Mechanisms Leading to Tuberculosis-Related Respiratory Sequelae in Adolescents

Rhode Island Hospital, Providence RI

Investigators

Abstract

PROJECT SUMMARY Each year around the world, approximately 800,000 incident cases of tuberculosis (TB) disease occur in adolescents (people 10-19 years old), the vast majority of whom have pulmonary TB. Despite successful treatment completion, pulmonary TB can lead to long-term respiratory disability, more commonly known as post- tuberculosis lung disease (PTLD). PTLD – defined as abnormal lung function and/or frequent respiratory symptoms – occurs after successful treatment in 50% of individuals who have pulmonary TB as adults (“adult TB survivors”). According to preliminary data from research project leader (RPL) Dr. Silvia Chiang’s study from Lima, Peru, 34% of people who have TB as adolescents (“adolescent TB survivors”) have PTLD 1-2 years after successful treatment. The high adolescent TB incidence, the likely high prevalence of adolescent PTLD, and the many years that adolescent TB survivors may live with respiratory disability add up to a substantial burden of disability-adjusted life years (DALYs) lost to this condition. Nonetheless, the clinical manifestations, mechanisms, and risk factors of PTLD in adolescent TB survivors remain unclear, and they likely differ from those in adult TB survivors due to biological and behavioral differences, such as adolescents’ ongoing lung growth and delayed presentation to care. Unless we fill these knowledge gaps, we will be unable to effectively recognize, treat, and prevent this condition. This proposed research project will elucidate the clinical manifestations, mechanisms, and risk factors of adolescent PTLD. COBRE biostatistics faculty will lead advanced analyses of data from the RPL’s cohort of 101 adolescent TB survivors in Lima. Participants were enrolled in the first month of TB treatment and evaluated up to three years after treatment completion. For each participant, we collected extensive sociodemographic and clinical data, including comorbidities, substance use, anthropometrics, chest x-ray findings, and degree of treatment adherence. Participants underwent two post-treatment lung health evaluations, which consisted of spirometry, oscillometry, fractionated exhaled nitric oxide (FeNO), and a validated 50-item respiratory health questionnaire (including symptom assessment). In Specific Aim 1, we will classify adolescent TB survivors into clinical phenotypes using k-means clustering. A machine learning technique, cluster analysis can identify distinct subgroups within a dataset when the characteristics that separate those subgroups are unknown. In Specific Aim 2, we will elucidate mechanisms of PTLD through path analysis. All published PTLD data are, to date, observational; thus, this approach is innovative because it allows the identification of causal pathways by producing estimates of the direct and indirect effects of variables on PTLD. In Specific Aim 3, we will build a clinical prediction model for PTLD. Given resource constraints in TB-endemic settings, a simple, inexpensive tool is needed to triage adolescents who need continued post-treatment evaluation and monitoring. Successful completion of these specific aims will facilitate the RPL’s transition to research independence by providing preliminary data for her R01 proposal.