Ischemia Monitoring by Analysis of Depolarization Changes G Amit 1 , LR Davrath 1 , S Abboud 3 , H Hod 2 , E Toledo 1 , S Matetzky 2 1 Research & Development, BSP Ltd, Tel-Aviv, Israel 2 Heart Institute, Chaim Sheba Medical Center, Tel-Hashomer, Israel 3 Department of Biomedical Engineering, Tel-Aviv University, Israel Abstract Myocardial ischemia causes changes in the depolarization phase of the cardiac cycle, which can be quantified by analysis of high-frequency QRS components (HFQRS). We introduce a novel HFQRS analysis technology and evaluate its performance in monitoring transient ischemic episodes in patients hospitalized due to chest pain. Continuous monitoring by high-resolution 12-lead ECG was performed in 43 patients admitted to a chest pain unit, followed by cardiac imaging. Indices of HFQRS based on ischemia-specific morphological changes and conventional ST segment levels were extracted from signal-averaged ECG. HFQRS indices were positive for 5 of 10 patients that were diagnosed with coronary artery disease, while ST analysis was negative for all patients. The severity of the HFQRS indices was directly related to the likelihood of ischemic events. HFQRS is a promising technology for monitoring and early detection of acute coronary syndrome. 1. Introduction Chest pain is one of the leading reasons for hospital emergency department (ED) visits worldwide. In the US, approximately 6 million people annually undergo evaluation in the ED for acute chest pain. Despite the wealth of knowledge available about acute coronary syndrome (ACS), this condition continues to be among the most difficult to predict or diagnose. Nearly half of patients hospitalized for unstable angina eventually receive a non-cardiac-related diagnosis. Nonetheless, 5% of patients with myocardial infarction (MI) are inappropriately discharged from the ED [1]. Although ECG is a mainstay in the diagnosis of ACS, the initial 12-lead ECG in the ED is often non-diagnostic in ACS patients, especially in non-ST elevation MI and unstable angina [2]. Serial ECG recordings or continuous monitoring of ST-segment changes has been suggested to assist in evaluation of patients with chest pain. In the early hours of acute MI, the ST-segment is often dynamic, exhibiting cyclic patterns of thrombotic occlusion and spontaneous reperfusion. Even tough ST-segment monitoring for 8 to 12 hours of patients with chest pain is recommended by the AHA practical standards [3], studies have shown that its additive value is limited in the diagnostic evaluation of intermediate-risk patients managed in the chest-pain unit [4]. Conventional analysis of ST segment deviations aims at detecting repolarization abnormalities. However, ischemia may also bring about changes in the depolarization phase. These depolarization changes can be detected and quantified using analysis of the high- frequency components of the QRS complex (HFQRS). HFQRS have been previously shown to provide incremental diagnostic value in detecting demand ischemia [5]. HFQRS changes were also used to identify supply ischemia in patients undergoing percutaneous transluminal coronary angioplasty [6]. The aim of this study was to develop an analysis technique, based on HFQRS, for detecting transient ischemia and to evaluate its performance in continuous monitoring of patients hospitalized due to chest pain. 2. Methods 2.1. Patients and protocol The study was approved by the local ethics committee for medical research. The study included 52 consecutive patients (age 57–11 yo, 39 men) admitted to the chest paint unit (CPU). These patients, classified as having low to medium risk of ACS, were admitted for medical observation, continuous 12-lead ST monitoring, repeated biomarker testing and cardiac imaging tests for diagnosis of ischemic heart disease. Imaging tests included SPECT myocardial perfusion imaging, stress echocardiography or CT-angiography. Patients with positive ACS diagnosis underwent coronary angiography. Patients were continuously monitored by high-resolution 12-lead ECG (HyperQ™ System, BSP, Israel) during CPU stay ISSN 0276-6574 377 Computers in Cardiology 2008;35:377-380.