Performance of a new "physician-less" automated system of prehospital ST-segment elevation myocardial infarction diagnosis and catheterization laboratory activation.

The door-to-balloon times frequently exceed the recommended delay. We therefore evaluated the performance of a novel "physician-less" cardiac catheterization laboratory (CCL) activation system relying on the automated electrocardiographic diagnosis alone. From January 2010 to 2012, first responders performed electrocardiograms in the field for all patients with a complaint of chest pain or dyspnea. An automated machine diagnosis of "acute myocardial infarction" resulted in immediate CCL activation and direct transfer without human reinterpretation or transmission of the electrocardiogram. Any activation resulting from a nondiagnostic ECG (no ST-segment elevation) was deemed inappropriate and classified as resulting from either human or machine error. Of 155 activations, 136 (88%) were electrocardiographically appropriate. Of these, 128 patients had a final diagnosis of ST-segment elevation myocardial infarction. A door-to-balloon time of <90 minutes was achieved in 99%, the procedural success was high (94%), and the overall mortality was low (3%). Of the electrocardiographically appropriate activations, 8 (5%) were false-positive results. The remaining 19 activations (12%) were inappropriate. Compared with the electrocardiographically appropriate activations, those with inappropriate activations had significantly greater rates of hypertension (p = 0.0070) and known coronary artery disease (p = 0.0008) and higher presenting heart rates (p <0.0001). The causes for inappropriate activation were approximately evenly split between human and machine error. In conclusion, a combination of prehospital automated ST-segment elevation myocardial infarction diagnosis and "physician-less" CCL activation was safe and effective in ensuring target door-to-balloon times in virtually all patients and resulted in an acceptable rate of inappropriate CCL activation.