Improved method for collection of sputum for tuberculosis testing to ensure adequate sample volumes for molecular diagnostic testing

The quality and quantity of sputum collected has an important impact on the laboratory diagnosis of pulmonary TB. We conducted a pilot study to assess a new collection cups for the collection of sputum for the diagnosis of pulmonary tuberculosis. The pilot study utilized the standard collection cup in South Africa demonstrating a mean collection volume of 2.86 ± 2.36SD ml for 198 samples; 19% of the specimens contained < 1 ml and 12% contained > 5 ml. We designed and tested two novel sputum cups with a narrow bottom section and clear minimum and maximum markings to allow patients and clinicians to know whether sufficient sputum volume has been produced. The cups differed in their shape and manufacturing approach. The two options also support different mixing approaches being considered for a highly sensitive companion TB-screening assay being developed at Northwestern University (XtracTB assay). Sputum was collected from 102 patients at Nolungile Youth Centre, Khayelitsha, Cape Town, South Africa for a total of 204 samples. The mean volumes collected from the two cups were 2.70 ± 0.88SD ml and 2.88 ± 0.89SD ml. While the mean volumes of current and novel cups are similar, the volume ranges collected with the novel cups were narrower, and 98% of the specimen volumes were within the target range. Only 4 samples contained > 5 ml, but none were > 6 ml, and none of the specimens contained < 1 ml. The number of coughs that produced the samples, patient HIV and TB status plus qualitative descriptions of the sputum specimens were also evaluated.