Assessment of Artificial Intelligence in Echocardiography Diagnostics in Differentiating Takotsubo Syndrome From Myocardial Infarction.

Importance: Machine learning algorithms enable the automatic classification of cardiovascular diseases based on raw cardiac ultrasound imaging data. However, the utility of machine learning in distinguishing between takotsubo syndrome (TTS) and acute myocardial infarction (AMI) has not been studied. Objectives: To assess the utility of machine learning systems for automatic discrimination of TTS and AMI. Design, Settings, and Participants: This cohort study included clinical data and transthoracic echocardiogram results of patients with AMI from the Zurich Acute Coronary Syndrome Registry and patients with TTS obtained from 7 cardiovascular centers in the International Takotsubo Registry. Data from the validation cohort were obtained from April 2011 to February 2017. Data from the training cohort were obtained from March 2017 to May 2019. Data were analyzed from September 2019 to June 2021. Exposure: Transthoracic echocardiograms of 224 patients with TTS and 224 patients with AMI were analyzed. Main Outcomes and Measures: Area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the machine learning system evaluated on an independent data set and 4 practicing cardiologists for comparison. Echocardiography videos of 228 patients were used in the development and training of a deep learning model. The performance of the automated echocardiogram video analysis method was evaluated on an independent data set consisting of 220 patients. Data were matched according to age, sex, and ST-segment elevation/non-ST-segment elevation (1 patient with AMI for each patient with TTS). Predictions were compared with echocardiographic-based interpretations from 4 practicing cardiologists in terms of sensitivity, specificity, and AUC calculated from confidence scores concerning their binary diagnosis. Results: In this cohort study, apical 2-chamber and 4-chamber echocardiographic views of 110 patients with TTS (mean [SD] age, 68.4 [12.1] years; 103 [90.4%] were female) and 110 patients with AMI (mean [SD] age, 69.1 [12.2] years; 103 [90.4%] were female) from an independent data set were evaluated. This approach achieved a mean (SD) AUC of 0.79 (0.01) with an overall accuracy of 74.8 (0.7%). In comparison, cardiologists achieved a mean (SD) AUC of 0.71 (0.03) and accuracy of 64.4 (3.5%) on the same data set. In a subanalysis based on 61 patients with apical TTS and 56 patients with AMI due to occlusion of the left anterior descending coronary artery, the model achieved a mean (SD) AUC score of 0.84 (0.01) and an accuracy of 78.6 (1.6%), outperforming the 4 practicing cardiologists (mean [SD] AUC, 0.72 [0.02]) and accuracy of 66.9 (2.8%). Conclusions and Relevance: In this cohort study, a real-time system for fully automated interpretation of echocardiogram videos was established and trained to differentiate TTS from AMI. While this system was more accurate than cardiologists in echocardiography-based disease classification, further studies are warranted for clinical application.

Telemedical cardiac risk assessment by implantable cardiac monitors in patients after myocardial infarction with autonomic dysfunction (SMART-MI-DZHK9): a prospective investigator-initiated, randomised, multicentre, open-label, diagnostic trial.

BACKGROUND: Cardiac autonomic dysfunction after myocardial infarction identifies patients at high risk despite only moderately reduced left ventricular ejection fraction. We aimed to show that telemedical monitoring with implantable cardiac monitors in these patients can improve early detection of subclinical but prognostically relevant arrhythmic events. METHODS: We did a prospective investigator-initiated, randomised, multicentre, open-label, diagnostic trial at 33 centres in Germany and Austria. Survivors of acute myocardial infarction with left ventricular ejection fraction of 36-50% had biosignal analysis for assessment of cardiac autonomic function. Patients with abnormal periodic repolarisation dynamics (≥5·75 deg2) or abnormal deceleration capacity (≤2·5 ms) were randomly assigned (1:1) to telemedical monitoring with implantable cardiac monitors or conventional follow-up. Primary endpoint was time to detection of serious arrhythmic events defined by atrial fibrillation 6 min or longer, atrioventricular block class IIb or higher and fast non-sustained (>187 beats per min; ≥40 beats) or sustained ventricular tachycardia or fibrillation. This study is registered with ClinicalTrials.gov, NCT02594488. FINDINGS: Between May 12, 2016, and July 20, 2020, 1305 individuals were screened and 400 patients at high risk were randomly assigned (median age 64 years [IQR 57-73]); left ventricular ejection fraction 45% [40-48]) to telemedical monitoring with implantable cardiac monitors (implantable cardiac monitor group; n=201) or conventional follow-up (control group; n=199). During median follow-up of 21 months, serious arrhythmic events were detected in 60 (30%) patients of the implantable cardiac monitor group and 12 (6%) patients of the control group (hazard ratio 6·33 [IQR 3·40-11·78]; p<0·001). An improved detection rate by implantable cardiac monitors was observed for all types of serious arrhythmic events: atrial fibrillation 6 min or longer (47 [23%] patients vs 11 [6%] patients; p<0·001), atrioventricular block class IIb or higher (14 [7%] vs 0; p<0·001) and ventricular tachycardia or ventricular fibrillation (nine [4%] patients vs two [1%] patients; p=0·054). INTERPRETATION: In patients at high risk after myocardial infarction and cardiac autonomic dysfunction but only moderately reduced left ventricular ejection fraction, telemedical monitoring with implantable cardiac monitors was highly effective in early detection of subclinical, prognostically relevant serious arrhythmic events. FUNDING: German Centre for Cardiovascular Research (DZHK) and Medtronic Bakken Research Center.

