Defining the usefulness of oral swabs in tuberculosis diagnosis

Rapid microbiological confirmation of Mycobacterium tuberculosis (M tuberculosis) in sputum with nucleic acid amplification tests (TB-NAAT) is an important step to diagnose tuberculosis and to identify possible drug resistance. These tests significantly increase the diagnostic yield when compared with sputum smear and culture and reduce the time to treatment for people with drug-susceptible or drug-resistant tuberculosis.1WHOManual for selection of molecular WHO recommended rapid diagnostic tests for detection of tuberculosis and drug-resistant tuberculosis. World Health Organization, Geneva2022Google Scholar Obtaining respiratory specimens remains challenging in children younger than 8 years (because they are unable to expectorate and often swallow sputum) and in people who are not coughing or have non-productive cough. Sputum induction, nasopharyngeal aspirate, and gastric aspirate are not complex procedures, but require training, equipment, and appropriate infection prevention and control measures. Diagnostic biomarkers in blood are not fully developed and urine lipoarabinomannan assay is helpful only in selected patients.2Nogueira BMF Krishnan S Barreto-Duarte B et al.Diagnostic biomarkers for active tuberculosis: progress and challenges.EMBO Mol Med. 2022; 14e14088 Crossref Scopus (9) Google Scholar These biomarkers do not provide information on the M tuberculosis drug-resistance pattern, a key advantage of pathogen-based testing. Sampling the upper airway with swabs is an attractive option in settings where obtaining complex and invasive specimens such as gastric aspirate, induced sputum, and pharyngeal aspirates that require significant assistance is avoided. The findings of a systematic review of the sensitivity and specificity of swabs of the buccal mucosa and tongue performed by E Chandler Church and colleagues and published in The Lancet Global Heath is timely as we consider the potential role of these less invasive sampling approaches in the diagnostic algorithm for tuberculosis in adults and children.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar The authors included 15 studies in adults and five in children. The sensitivity of oral swabs in adults ranged from 36% to 91%. As expected, due to the paucibacillary nature of disease, sensitivity was lower in children and ranged from 5% to 42%. The overall specificity was high, between 66% and 100% across all studies in children and adults.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar These findings are similar to the previously published 56·7% (95% CI 44·3–68·2) pooled sensitivity and 91·3% (81·0–96·3) pooled specificity.4Savage HR Rickman HM Burke RM et al.Accuracy of upper respiratory tract samples to diagnose Mycobacterium tuberculosis: a systematic review and meta-analysis.Lancet Microbe. 2023; 4: e811-e821Summary Full Text Full Text PDF Google Scholar The authors highlight knowledge gaps in the preanalytical stages, including the appropriate swab type and site (tongue vs buccal mucosa), storage, and laboratory optimisation techniques needed.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar Ideally, feasibility should be assessed using commercially available automated TB-NAAT. Several studies in this review used in-house tests, possibly contributing to the lower sensitivity and specificity of oral specimens compared with using Xpert MTB/RIF Ultra tests.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar Disease severity, increased bacillary load, and ability to expectorate are important determinants of sensitivity of this sampling strategy when compared with sputum.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar In people with high bacillary load, swabs are likely to perform as well as sputum. Use is less clear when bacillary load is low, such as in people who are unable to expectorate and people presenting for mass screening.5Lima F Santos AS Oliveira RD et al.Oral swab testing by Xpert MTB/RIF Ultra for mass tuberculosis screening in prisons.J Clin Tuberc Other Mycobact Dis. 2020; 19100148 Google Scholar Fewer studies in children were included than studies in adults, and the sensitivity of oral swabs ranged from 5 to 42% in children. The three studies from South Africa enrolled mostly hospitalised children, indicating more severe disease with a higher likelihood of having confirmed tuberculosis than would be seen in children at primary care level.