Relative value of clinical variables, treadmill stress testing, and holter ST monitoring for postinfarction risk stratification

Abstract The aim of this study was to compare the roles of clinical assessment, treadmill stress testing, and Holter ST analysis for postinfarction risk stratification in patients treated with thrombolysis. The study group consisted of 256 consecutive patients, all of whom underwent Holter ST monitoring early (mean 83 hours, range 48 to 180) after admission. Of these, 12 were excluded from the analysis either because Holter recordings were of insufficient quality (n = 6), or because reinfarction occurred within 24 hours of monitoring (n = 6). In the remaining 244 patients, 43 sustained a recurrent event (death, reinfarction, unstable angina) over the 8-month (range 3 to 12) follow-up period, and an additional 14 patients required revascularization. At multivariate event-free survival analysis, Killip class ≥2 and Holter ST shift were independently predictive of the outcome. The strongest predictor was Holter ST shift at a cumulative duration of >60 minutes. Of the 232 patients eligible for stress testing (12 sustained an event between Holter monitoring and the scheduled stress test), 196 were able to perform the test. The variable “inability to perform a stress test” was not independently predictive of outcome and did not influence the multivariate analysis. When clinical, Hotter, and stress test variables were taken into account in patients who performed a stress test, Killip class was the only independent predictor of outcome (event-free survival). When revascularization was included as an end point, Holter ST shift was the only independent predictor of outcome. In conclusion, a significant proportion of recurrent events after thrombolysis occurs very early, before stress testing can be performed. The only independent predictors of outcome identified in this study were Holter ST shift and Killip class. Stress testing could only be performed in selected patients and added no additional prognostic information.