Review 2: "Persistent Mycobacterium Tuberculosis Bioaerosol Release in a Tuberculosis-Endemic Setting"

RR:C19 Evidence Scale rating by reviewer:

• Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.

***************************************

Review: This work describes a clinical collection study of participants in represented from either a tuberculosis-endemic area, or from a non-endemic area. Participants in the study were sampled for exhaled breath aerosols for the presence of Mycobacterium tuberculosis. The results of the sampling identified aerosolized M. tuberculosis presence in participant sampling results. The production of the aerosolized M. tuberculosis was not correlative of classical determinants of clinical disease, nor other parameters for that matter, measured in the participants, suggesting that current paradigms of Tb transmission should be reexamined. Authors meted out limitations associated with this study, including low participant number, and mycobacterial staining associated with detection. However, the clear limitations associated with 1) aerosol sampling, and 2) the lack of confirmatory culture reduces the enthusiasm with author suggestions from study results. First, the aerosol sampling rate for personal samples was between 100-300 liters/minute into an improved wetted cyclone into a small volume (15 ml) of medium. The average respiratory minute volume for humans is 3-4 liters/minute, nearly two logs less outflow than that of sampling rate of the sampler used in this study, suggesting that this sampling configuration is concomitantly sampling the ambient air (maybe in the clinic?), which comprises most of the sample collected resulting in the personal sample comprising <1% of the collected sample by volume. How then, can one reliably make conclusions about source of the M. tuberculosis when such a small percentage of the volume of the air sample is so low?  Further, the primary detection method used for M. tuberculosis detection is a stain that identifies the mycobacteria morphologically. Identification of M. tuberculosis by this method, albeit quicker than culture, does not indicate viability of the target organism, nor distinguishes source of the organism in the sample (nor would culture, but in the context of the subject matter of this manuscript, is an important noted weakness of this work). Accordingly, it is exceedingly difficult to resolve the sampling volume/ambient sampling, and the corresponding detection technology as supportive of authorship suggestions and/or conclusionary statements in this study. Lack of correlative measures between M. tuberculosis detection and clinical Tb disease among the participants is symptomatic of  identified weaknesses in approach and unresolved questions surrounding sampling strategy employed in this study.