Digital 12-lead ECG Research Articles

Abstract 14012: The Importance of Right Bundle Branch Block in Suspected ST-Elevation Myocardial Infarction is Independent of ST-T Deviations

Background: In patients with suspected acute coronary syndrome (ACS) and bundle branch block (BBB) an emergent invasive strategy is recommended. While left bundle branch block (LBBB) is a well-known high-risk feature in patients with acute myocardial infarction, data supporting this strategy for right bundle branch block (RBBB) are scarce. Hypothesis: RBBB poses a high-risk feature in suspected ST-elevation myocardial infarction (STEMI). Methods: 2,139 patients with suspected STEMI were triaged to acute coronary angiography based on a prehospital, tele-transmitted digital 12-lead ECG. All discharge diagnoses were independently adjudicated. Sensitivity and specificity for STEMI-ECG-criteria were compared in RBBB and non-BBB patients. Adjusted hazard ratios for 1-year overall mortality were computed. Results: STEMI was adjudicated in 1,832/2,139 (85.6%) of all patients and in 102/117 (87.2%) of RBBB patients. ST-segment deviation followed typical patterns in most RBBB patients (n=36 for anterior, n=49 for inferior and n=15 for lateral STEMI) and among these a final diagnosis of STEMI was adjudicated in 88. Out of 17 patients without significant ST-changes, 14 (82%) were adjudicated as STEMI. Diagnostic accuracy of STEMI-criteria was comparable in RBBB and non-RBBB patients for anterior (sensitivity: 51.1% vs 59.1%, p=0.14; specificity: 66.7% vs 52.1%, p=0.33) and inferior STEMI (sensitivity: 35.2% vs 36.6%, p=0.80; specificity: 58.3% vs 49.5%, p=0.55). There was a statistical difference for lateral STEMI (sensitivity: 14.8% vs 4.4%, p=0.001; specificity: 75.0% vs 98.4%, p=0.001). Patients with RBBB had significantly higher1-year mortality compared to non-BBB patients. Adjusted hazard ratio for 1-year mortality comparing RBBB to non-BBB patients was 2.3% (95% CI 1.25-4.21, p=0.007). Conclusion: ECG-criteria for the detection of STEMI show comparable diagnostic accuracy in RBBB and non-BBB patients. However, STEMI was frequently present in RBBB patients not fulfilling diagnostic ECG-criteria. In addition, RBBB-patients showed poorer outcome after 1 year. Our data therefore reinforce current guidelines in designating ACS patients with RBBB, independent of ST-T deviations, high-risk and candidates for acute invasive evaluation.

Abstract 18317: Implications of Noise on Deep Learning Models for 12-Lead ECG Construction

Introduction: ECG construction with deep learning can help standardize ECGs and remove noise and artifacts, transforming potentially unusable ECGs into clinically useful ones. However, noise levels, often unpredictable in clinical settings, may impact model performance. Hypothesis: This study explores the relationship between ECG noise levels and deep learning model performance in constructing noise-free ECGs. We hypothesize that beyond a certain noise threshold, the model's output ceases to be clinically usable. Aims: Our study aims to determine noise effects on the performance of deep learning models in constructing 12-Lead ECGs, providing insights into the model's robustness and the optimal conditions for its application. Methods: Using 250 digital 12-Lead ECGs from the PTB database, we injected Gaussian noise (mean 0.0, standard deviations between 0.0 and 1.0) into these ECGs, all processed and scaled to unitless values (-1 to +1) and resampled at 100Hz. A model was asked to construct a noise-free ECG from a full 12-lead ECG with noise injection, which was compared to the original noise-free ECG. Results: The findings reveal two distinct linear phases in the relationship between noise and model performance (measured by mean absolute error, MAE). Between standard deviations of 0.0 and 0.03, the MAE increased marginally, while a sharp increase occurred between 0.03 and 1.0. Importantly, critical features of the original ECG were retained for noise levels up to 0.08, implying a noise threshold for clinical usability. Conclusions: Deep learning models, though not entirely resistant to noise, demonstrate efficacy in constructing ECGs up to certain noise levels (standard deviation 0.03). Beyond this, performance declines markedly. These findings outline the limits and potential of deep learning models like ECGio in clinical practice. Future research should focus on resilience to higher noise levels or noise-reduction preprocessing strategies.

