Abstract
PURPOSE: To evaluate the accuracy of metagenomic next-generation sequencing (mNGS) for rapid diagnosis of spinal tuberculosis using spinal tissue specimens. METHODS: Medical data regarding suspected spinal tuberculosis were retrospectively analyzed. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) of mNGS were calculated to determine its diagnostic efficacy, and these values were compared with culture and composite reference standard (CRS). RESULTS: Two hundred and three cases of spinal tuberculosis were included for analysis. The sensitivity, specificity, PPV, NPV, and AUC of mNGS compared with culture were all very good. When CRS was used for the comparison, the sensitivity of mNGS and culture were 71.2% and 73.0%; the specificity and PPV were 100% in all cases; the NPV were 74.2% and 75.4%; the AUCs were all 0.86. The sensitivity and NPV of culture were slightly higher than that of mNGS; however, the diagnostic efficacy of mNGS and culture was consistent (P > 0.05). CONCLUSION: Spinal tissue specimens for mNGS testing had very good accuracy for diagnosing spinal tuberculosis.