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar The sensitivity in these studies might reflect the expected yield of this sampling strategy when compared with induced sputum and gastric aspirate in hospitalised children. The largest study of 291 South African children reported sensitivity of oral swabs with Xpert MTB/RIF Ultra as 22% (95% CI 15–32) with confirmed tuberculosis (defined as having a positive Xpert MTB/RIF Ultra or culture on a respiratory specimen other than the oral swab specimens) being the reference. The authors did not report on disease severity according to chest x-ray, but 72 (80·0%) of 90 children with confirmed tuberculosis and 119 (75·8%) of 157 children with unconfirmed tuberculosis were in hospital, suggesting severe disease.6Cox H Workman L Bateman L et al.Oral swab specimens tested with Xpert MTB/RIF Ultra Assay for diagnosis of pulmonary tuberculosis in children: a diagnostic accuracy study.Clin Infect Dis. 2022; 75: 2145-2152Crossref PubMed Scopus (9) Google Scholar The reported sensitivity of oral specimens in the review by Church and colleagues is slightly lower than that reported for stool with Xpert MTB/RIF Ultra with pooled sensitivity of 53% (95% CI 35–70) and pooled specificity of 98% (93–99).7WHOPractical manual of processing stool samples for diagnosis of childhood TB. World Health Organization, Geneva2022Google Scholar However, despite the lower sensitivity, the advantage of oral specimens is clear, as testing is not dependent on bowel movement and the health-care worker can collect the sample on the same day. This advantage is important as it prevents failure to collect the specimens as observed when using stool samples. In a study in hospitalised children from six African counties, 1140 (97·5%) of 1169 children had a nasopharyngeal aspirate done but only 942 (80·6%) of 1169 had a stool sample collected during hospitalisation.8Marcy O Wobudeya E Font H et al.Effect of systematic tuberculosis detection on mortality in young children with severe pneumonia in countries with high incidence of tuberculosis: a stepped-wedge cluster-randomised trial.Lancet Infect Dis. 2023; 23: 341-351Summary Full Text Full Text PDF Google Scholar Oral swabs hold promise but the reported sensitivity is below the WHO cutoff for diagnostic tools.1,3–4 Hopefully sensitivity can be improved though optimising specimen collection, including self-collection, and testing methods.3Church EC Steingart KR Cangelosi GA Ruhwald M Kohli M Shapiro AE Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic review.Lancet Glob Health. 2024; 12: e45-e54Summary Full Text Full Text PDF Google Scholar, 9Andama A Whitman GR Crowder R et al.Accuracy of tongue swab testing using Xpert MTB-RIF Ultra for tuberculosis diagnosis.J Clin Microbiol. 2022; 60e0042122 Crossref PubMed Scopus (9) Google Scholar Future studies should include patients with confirmed tuberculosis disease, ranging from non-severe to more severe disease, to measure the sensitivity of oral specimens on the basis of disease severity and therefore place in the diagnostic algorithm. Studies in children should include participants managed in primary care, who are less likely to have severe or confirmed tuberculosis than hospitalised children. Additionally, oral sample yield feasibility and acceptability should be compared with stool samples. HR has received consulting fees from Merck, Sharp & Dohme, and Aid for Aids, payment or honoraria from AbbVie, and has participated on a Data Safety Monitoring Board or Advisory Board for TB Speed and Empirical. MMvDZ has received grants from the EDCTP2 program supported by the European Union (TMA2019SFP-2836 TB lung-FACT2) and the Fogarty International Centre of the National Institutes of Health (K43TW011028). Oral swabs with a rapid molecular diagnostic test for pulmonary tuberculosis in adults and children: a systematic reviewSensitivity varies in both adults and children when diverse methods are used. Variability in sampling location, swab type, and type of NAAT used in accuracy studies limits comparison. Although data are suggestive that high accuracy is achievable using oral swabs with molecular testing, more research is needed to define optimal methods for using oral swabs as a specimen for tuberculosis detection. The current data suggest that tongue swabs and swab types that collect increased biomass might have increased sensitivity. Full-Text PDF Open Access