Rising CD4 counts and suppressed viral load on antiretroviral treatment have had a far greater impact on reducing new cases of tuberculosis (TB) in people with HIV in Switzerland than testing and preventive treatment for latent TB, researchers from the Swiss HIV Cohort report in the journal Clinical Infectious Diseases.
Background
Exposure to TB bacteria may lead to the rapid development of active TB if the immune system is weakened by HIV or malnutrition. But in most people exposed to TB, the bacteria are walled off by immune cells after entering the lungs. This condition is called latent TB and may last for many years. People with latent TB have no symptoms of infection but may test positive on a skin test or blood test for TB.
After testing positive, latent TB infection can be cleared by a course of antibiotic treatment with either isoniazid alone or rifampicin alone or isoniazid plus either rifampicin or rifapentine.
Without preventive treatment, latent TB can turn into active TB if the immune system can’t keep the TB bacteria isolated. The most common reason for reactivation of TB is immune suppression, either due to HIV or another stressor.
Although testing for latent TB and preventive treatment are recommended for people with HIV, the real-world effectiveness of testing and preventive treatment in settings with low rates of TB has been uncertain.
In previous decades, TB was one of the most common opportunistic infections in people with HIV in higher-income settings. Some studies in Europe have shown declines in the proportion of people diagnosed with TB who also have HIV (see this British study) but a UK study of the period up to 2017 found that TB rates remained high among Black African people with HIV.
The study
Swiss HIV Cohort researchers investigated the impact of testing for latent TB, preventive treatment for TB, antiretroviral treatment and viral suppression on the prevalence of latent TB and active TB in their cohort between 1988 and 2022. They counted the number of people diagnosed with TB within six months of enrolment in the cohort as prevalent cases of active TB. TB diagnoses were not counted as prevalent cases if they were relapses in people recently treated for TB or cases of reinfection after previous successful treatment of TB.
Cases of latent TB were defined as those people with HIV who tested positive on a skin test or an interferon gamma release assay (IGRA blood test) within six months of enrolment in the cohort. They also counted the number of people with latent TB who started preventive treatment for TB and assessed its effectiveness in preventing TB.
During the follow-up period, 21,528 people with HIV were enrolled in the Swiss HIV Cohort. Of these, 68% underwent testing for latent TB and 1,233 tested positive (8% of all people with HIV in the cohort). The prevalence of latent TB declined from 15% in 2001 to 4% in 2021.
Eight hundred and twenty-five people with HIV were diagnosed with active TB; 33% of these cases were classified as incident cases of TB, preventable cases of TB that occurred more than six months after enrolment into the cohort. Whereas 4% of those who joined the Swiss HIV Cohort in 1993 either had TB at enrolment or were diagnosed with TB within six months of enrolment, no one enrolled in 2020 after a recent HIV diagnosis either had TB at enrolment or went on to be diagnosed with TB within six months.
During a median of 14 years of follow-up after a positive test for latent TB, 29 people developed active TB, a cumulative incidence of 2.3%. Nine cases occurred in people who received preventive treatment. Preventive treatment reduced the relative risk of developing TB by 28%. The prevalence of active TB among newly enrolled cohort participants fell sharply from 2010 onwards while the rate of incident active TB fell continuously throughout the follow-up period.
However, when the researchers looked at adherence to preventive treatment (for which they used adherence to antiretroviral treatment as a proxy), preventive treatment was associated with a 76% reduction in the relative risk of TB in those who were adherent to treatment. Lack of preventive treatment was consistently associated with a higher risk of developing TB across all time periods analysed (before 1996, 1997-2007, after 2007).
The researchers analysed risk factors for active TB in people who had recorded results of a test for latent TB and full information on potential risk factors. The analysis included 12,783 people, including 73 who developed active TB. Those who developed TB were significantly younger (34 vs 37 years, p<0.001), had lower CD4 counts (321 vs 396, p=0.028), were more likely to be Black (30% vs 13%) and were more likely to be female heterosexual (39% vs 28%) than people who did not develop TB.
Risk factor analysis showed that preventive treatment was associated with a 60% reduced risk of developing active TB. A viral load above 1,000 copies/ml increased the risk of developing active TB two-and-a-half times while a CD4 count below 500 was also associated with a raised risk. The risk of active TB was more than three times higher in people with CD4 counts between 350 and 500 and seven-and-a-half times higher in people with CD4 counts below 200.
People of non-White ethnicity were at higher risk of active TB. Black people had almost four times the risk of White people of developing active TB while the risk was 2.8 times higher in people of other non-White ethnicities. A sensitivity analysis by region of origin did not substantially affect the results of the multivariable analysis.
A multiple mediation analysis was used to examine how various factors mediated the effects of time after 1996 on the incidence of active TB. Without intervening factors, the risk would have increased over time, but the multiple mediation analysis showed that increase in CD4 count (-77%), and suppression of viral load were the strongest mediating factors (-49%).
In comparison, improvements in the sensitivity of latent TB testing and the efficacy of TB preventive treatment had miniscule indirect effects (<5%).
Conclusion
The researchers say their findings have two main implications. The first is that early HIV treatment is crucial and is the main driver of reduced TB risk in people with HIV. The median interval between the estimated date of infection and initiation of antiretroviral treatment has been less than one year in the Swiss HIV Cohort since 2010, while the median CD4 count at antiretroviral treatment initiation has remained above 350 since 2012.
The second is that testing for latent TB misses many people at high risk of TB. In this study, almost three out of four people (73%) who developed active TB had a negative result for a test for latent TB, due to the poor sensitivity of the tests in people with impaired immunity.
They also found that over half (56%) of people with a positive test result did not receive preventive treatment, either because it was not offered, or it was declined. Uptake and completion of preventive treatment declined sharply after 2014.
Zeeb M et al. Understanding the decline of incident, active tuberculosis in people with HIV in Switzerland. Clinical Infectious Diseases, 31 May 2023.