Diagnostic accuracy of nanopore sequencing using respiratory specimens in the diagnosis of pulmonary tuberculosis.

OBJECTIVES

The aim of this study was to assess the role of nanopore sequencing using respiratory specimens in the early diagnosis of pulmonary tuberculosis (PTB) and simultaneously compare it head-to-head with Mycobacterium tuberculosis (MTB) culture, and Xpert MTB/rifampin (RIF).

METHODS

The clinical data of 164 patients with suspected PTB were retrospectively reviewed to determine the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) of the acid-fast bacilli (AFB) smear, MTB culture, Xpert MTB/RIF, and nanopore sequencing and assess their diagnostic accuracy compared with culture combined with clinical diagnosis.

RESULTS

The overall sensitivity, specificity, PPV, NPV, and AUC of the AFB smear were 27.6%, 87.5%, 84.2%, 33.3%, and 0.58, respectively; for MTB culture, these values were 57.8%, 100.0%, 100.0%, 49.5%, and 0.79, respectively; for Xpert MTB/RIF, these values were 62.9%, 97.9%, 98.7%, 52.2%, and 0.80, respectively; and for nanopore sequencing, these values were 94.8%, 97.9%, 99.1%, 88.7%, and 0.96, respectively.

CONCLUSION

The diagnostic accuracy of nanopore sequencing was excellent in terms of PTB diagnosis and was considerably better than that of the Xpert MTB/RIF and MTB culture. Nanopore sequencing could be an effective alternative to Xpert MTB/RIF for the initial detection of PTB to improve the accuracy of PTB diagnosis.