Skip to main content
. 2012 Mar 20;184(5):E260–E268. doi: 10.1503/cmaj.110773

Table 2:

Performance of high-sensitivity troponin T assay in diagnosing myocardial infarction at presentation and two hours later

Cut-off value of assay; performance measure Time of assay; % (95% CI)
At presentation 2 h after presentation
≥ 14 ng/L (99th percentile)*
 Sensitivity 88.3 (83.3–91.8) 92.2 (88.1–85.0)
 Specificity 81.7 (80.5–82.7) 79.7 (78.6–80.5)
 Positive-predictive value 57.5 (54.5–59.7) 55.9 (53.4–57.6)
 Negative-predictive value 96.2 (94.7–97.3) 97.3 (95.9–98.3)
 Accuracy 83.2 (81.2–84.7) 82.4 (80.6–83.7)
≥ 5 ng/L (limit of detection)
 Sensitivity 93.7 (89.4–96.4) 95.6 (91.7–97.8)
 Specificity 58.4 (57.3–59.2) 53.7 (52.6–54.3)
 Positive-predictive value 38.6 (36.9–39.8) 36.6 (35.1–37.4)
 Negative-predictive value 97.1 (95.1–98.3) 97.8 (95.8–98.9)
 Accuracy 66.1 (64.3–67.3) 62.8 (61.1–63.8)
≥ 3 ng/L (limit of blank)
 Sensitivity 95.6 (91.8–97.8) 98.1 (94.9–99.4)
 Specificity 47.8 (46.7–48.4) 42.2 (41.3–42.6)
 Positive-predictive value 33.9 (32.5–34.6) 32.3 (31.2–32.7)
 Negative-predictive value 97.5 (95.3–98.8) 98.7 (96.6–99.6)
 Accuracy 58.3 (56.6–97.8) 45.5 (45.0–46.9)

Note: CI = confidence interval.

*

At or above the 99th percentile of the general population.

Limit of detection = the lowest quantity or concentration of an analyte that can be reliably detected with a given analytical method.

Limit of blank = the highest apparent concentration of an analyte expected to be found in a sample containing no analyte.