JR, Fielding K, Murray J, Godfrey-Faussett P, Shearer S. How soon after infection with HIV does the risk of tuberculosis start to increase? A retrospective cohort study in South African gold miners. J Infect Dis 2005;191:150-8.