Assessment and significance of ST-segment changes detected by ambulatory electrocardiography after acute myocardial infarction.

This study assessed the prognostic value of ST-segment changes detected by ambulatory electrocardiographic monitoring during the early in-hospital period after acute myocardial infarction. New methods for defining the ST-segment reference level and for measuring ST-segment elevation were used. ST-segment depression was defined as a change in ST level by > or = 0.1 mV 80 ms after the J point, elapsing > or = 1 minute. ST-segment elevation was defined as a deviation by > or = 0.15 mV, elapsing > or = 1 minute, and measured at the J point. An interval of > or = 2 minutes was required before another discrete episode was counted. Four ST-segment reference levels were automatically calculated: (1) "isoelectric," (2) "nearest to normal," (3) "24-hour median," and (4) "first-hour median." During a mean follow-up period of 3 years (mean 36 +/- 15 months), 47 cardiac events occurred in 38 patients: 18 deaths, 9 nonfatal reinfarctions, and 20 revascularization procedures. More deaths occurred in patients with than without ST elevation-24-hour median (22% vs 5%, p = 0.03), and in patients with than without ST depression-isoelectric (61% vs 32%, p = 0.02), and in patients with than without ST-depression-24-hour median (61% vs 23%, p = 0.003). "All cardiac events" (deaths, infarctions, or revascularization procedures) occurred more often in patients with than without ST depression-isoelectric (55% vs 22%, p = 0.003), and in patients with than without ST-depression-24-hour median (47% vs 17%, p = 0.004). Sensitivity, specificity, and accuracy of ST depression/elevation-24-hour median to assess mortality were 78%, 71%, and 73%, respectively.