SAGE Open Medicine (Aug 2022)

Accuracy of GenoQuick MTB test in detection of in sputum from TB presumptive patients in Uganda

  • Sylvia Kaswabuli,
  • Emmanuel Musisi,
  • Patrick Byanyima,
  • Abdul Sessolo,
  • Ingvar Sanyu,
  • Josephine Zawedde,
  • William Worodria,
  • Laurence Huang,
  • Alfred Okeng,
  • Freddie Bwanga

DOI
https://doi.org/10.1177/20503121221116861
Journal volume & issue
Vol. 10

Abstract

Read online

Objective: The objective of the study was to determine the diagnostic performance of the GenoQuick MTB test on heated sputum against the conventional Lowenstein–Jensen Mycobacterium tuberculosis culture as the reference method for tuberculosis diagnosis. Introduction: Fast, reliable, and easy-to-use tests for tuberculosis diagnosis are essential to achieving the Sustainable Development Goal of diagnosing and treating 90% of tuberculosis patients by 2030. We evaluated the diagnostic performance of the GenoQuick MTB, a polymerase chain reaction–lateral flow test, in Uganda, a resource-constrained, high tuberculosis- and HIV-burden setting. Methods: Fresh sputum samples from presumptive tuberculosis patients at Mulago Hospital were tested for M. tuberculosis using smear microscopy, GenoQuick MTB test, and Lowenstein–Jensen culture. For the GenoQuick MTB test, mycobacterial DNA was extracted by heating sputum at 95°C for 30 min while DNA amplification and detection were done following the manufacturer’s protocol (Hain Lifescience, Nehren, Germany). Sensitivity, specificity, and kappa agreements were calculated against Lowenstein–Jensen M. tuberculosis culture as a reference test using STATA V12. Results: Of the 86 tested samples, 30.2% had culture-confirmed pulmonary tuberculosis. Overall, sensitivity was higher for GenoQuick MTB (81%, 95% confidence interval: 60%−93%) than for smear microscopy (69%, 95% confidence interval: 48%−86%). Among people living with HIV, sensitivity was identical for GenoQuick MTB and smear tests (75%, 95% confidence interval: 42%−95%). Contrastingly, smear had a higher overall specificity (98%, 95% confidence interval: 91%−100%) than for GenoQuick MTB (92%, 95% confidence interval: 81%−97%). A similar trend of specificity was observed among the people living with HIV for smear microscopy (100%, 95% CI: 87%−100%) and for GenoQuick MTB (96%, 95% confidence interval: 81%−100%). Conclusion: The GenoQuick MTB test could be a potential tuberculosis diagnostic test given its higher sensitivity. Evaluation of this test in larger studies is recommended.