Unstable post-infarction ischemia : identification and risk stratification with special emphasis on noninvasive methods

Abstract

Although in-hospital mortality after acute myocardial infarction has decreased, long-term mortality is still high. This study tested the prognostic value of detecting residual myocardial ischemia by different techniques in 74 patients recovering from an acute myocardial infarction. The prevalence and prognostic significance of ST-segment depression detected by ambulatory electrocardiography was studied. The prognosis in 22 (30%) patients with ST-segment depression was worse than in 52 (70%) without, regarding short-term cardiac events (defined as death or reinfarction or revascularization) (P<0.01), long-term mortality (P=0.01) and long-term cardiac events (P<0.001). Compared with clinical variables in a multivariate regression analysis, ST-segment depression was the strongest covariate assessing prognosis. Due to the high prevalence of prolonged or fixed ST-segment depression during ambulatory electrocardiography, new methods to detect ST-segment changes were assessed. ST-segment changes were analyzed from four reference levels. ST-segment elevation was measured 0 to 5 ms after the J point. The presence of ST-segment elevation measured from a 24-hour median level was statistically associated with mortality (P=0.03). The sensitivity, specificity and accuracy of ST-depression/ST-elevation measured from the 24-hour median level in predicting mortality were 78%, 71% and 73%, respectively. These values are higher than those reported for ST-segment depression detected during exercise testing.

The prognostic value of ST-segment depression detected by ambulatory electrocardiography was compared with the outcome of exercise testing. Both tests were able to predict death, death or nonfatal infarction and cardiac events. The sensitivity of both methods to assess death and death or nonfatal infarction was similar. The specificity of ambulatory electrocardiography was superior in predicting death (P=0.01) and death or nonfatal infarction (P=0.001). When both techniques were combined, the studied population could be classified in groups with low, medium and high risks. ST-segment depression detected by ambulatory electrocardiography was a stronger covariatein predicting mortality than exercise-induced ST-segment depression. The prognostic value of predischarge stress echocardiography was studied. The method was compared with exercise testing and clinical variables. A positive stress echocardiography was associated with a poor prognosis (mortality:P=0.0002; cardiac events: P<0.0001). Even revascularization procedures, subsequently decided upon the results of exercise testing and clinical symptoms, were predicted by the initial stress echocardiogram (P=0.02). A new-onset wall motion abnormality during stress echocardiography was the strongest variable predicting death and death or nonfatal infarction. The cardiac sympatho-vagal regulation was studied immediately after the acute phase of AMI. Heart rate variability was evaluated in the time and frequency domains. Heart rate variability was compared in survivors, nonsurvivors and a group of 24 healthy controls. Heart rate variability was higher in survivors than in nonsurvivors (P=0.005), higher in controls than in survivors (P=0.05), higher in controls than in nonsurvivors (P=0.0001) and higher inpatients without cardiac events and reinfarction (P=0.03 and P=0.03, respectively). In a multivariate regression analysis, including clinical variables and left ventricular ejection fraction in the model, heart rate variability retained its independent and additive prognostic value. The prognostic value of several clinical variables and variables derived from diverse noninvasive methods was investigated in a two-stage multivariate stepwise regression analysis. The following covariates were tested: clinical variables, ST-segment changes detected on ambulatory electrocardiography and exercise testing, heart rate variability, ejection fraction, wall motion score index, worsened or new-onset wall motion abnormality on stress echocardiography and different physiological variables obtained during exercise testing. At the second stage of this multivariate analysis only new-onset wall motion abnormality, ST-segment depression on ambulatory electrocardiography and decreased heart rate variability had additive and independent value to predict mortality. In conclusion, residual myocardial ischemia detected by different techniques was the strongest factor predicting long-term mortality and should be studied in all patients recovering from an acute myocardial infarction.