Pulmonary restriction predicts long-term pulmonary impairment in people with HIV and tuberculosis

Abstract

Background: While tuberculosis is considered a risk factor for chronic obstructive pulmonary disease, a restrictive pattern of pulmonary impairment may actually be more common among tuberculosis survivors. We aimed to determine the nature of pulmonary impairment before and after treatment among people with HIV and tuberculosis and identify risk factors for long-term impairment.

Methods: In this prospective cohort study conducted in South Africa, we enrolled adults newly diagnosed with HIV and tuberculosis who were initiating antiretroviral therapy and tuberculosis treatment. We measured lung function and symptoms at baseline, 6, and 12 months. We compared participants with and without pulmonary impairment and constructed logistic regression models to identify characteristics associated with pulmonary impairment.

Results: Among 134 participants with a median CD4 count of 110 cells/μl, 112 (83%) completed baseline spirometry at which time 32 (29%) had restriction, 13 (12%) had obstruction, and 9 (7%) had a mixed pattern. Lung function was dynamic over time and 30 (33%) participants had impaired lung function at 12 months. Baseline restriction was associated with greater symptoms and with long-term pulmonary impairment (adjusted odds ratio 5.44, 95% confidence interval 1.16-25.45), while baseline obstruction was not (adjusted odds ratio 1.95, 95% confidence interval 0.28-13.78).

Conclusions: In this cohort of people with HIV and tuberculosis, restriction was the most common, symptomatic, and persistent pattern of pulmonary impairment. These data can help to raise awareness among clinicians about the heterogeneity of post-tuberculosis pulmonary impairment, and highlight the need for further research into mediators of lung injury in this vulnerable population.

Keywords: HIV; Pulmonary disease; Respiratory function tests; Tuberculosis.

Fig. 1

a Baseline COPD assessment test (CAT) scores and b baseline CT radiologic scores according to baseline diagnosis (*p value < 0.05)

Fig. 2

Changes in pulmonary diagnoses at baseline, 6 months and 12 months. For 44 participants with a normal baseline pattern, 35 (80%) remained normal at 12 months, while 6 (14%) developed incident obstruction, 1 (2%) developed restriction, and 2 (5%) developed a mixed pattern. For 10 participants with obstruction at baseline, 7 (70%) had normalized at 12 months, while 3 (30%) had persistent obstruction. For 22 participants with restriction at baseline, 12 (55%) had normalized at 12 months, while 8 (36%) had persistent restriction, 1 (5%) had developed obstruction in addition to restriction (i.e., a mixed pattern), and 1 (5%) had obstruction alone. For 5 participants who had a mixed pattern both at baseline, 2 (40%) had normalized at 12 months, 2 (40%) had persistent restriction, and 1 (20%) had a persistent mixed pattern