Background: In evolving myocardial infarction, assessment of the sum of early resolution of ST-segment elevation (sumSTR) has become an established method to predict outcome. We have found previously that mortality is predicted more accurately by the existing ST-segment deviation in the single electrocardiograph (ECG) lead with maximum deviation (maxSTE) 90 min after start of thrombolysis. This report compares the power to predict medium-term mortality by these two approaches.
Methods: An ST-segment resolution substudy was done in conjunction with the Intravenous nPA for Treatment of Infarcting Myocardium Early (InTIME) II Study, which compared mortality in patients with acute myocardial infarction randomly assigned lanoteplase or alteplase. In 2719 patients, a 12-lead ECG was assessed at baseline and 90 min after the start of thrombolytic therapy.
Findings: MaxSTE achieved a better combination of sensitivities and specificities for mortality prediction than sumSTR. The area under the receiver-operating characteristic curves for 180-day mortality prediction was 0.680 for maxSTE and 0.622 for sumSTR (difference 0.058; 95% CI 0.027-0.088). Risk groups categorised at low, medium, or high risk by maxSTE comprised 43%, 32%, and 24% of patients and those by complete, partial, or no sumSTR comprised 40%, 36%, and 24% of all patients. The 180-day mortality rates for the three maxSTE risk groups were 3.1%, 7.1%, and 16.2%, and those for the sumSTR groups were 4.8%, 8.1%, and 11.7%. The 12-month Kaplan-Meier estimates were 4.1%, 8.8%, and 18.6%, and 5.9%, 9.9%, and 13.7%, respectively.
Interpretation: MaxSTE predicts early and medium-term mortality more accurately than does sumSTR. The prognosis for an individual patient can be accurately estimated simply by the ST-segment deviation present in one ECG lead recorded 90 min after thrombolysis.