Table 3.
Performance of a modified ESC algorithm and a historical-hs-cTnT algorithm with the use of a historical hs-cTnT value as the 0-hour value for the prediction of all-cause mortality
| No of eligible patients | 8432 (100) | |
| Death ≤30 days after the index visit, n (%) | 13 (0.2) | |
|
Algorithm using hs-cTnT measured at the same visit (modified ESC algorithm) | ||
| Total no of patients ruled-out, n (%) | TN (%) | FN (%) |
| 8100 (96) | 8088 (99.9) | 12 (0.1) |
| Total no of patients not ruled-out, n (%) | TP (%) | FP (%) |
| 332 (3.9) | 1 (0.3) | 331 (99.7) |
| Rule-out | ||
| No of events, (%) | 12 (0.1) | |
| 30-day risk of all-cause mortality (95% CI) | 0.1% (0.1% to 0.3%) | |
| NPV, % (95% CI) | 99.9 (99.7 to 99.9) | |
| LR− (95% CI) | 0.96 (0.28 to 1.04) | |
| Sensitivity, % (95% CI) | 7.7 (0.4 to 37.9) | |
|
Algorithm using a historical hs-cTnT value as the 0-hour hs-cTnT value (historical-hs-cTnT algorithm) | ||
| Total no of patients ruled-out, n (%) | TN (%) | FN (%) |
| 6700 (79) | 6696 (99.9) | 4 (0.1) |
| Total no. of patients not ruled-out, n (%) | TP (%) | FP (%) |
| 1732 (20) | 9 (0.5) | 1723 (99.5) |
| Rule-out | ||
| No of events, (%) | 4 (0.1) | |
| 30-day risk of all-cause mortality (95% CI) | 0.1% (0.0% to 0.2%) | |
| NPV, % (95% CI) | 99.9 (99.8 to 100) | |
| LR− (95% CI) | 0.39 (0.04 to 1.01) | |
| Sensitivity, % (95% CI) | 69.2 (38.9 to 89.6) | |
ESC, European Society of Cardiology; FN, false negative; FP, false positive; hs-cTN, high-sensitivity cardiac troponin; LR−, negative likelihood ratio; NPV, negative predictive value; TN, true negative; TP, true positive.