Algorithm for real-time analysis of intracoronary electrocardiogram

INTRODUCTION: Since its first implementation in 1985, intracoronary (ic) electrocardiogram (ECG) has shown ample evidence for its diagnostic value given the higher sensitivity for myocardial ischemia detection in comparison to surface ECG. However, a lack of online systems to quantitatively analyze icECG in real-time prevents its routine use. The present study aimed to develop and validate an autonomous icECG analyzing algorithm. MATERIALS AND METHODS: This is a retrospective observational study in 100 patients with chronic coronary syndrome. From each patient, a non-ischemic as well as ischemic icECG at the end of a 1-min proximal coronary balloon occlusion was available. An ECG expert as well as the newly developed algorithm for autonomous icECG analysis measured the icECG ST-segment shift in mV for each icECG tracing. RESULTS: Intraclass correlation coefficient (ICC) demonstrated low variability between the two methods (ICC = 0.968). Using the time point of icECG recording as allocation reference for absent or present myocardial ischemia, ROC-analysis for ischemia detection by the manually determined icECG ST-segment shift showed an area under the curve (AUC) of 0.968 ± 0.021 (p < 0.0001). AUC for the algorithm analysis was 0.967 ± 0.023 (p < 0.0001; p = 0.925 for the difference between the ROC curve AUCs). Time to complete analysis was below 1,000 ms for the autonomous icECG analysis and above 5 min for manual analysis. CONCLUSION: A newly developed autonomous icECG analysing algorithm detects myocardial ischemia with equal accuracy as manual ST-segment shift assessment. The algorithm provides the technical fundament for an analysing system to quantitatively obtain icECG in real-time.