Of the 963 people enrolled, 857 completed the baseline assessment. Interferon-gamma release assays and standardized questionnaires were used to assess latent tuberculosis infection and collect participant characteristics. The baseline prevalence of infection was 6.8%, with a 95% confidence interval of 5.3% to 8.7%. The cohort was conducted in a setting described by the researchers as having a relatively low incidence of tuberculosis and HIV, and its stated aim was to examine a possible approach to tuberculosis screening and prevention in that context. It was an observational cohort study rather than a trial of screening, preventive treatment or another intervention.

The researchers used Firth’s penalized maximum likelihood logistic regression in univariable and multivariable analyses to address potential complete separation in the data. In the multivariable model, CD4 group was the only independent predictor of latent tuberculosis infection. Participants with CD4 counts below 400 cells per microliter had lower odds of infection than those with counts of at least 400 cells per microliter, with an adjusted odds ratio of 0.505, a 95% confidence interval of 0.258 to 0.929 and a P value of 0.027. This model estimate represents an association between baseline CD4 group and infection status; it does not show that changes in CD4 count caused latent tuberculosis infection.

Follow-up provided a separate estimate of assay conversion. Among the 554 participants who were negative by interferon-gamma release assay at baseline, 21 converted to positive, a proportion of 3.8% with a 95% confidence interval of 2.2% to 5.4%. The corresponding annual conversion rate was 1.75 per 100 person-years, with a 95% confidence interval of 1.09 to 2.68. These results describe conversions detected during follow-up rather than the precise time or mechanism of infection. One incident active tuberculosis case was identified, giving an incidence density of 0.084 cases per 100 person-years and a 95% confidence interval of 0.002 to 0.307. The very small number of active cases means the estimate of active-disease incidence is highly uncertain.

The findings provide local measurements of latent tuberculosis infection prevalence, conversion and active disease incidence among people living with HIV, as well as a model-based association between CD4 group and baseline infection status. They do not establish whether a strategy based on latent tuberculosis infection screening would prevent disease, improve clinical outcomes or be cost-effective. Important limitations include the restriction to two districts of one city, the incomplete baseline assessment among the 963 enrolled participants, the 554-person group available for conversion analysis and the occurrence of only one incident active tuberculosis case. The authors therefore described the results as preliminary and hypothesis-generating and said larger multicenter studies, extended follow-up and health-economic evaluations would be needed to determine whether a latent-tuberculosis-infection-guided approach could be viable in this low-burden context.

Subject of Research: Prospective cohort assessment of latent tuberculosis infection among people living with HIV in Shanghai

Article Title: Burden of latent tuberculosis infection among people living with HIV in Shanghai, China: a 4-year cohort study

Article References: Rao, L., Gu, K., Yang, Y., Le, B., Zhang, Z., Chen, J., Ning, Z., Yue, Q., Zhu, Z., & Shen, X. (2026). Burden of latent tuberculosis infection among people living with HIV in Shanghai, China: a 4-year cohort study. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14178-6

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14178-6

Keywords: latent tuberculosis, HIV, cohort study, Shanghai, IGRA, CD4 count, tuberculosis conversion, active tuberculosis, screening, Burden, latent, tuberculosis