Back

Diagnostic accuracy of tongue swab testing in persons with sputum Xpert Ultra Trace results

Shapiro, A. E.; Dalmat, R.; Mukwatamundu, J.; Kamoga, C.; Ngwane, M. W.; Steadman, A. E.; Budiawan, E.; Stein, G.; Nalutaaya, A.; Mukiibijames, J.; Nantale, M.; Biche, P.; Visek, C.; Sung, J.; Magcaba, Z.; Ngcobo, N.; Morton, J. F.; Lenn, M. R.; Aucock, S.; Draper, R.; Ganguloo, A.; Wilson, D. P.; Katamba, A.; Kendall, E. A.; Drain, P. K.

2025-07-16 infectious diseases
10.1101/2025.07.15.25331566 medRxiv
Show abstract

BackgroundMolecular amplification of tongue swab samples is a non-sputum-based investigational approach to diagnose pulmonary tuberculosis (pTB). An improved manual qPCR method for tongue swabs recently achieved >90% sensitivity overall in diagnosing TB, compared to a sputum microbiologic reference standard. Performance characteristics in persons with low-positive results on sputum molecular tests are unknown. MethodsAdults in South Africa and Uganda with sputum Xpert MTB/RIF Ultra Trace (TR+) results were recruited for confirmatory evaluation and follow-up. They underwent symptom evaluation, examination, chest X-ray, further sputum testing (repeat Xpert Ultra and two solid and liquid mycobacterial cultures), and two tongue swabs. Tongue swabs were tested using qPCR amplification of the IS6110 gene. A single copy detected on >=1 swab was considered TB-positive. TR+ persons not diagnosed with TB at baseline were re-evaluated at 1 and 3 months. We determined the sensitivity and specificity of tongue swabs against TB culture alone, a microbiologic reference standard (MRS: any positive result from Xpert Ultra or TB culture) and a composite reference standard (CRS: a clinical recommendation for TB treatment or any positive culture) at baseline. Results225 enrolled TR+ participants (115 (51%) women, median age 38 [IQR 30-47], 130 (58%) people living with HIV (PWH)) provided at least 1 tongue swab at baseline. With a culture reference standard, 45 (20%) were positive for TB at baseline testing; 58 (26%) were positive for TB by MRS and 83 (37%) by CRS. Sensitivity and specificity of tongue swabs against culture were 25% [95% CI 13-40%] and 94% [90-97%], vs. MRS were 25% [95% CI 14-38%] and 96% [91-98%], and vs. CRS were 16% [9-26%] and 94% [89-98%]. ConclusionTongue swabs had low sensitivity and moderately high specificity for TB in persons with a Trace Xpert Ultra result. Tongue swabs have limited value for diagnosing people with low-positive molecular test results of uncertain clinical significance.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.