Evaluation of a novel portable capacitive ECG system in the clinical practice for a fast and simple ECG assessment in patients presenting with chest pain: FIDET (Fast Infarction Diagnosis ECG Trial).

BACKGROUND: Electrocardiogram (ECG) assessment plays a crucial role in patients presenting with chest pain and suspected acute coronary syndrome (ACS). In a pilot study, we previously evaluated a capacitive ECG system (cECG) as a novel ECG technique for a fast and simple ECG assessment in patients with ST-elevation myocardial infarction (STEMI). In a next step, the sensitivity and specificity of this novel ECG technique have to be assessed in patients with ACS. HYPOTHESIS: The Fast Infarction Diagnosis ECG Trial (FIDET) is a prospective, bi-center, observer-blinded noninferiority study to evaluate the cECG compared to the conventional ECG (kECG) in the clinical practice for ECG assessment in consecutive patients presenting with suspected ACS. METHODS: In 250 patients who were admitted to the hospital, because of an ACS [including STEMI and non-ST-elevation acute coronary syndrome (NSTE-ACS)], both a kECG and a cECG recording were performed within a time lag of less than 10 min. END POINTS: The primary end point will be sensitivity and specificity of the cECG compared to the kECG in diagnosing a STEMI with a margin of noninferiority of 7.5 %. Secondary end points include sensitivity and specificity of the cECG compared to the kECG in diagnosing an NSTE-ACS, safety of the cECG system (adverse event, serious adverse event and suspected unexpected serious adverse reaction), parameters of the ECG measurement (PQ-interval, QT-interval, ST-amplitude and heart rate) and measurement duration of the two methods. CONCLUSION: FIDET is designed as a noninferiority study to show that a novel cECG system is suitable for the diagnosis of myocardial infarction in the clinical context and might even have benefits, for example by offering a faster and easier ECG assessment.

Simulation of Ca-calmodulin-dependent protein kinase II on rabbit ventricular myocyte ion currents and action potentials.

Ca-calmodulin-dependent protein kinase II (CaMKII) was recently shown to alter Na(+) channel gating and recapitulate a human Na(+) channel genetic mutation that causes an unusual combined arrhythmogenic phenotype in patients: simultaneous long QT syndrome and Brugada syndrome. CaMKII is upregulated in heart failure where arrhythmias are common, and CaMKII inhibition can reduce arrhythmias. Thus, CaMKII-dependent channel modulation may contribute to acquired arrhythmic disease. We developed a Markovian Na(+) channel model including CaMKII-dependent changes, and incorporated it into a comprehensive myocyte action potential (AP) model with Na(+) and Ca(2+) transport. CaMKII shifts Na(+) current (I(Na)) availability to more negative voltage, enhances intermediate inactivation, and slows recovery from inactivation (all loss-of-function effects), but also enhances late noninactivating I(Na) (gain of function). At slow heart rates, with long diastolic time for I(Na) recovery, late I(Na) is the predominant effect, leading to AP prolongation (long QT syndrome). At fast heart rates, where recovery time is limited and APs are shorter, there is little effect on AP duration, but reduced availability decreases I(Na), AP upstroke velocity, and conduction (Brugada syndrome). CaMKII also increases cardiac Ca(2+) and K(+) currents (I(Ca) and I(to)), complicating CaMKII-dependent AP changes. Incorporating I(Ca) and I(to) effects individually prolongs and shortens AP duration. Combining I(Na), I(Ca), and I(to) effects results in shortening of AP duration with CaMKII. With transmural heterogeneity of I(to) and I(to) downregulation in heart failure, CaMKII may accentuate dispersion of repolarization. This provides a useful initial framework to consider pathways by which CaMKII may contribute to arrhythmogenesis.