BackgroundThe size and concentration of exhaled bio-aerosols can affect risk of tuberculosis transmission. Understanding bio-aerosol production might provide a way to assess heterogeneity in infectivity and mathematical models, central to informing control practice and policy. This study aimed to measure bio-aerosol production in patients with tuberculosis and assess variability in bio-aerosol production during normal breathing. MethodsBio-aerosol measurements (particles >0·3 μm diameter) were taken during 15 tidal breaths from four adult groups (age >16 years): healthy and uninfected volunteers, patients infected with Mycobacterium tuberculosis, patients with extrathoracic tuberculosis, and patients with intrathoracic tuberculosis. Repeat measurements were obtained before treatment and every 4–8 weeks during treatment. Because M tuberculosis bacilli are 0·2–0·5 μm in width and 2–4 μm in length, we counted droplets of size 1–5 μm, which would plausibly serve as vehicles for M tuberculosis transmission. Our definition of high particle production was a 1–5 μm count above the median among all participants. Stability of counts during treatment was assessed with Lin's concordance correlation coefficients. Logistic regression was used to assess risk factors for high pretreatment production. Ethics approval was given by the National Research Ethics Service Committee—City & East (reference number 11/LO/1601). All participants provided written informed consent. FindingsData from 188 participants were obtained at pretreatment (baseline): bio-aerosol counts (1–5 μm) varied substantially between individuals (0–157·8 counts/L). Although counts were stable within individuals over time, high initial counts declined during treatment. Multivariable analysis showed that intrathoracic tuberculosis was associated with a significant increase in the odds of high 1–5 μm bio-aerosol production (adjusted odds ratio 3·5, 95% CI 1·6–7·8; p=0·002) compared with healthy and uninfected individuals. Participants with extrathoracic tuberculosis and those with M tuberculosis infection had increased odds of high bio-aerosol production, although this was not statistically significant. InterpretationWe provide the first evidence, to our knowledge, that intrathoracic tuberculosis increases bio-aerosol particle production in a particle size range that could plausibly carry M tuberculosis. Although cough-generated bio-aerosols are often assumed to mediate tuberculosis transmission, we found increased bio-aerosol production during normal breathing in patients with intrathoracic tuberculosis. The substantial variation in production among patients with tuberculosis might relate to the degree of infectivity. Thus, presence or absence of cough and sputum smear status might be inadequate as measures of infectiousness and transmission risk. FundingClinical Research and Development Committee (University College London Hospital [UCLH] Charities).
Read full abstract