Long-term mortality after tuberculosis treatment among persons living with HIV in Haiti

Yvetot Joseph, Zhiwen Yao, Akanksha Dua, Patrice Severe, Sean E. Collins, Heejung Bang, Marc Antoine Jean-Juste, Oksana Ocheretina, Alexandra Apollon, Margaret L. McNairy, Kathryn Dupnik, Etienne Cremieux, Anthony Byrne, Jean W. Pape, Serena P. Koenig

Research output: Contribution to journalArticlepeer-review


Introduction: Long-term mortality among TB survivors appears to be higher than control populations without TB in many settings. However, data are limited among persons with HIV (PWH). We assessed the association between cured TB and long-term mortality among persons with PWH in Haiti. Methods: A prospective cohort of PWH from the CIPRA HT-001 trial was followed from study enrolment (August 2005 to July 2008) to study closure (December 2018) to compare mortality between participants with and without TB. The index date for the survival analysis was defined as 240 days after TB diagnosis or randomization date. Time to death was described using Kaplan–Meier curves, and log-rank tests were used to compare time to death between the TB and no-TB cohorts. The association between TB and long-term mortality was estimated with multivariable Cox models. Results: Of the 816 participants in the CIPRA HT-001 trial, 77 were excluded for a history of TB prior to study enrolment and 31 were excluded due to death or attrition prior to the index date, leaving 574 in the no-TB and 134 in the TB cohort. Twenty-four (17.9%) participants in the TB and 48 (8.4%) in the no-TB cohort died during follow-up. Five and 10-year mortality rates were 14.2% and 17.9% respectively, in the TB cohort, and 6.1% and 8.4% in the no-TB cohort. In Kaplan–Meier analysis, participants in the TB cohort had a significantly shorter time to death (log-rank p < 0.001). In multivariable analysis, TB treatment was the only predictor of mortality (HR: 2.78; 95% CI: 1.61, 4.79). Sensitivity analyses, which included only baseline TB cases, an index date of two years after TB diagnosis, and study enrolment and case-control matching yielded results that were consistent with primary analyses. Conclusions: PWH who are successfully treated for TB have higher long-term mortality than those who are never diagnosed with TB, even after accounting for acute TB-related mortality. A better understanding of the underlying mechanisms associated with TB sequelae is critically needed to guide specific interventions. Until then, more aggressive measures for health promotion and disease prevention are essential to improve long-term survival for PWH after TB treatment.

Original languageEnglish (US)
Article numbere25721
JournalJournal of the International AIDS Society
Issue number7
StatePublished - Jul 2021
Externally publishedYes


  • AIDS
  • HIV
  • long-term mortality
  • opportunistic infections
  • tuberculosis

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health
  • Infectious Diseases


Dive into the research topics of 'Long-term mortality after tuberculosis treatment among persons living with HIV in Haiti'. Together they form a unique fingerprint.

Cite this