Background: Deciding which patients should receive thrombolytic therapy or percutaneous transluminal coronary angioplasty (PTCA) for acute myocardial infarction (AMI) can be difficult, especially for less-obvious candidates and when consulting physicians are off site.
Objective: To test whether the electrocardiograph-based Thrombolytic Predictive Instrument (TPI) improves use of thrombolytic and overall reperfusion therapy.
Design: 22-month randomized, controlled, clinical effectiveness trial.
Setting: Emergency departments at 28 urban, suburban, and rural hospitals in the United States.
Patients: Persons presenting to the emergency department with AMI and ST-segment elevation on an electrocardiogram (ECG).
Intervention: TPI predictions automatically printed on ECG text headers.
Measurements: Percentages of patients receiving thrombolytic therapy, thrombolytic therapy within 1 hour of initial ECG, and overall reperfusion (thrombolytic therapy or PTCA).
Results: Of 2875 patients with AMI, 1243 (43.2%) had ST-segment elevation. Of these, 1197 were randomly assigned to study groups; 732 (61.2%) had inferior AMI, and 465 (38.8%) had anterior AMI. A total of 60.5% of controls and 62.1% of TPI patients (P = 0.2) received thrombolytic therapy, 52.5% of controls and 53.3% of TPI patients received thrombolytic therapy within 1 hour (P > 0.2), and 67.6% of controls and 70.3% of TPI patients received overall reperfusion (P = 0.2). Of patients with inferior AMI in the control group versus the TPI group, 61.1% versus 67.6% (P = 0.03) received thrombolytic therapy, 53.2% versus 58.6% (P = 0.08) received thrombolytic therapy within 1 hour, and 67.7% versus 74.7% (P = 0.03) received overall reperfusion. Of patients with anterior AMI in the control group versus the TPI group, 59.5% versus 53.9% (P > 0.2) received thrombolytic therapy, 51.4% versus 45.3% (P > 0.2) received thrombolytic therapy within 1 hour, and 67.6% versus 63.8% (P > 0.2) received overall reperfusion. Among women (n = 398) in the control group versus the TPI group, 48.1% versus 58.2% (P = 0.03) received thrombolytic therapy, 40.5% versus 48.4% (P = 0.10) received thrombolytic therapy within 1 hour, and 55.7% versus 65.7% (P = 0.04) received overall reperfusion. Of patients who required physician consultation by telephone (n = 271) in the control group versus the TPI group, 47.3% versus 63.2% (P = 0.01) received thrombolytic therapy, 41.1% versus 53.6% (P = 0.04) received thrombolytic therapy within 1 hour, and 50.7% versus 66.4% (P = 0.01) received overall reperfusion.
Conclusions: The TPI increased use of thrombolytic therapy, use of thrombolytic therapy within 1 hour, and use of overall coronary reperfusion by 11% to 12% for patients with inferior AMI, 18% to 22% for women, and 30% to 34% for patients with an off-site physician. Although its effect was minimal on patients with high baseline reperfusion rates, the TPI increased use and timeliness of reperfusion in often-missed groups and when involved physicians were off site.