Table 2.
Performance of WGS for the prediction of susceptibility to first-line drugs among Swedish consecutive routine M. tuberculosis isolates 2016–2022.
| Phenotypically Resistant |
Phenotypically Sensitive |
|||||||
|---|---|---|---|---|---|---|---|---|
| Drug (Number of Isolates Tested) | Resistant WGS Prediction | Sensitive WGS Prediction | Resistant WGS Prediction | Sensitive WGS Prediction | Sensitivity (%) | Specificity (%) | PPV * (%) | NPV * (%) |
| Isoniazid (n = 2268) | 259 | 10a | 1 | 1998 | 96.28 | 99.95 | 99.62 | 99.50 |
| Rifampicin (n = 2275) | 92 | 1 | 8b | 2174 | 98.92 | 99.63 | 92.00 | 99.95 |
| Ethambutol (n = 2265) | 37 | 3 | 25 | 2200 | 92.50 | 98.88 | 59.68 | 99.86 |
| Pyrazinamide (n = 2264) | 99 | 8c | 2 | 2155 | 92.52 | 99.91 | 98.02 | 99.63 |
a = Two isolates with rare katG mutations, one isolate with a deletion in ndh. b = These isolates all harbor a so-called “disputed” rpoB mutation. c = Seven lineage 1 isolates, one lineage 4 isolate with a panD mutation. * PPV = positive predictive value, * NPV = negative predictive value.