Abstract P107: Automated Detection of the Culprit Artery Associated With ST Elevation Myocardial Infarct Using a New Electrocardiogram Diagnostic Algorithm
Introduction: The ultimate goal of ST elevation myocardial infarction (STEMI) management is to open the occluded artery and restore blood flow. Recent AHA recommendations suggest that automated diagnostic algorithms should identify the occluded artery to help guide prehospital and hospital therapeutic interventions for STEMI.
Purpose: To study the performance of a new ECG diagnostic algorithm DXL (Philips Healthcare) in identifying the infarct-related-culprit artery in STEMI.
Methods: The DXL algorithm was evaluated retrospectively using a database of digitized admission electronic ECG’s linked to outcome and angiographic findings. The DXL algorithm was developed on a separate ECG database using criteria recommended in the AHA guidelines. The source population (n=711) was all suspected ACS patients presenting consecutively to the ED of a single hospital over a three year period with the discharge diagnosis of acute MI and an angiogram confirmed flow limiting lesion. After excluding subjects with NSTEMI and ECG confounders for STEMI (paced rhythm, LBBB, LVH, wide complex tachycardia), the final study cohort included 188 subjects who met the new AHA limits for STEMI (V2–3: female ≥0.15 mV; male ≥0.25 mV & <40 years old, ≥0.2 mV & ≥40 years old) with an angiogram recorded infarct-related culprit artery (LAD, RCA, LCx). Culprit artery identified by angiogram served as the gold standard. The DXL algorithm performance was measured using sensitivity, specificity, positive and negative likelihood ratios along with their 95% confidence intervals (CI).
Results: Artery specific performance was as follows:
Conclusion: Computer based culprit artery identification using the DXL algorithm is accurate and provides a new risk stratification tool to support the management of STEMI patients in the prehospital and hospital setting. Recognition of culprit artery could potentially further reduce discovery-to-perfusion time