Table 3.
Sensitivity | Specificity | PPV | NPV | |
---|---|---|---|---|
(95%CI) | (95%CI) | (95%CI) | (95%CI) | |
EBV-associated CNS disorder 2 | 100 | NA3 | NA | 100 |
(72.25-100) | (90.11–100) | |||
CMV encephalitis | 100 | NA | NA | 100 |
(74.12–100) | (92.87–100) | |||
PML | 92.86 | 100 | 100 | 98.36 |
(68.53–98.73) | (93.98–100) | (77.19–100) | (91.28–99.71) | |
CNS tuberculosis | 40 | 97.1 | 50 | 95.71 |
(11.76–76.93) | (90.03–99.20) | (15.00–85.00) | (88.14–98.53) | |
Toxoplasma encephalitis | 100 | 100 | 100 | 100 |
(60.97–100) | (94.65–100) | (60.97–100) | (94.65–100) | |
Cryptococcal meningitis | 100 | 100 | 100 | 100 |
(67.56–100) | (94.50–100) | (67.56–100) | (94.50–100) |
CSF, cerebrospinal fluid; mNGS, metagenomic next-generation sequencing; PPV, positive predictive value; NPV, negative predictive value; EBV, Epstein–Barr virus; CMV, cytomegalovirus; PML, progressive multifocal leukoencephalopathy; and CNS, central nervous system. 1For the purpose of illustration, all six patients with undetermined causes of neurological symptoms had been excluded from analysis. Finally, diagnostic performance of CSF mNGS was calculated from 74 samples. 2CNS lymphoma or EBV encephalitis. 3EBV and CMV were commonly detected by CSF mNGS. Some of the positive detections were not associated with CNS disorder, and also could not be considered as false positive either. Therefore, the specificity and PPV were not calculated.