Diagnostic accuracy of Xpert MTB/RIF Ultra and culture assays to detect Mycobacterium Tuberculosis using OMNIgene-sputum processed stool among adult TB presumptive patients in Uganda

Abdulwahab Sessolo*, Emmanuel Musisi, Sylvia Kaswabuli, Josephine Zawedde, Patrick Byanyima, Wilber Sabiiti, Stanley Walimbwa, Joseph Ola, Ingvar Sanyu, Rejani Lalitha, Moses Kamya, Lucian Davis, William Worodria, Laurence Huang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Downloads (Pure)

Abstract

Background
Stool is a potential sample for diagnosing Mycobacterium tuberculosis (Mtb) in patients with difficulty in expectorating. However, high mycobacterial culture contamination rates and Xpert MTB/RIF Ultra test error rates on stool samples have limited its use. OMNIgene SPUTUM (OM-S) is a sample transport reagent with characteristics of sputum decontamination while maintaining viable Mtb. We evaluated the impact of OM-S on Mtb diagnostic yield from stool using smear microscopy, Xpert MTB/RIF Ultra, and culture among presumptive TB patients.

Methods
Paired stool and expectorated sputum samples were collected from consecutive Ugandan adults undergoing diagnostic evaluation for pulmonary TB between June 2018 and June 2019. Stool was divided into 2 portions: one was homogenized in OM-S (OM-S stool) and the other in PBS (PBS stool) as control. Both sputum and processed stool were tested for Mtb using concentrated smear fluorescence microscopy (CFM), Xpert MTB/RIF Ultra (Xpert) and Mycobacteria Growth Indicator Tube (MGIT) culture. Sensitivity, specificity, and predictive values for each test were calculated against sputum MGIT culture as the reference standard.

Results
Of the 200 participants, 120 (60%) were male, 73 (37%) were HIV positive (median CD4 120 cells/uL (IQR 43–297)) and 128 (64%) had confirmed pulmonary TB by sputum MGIT culture. Seven (4%) OM-S stool Xpert samples reported errors while 47 (25%) and 103 (61%) were contaminated on OM-S stool MGIT and PBS stool MGIT, respectively. OM-S stool MGIT was able to accurately diagnose 56 of the contaminated PBS stool MGIT samples compared to only 5 of the contaminated OM-S stool MGIT samples diagnosed by PBS stool MGIT. Sensitivity (95% Confidence Interval, CI) 89% (83–94) for OM-S stool Xpert was higher compared to that of OM-S stool MGIT 60% (51–69) and PBS stool MGIT 42% (32–52). Specificity (95%CI) 91% (82–97) was also higher for OM-S stool Xpert compared to OM-S stool MGIT 64% (51–75) and PBS stool MGIT 26% (16–38).

Conclusion
Stool processed with OM-S showed potential to improve Mtb diagnostic yield and reduce rates of indeterminate results when tested on Xpert and MGIT culture. The method may thus be of value in Mtb detection among patients with difficulty to expectorate.
Original languageEnglish
Article numbere0284041
Number of pages14
JournalPLoS One
Volume18
Issue number4
DOIs
Publication statusPublished - 21 Apr 2023

Fingerprint

Dive into the research topics of 'Diagnostic accuracy of Xpert MTB/RIF Ultra and culture assays to detect Mycobacterium Tuberculosis using OMNIgene-sputum processed stool among adult TB presumptive patients in Uganda'. Together they form a unique fingerprint.

Cite this