Tuberculosis Infection in Early Childhood and the Association with HIV-exposure in HIV-uninfected Children in Rural Uganda

Abstract

Background: In high tuberculosis (TB) burden countries, a significant proportion of the latent TB reservoir is established by the age 5 years. There are critical knowledge gaps in our understanding of the age-specific prevalence of TB infection and the influence of HIV exposure on TB infection in the first 5 years of life among HIV-uninfected children in sub-Saharan Africa.

Methods: We measured TB infection with the Quantiferon Gold-in-Tube (QFT) and tuberculin skin tests (TST) in 447 children ≤60 months and their 284 HIV-infected and IV-uninfected mothers in rural Uganda.

Results: The overall prevalence of TB infection in children ≤60 months by TST was 24% (95% confidence intervals [CI]: 19.9-27.9). The prevalence of TST positivity was highest among children in their first year of life (36%; 95% CI: 26.0-45.9) and declined with age to 19% at 36-60 months of age, χ test for trend P = 0.014. In contrast, 4% (95% CI: 1.9-5.87%) of children had a positive QFT, and there was no trend detected with age, P = 0.576. QFT positivity was detected as early as 5 months. HIV-exposed uninfected children had significantly higher odds of TB infection by QFT (odds ratio [OR]: 21.2; P = 0.008; 95% CI: 2.2-204.7) or positive TST or QFT (OR, 2.4; P = 0.020; 95% CI: 1.2-5.1) compared with HIV-unexposed uninfected children, adjusting for age, BCG vaccination and a positive maternal TST or QFT.

Conclusions: An appreciable prevalence of TB infection was detected in early childhood. HIV-exposed uninfected children have a higher risk for TB infection compared with children born to HIV-uninfected mothers.

Figures

Figure 1. Proportion of positive Tuberculin Skin Tests (TST) and Quantiferon Gold In Tube Tests (QFT), stratified by age and mother's HIV status

A. Proportion of positive TST and QFTs among children ≤60 months, stratified by age-category. There is a negative trend in TST positivity with increasing age, chi-square test for trend p=0.014. The trend of QFT positivity and age was not statistically significant, chi-square test for trend p=0.576. B. Proportion of positive QFTs, stratified by a HIV-status of mother. Children had a higher proportion of positive QFTs for all age-categories, though this was not statistically significant for any age-categories. C. Proportion of positive TSTs, stratified by age and mother's HIV-status. Children born to HIV-infected mothers had a higher prevalence of TST positivity in the 24-35 month age-category (p=0.013) and the 36-60 month age group (p=0.001) compared to the children born to HIV-uninfected mothers in the same age-group. A positive TST is defined as induration ≥10mm.

Figure 2. Quantiferon Gold-In-Tube (QFT) and Tuberculin Skin (TST) positivity among children ≤60 months, using different definitions for TST cut offs for positivity

*age-specific cut-off proposed by Chan et al. (positive TST defined as ≥21mm from 0-1 years, ≥18mm from age 2-3 years, and ≥13mm from 4-5 years).