Quantification of viable bacterial load in artificial sputum spiked with Mycobacterium tuberculosis

Abstract Objective Artificial sputum spiked with Mycobacterium tuberculosis could serve for validation of procedures that determine viable mycobacterial load. Design Artificial sputum specimens prepared in-house were spiked with low, medium or high concentrations of Mycobacterium tuberculosis H37Rv stock solution. In a first series, a single technologist processed two batches of specimens daily with high load that were stored refrigerated or at room temperature for up to 8 days. In a second series, nine different technologists processed freshly made batches of specimens with low, medium or high loads. We recorded time to positivity (TTP) in duplicate liquid cultures made from each specimen. Results Specimens were well grouped around the mean TTP (hours; standard deviation) of low: 271.7 (25.9), medium: 233.5 (16.3), and two batches of high load: 186.9 (12.3) and 191.8 (9.0), respectively. A variance component model that included load, storage temperature, days of storage until processing, batch of specimens made, sample ID and technologist ID as random effects in a linear mixed-effects model identified only load, technologist and residual as significant contributors to overall TTP variance. Conclusion Artificial sputum specimens with reproducible and stable viable mycobacterial loads can be made that could serve for training and validation purposes.