PO-03-109 RIGHT VENTRICULAR PACING EXACERBATE ELECTRICAL DYSSYNCHRONY BEYOND INTRINSIC LEFT BUNDLE BRANCH BLOCK

Right ventricular (RV) pacing produces an iatrogenic form of left bundle branch block (LBBB). Intrinsic LBBB (I-LBBB) can cause adverse hemodynamic effects due to ventricular dyssynchrony, mainly in patients with heart failure and reduced ejection fraction (HFrEF). Although RV pacing is thought to create a similar activation pattern as LBBB, some studies suggest RV apical pacing can be harmful even in patients with I-LBBB. Currently, the mechanism responsible for this harm is not fully understood. We aimed to quantify and compare the degree of ventricular electrical dyssynchrony, as measured on the surface ECG, during I-LBBB, and RV pacing in patients with HFrEF (left ventricular EF ≤40%) and I-LBBB. Patients with HFrEF and prophylactic defibrillator therapy were prospectively recruited. For each patient, 3-minute, high-resolution (1024 Hz), digital 12-lead ECGs were recorded during I-LBBB rhythm and then RV pacing at 60 bpm or intrinsic heart rate (whichever was higher). High-resolution ECGs were analyzed to quantify abnormal intra-QRS peaks (QRSp) as a measure of ventricular electrical dyssynchrony. QRSp was defined as the mean number of peaks detected in the QRS complex that deviated from a smoothing-filtered version of the QRS. QRSp was automatically quantified for each precordial lead using validated custom software. Mean QRSp was defined as the average of QRSp in precordial leads. Thirty patients with HFrEF (66±9 yrs, 50% male, LVEF 30±8%) and I-LBBB (QRS duration [QRSd]: 172±22 ms) were included. Changes in mean QRSp between I-LBBB and RV apical pacing are shown in (Figure1A). RV pacing increased the mean QRSp in 27 patients (90%) as compared to I-LBBB, suggesting overall more electrical dyssynchrony during RV pacing. Changes in QRSd between I-LBBB and RV pacing are shown in (Figure 1B). RV pacing resulted in QRSd widening beyond I-LBBB QRSd in 18 patients (60%), while QRSd did not change in the remaining patients. During RV pacing (versus I-LBBB), Mean QRSp changed in more patients as compared to QRSd (27 patients versus 18 patients; p = 0.007). Although RV pacing produces an iatrogenic form of LBBB, ventricular electrical activation is more dyssynchronous during RV pacing as measured by changes in mean QRSp and QRSd. Compared to QRSd, mean QRSp is a more sensitive marker of change in ventricular electrical dyssynchrony between RV pacing and I-LBBB. Our results suggest RV pacing can exacerbate ventricular electrical dyssynchrony in those with I-LBBB.

Abstract 13586: In Patients With Left Bundle Branch Block, QRS Duration is a Stronger Predictor of Left Ventricular Hypertrophy and Dilation Than QRS Voltage

Introduction: ECG criteria for left ventricular (LV) hypertrophy in absence of left bundle branch block (LBBB) rely on QRS amplitudes. ECG correlates of LVH in setting of LBBB are not well established. We sought to evaluate quantitative LBBB ECG predictors of echocardiographic LV measurements [LV mass, end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF)]. Hypothesis: QRS duration (QRSd) is a better predictor of LV hypertrophy than QRS amplitudes. Methods: We included adult patients with typical LBBB having ECG and echocardiogram performed within 3 months of each other in 2010-2020. Digital 12-lead ECGs were processed to reconstruct orthogonal X, Y, Z leads using Kors matrix. Amplitudes and voltage-time-integrals (VTIs) from all 12 leads, X, Y, Z leads and 3D (root-mean-square) ECG, and QRSd were evaluated. We used age, sex and BSA-adjusted linear regressions to predict LV measurements, and separately generated ROC curves for identifying LV abnormalities. Results: We included 413 patients (53% women, age 73±12 yr). Among all the ECG variables, QRSd was most strongly correlated with all 4 LV measurements (all p<0.00001, see Table 1a showing QRSd compared to VTI and amplitude from 3D QRS). Table 1b shows area under ROC curves to identify LV abnormalities. QRS duration ≥150 ms in women had sensitivity 56.3%, specificity 64.4% for LVMi >95 g/m 2 and sensitivity 62.7%, specificity 67.8% for LVEDVi >61 mL/m 2 . QRS duration ≥160 ms in men had a sensitivity 63.1%, specificity 72.1% for LVMi >115 g/m 2 and sensitivity 58.3%, specificity 74.5% for LVEDVi >74 mL/m 2 . Conclusions: In patients with LBBB, QRS duration ≥150 in women and ≥160 in men is a superior predictor of adverse LV remodeling esp. eccentric LV hypertrophy/dilation. The weak correlation of QRS amplitude with hypertrophy may be due to the dominant effect of electrical dyssynchrony on voltage, and the heterogeneous effects of fibrosis, infiltrative cardiomyopathy, epicardial fat and LV dilation.

Abstract 11282: Artificial Intelligence Derived Algorithm to Screen for Ebstein’s Anomaly of the Tricuspid Valve Utilizing Echocardiography

Introduction: Ebstein’s anomaly (EA) is a rare congenital anomaly of the tricuspid valve associated with heart failure and arrhythmias. Currently available imaging modalities like transthoracic echocardiography, cardiac magnetic resonance imaging (MRI), or computerized tomography (CT) are not viable for screening large asymptomatic populations. Hypothesis: A deep learning algorithm, implemented using a convolutional neural network (CNN) applied to the standard 12-lead ECG can be used as an accurate screening tool for EA in children and adults. Method: We identified 1,154 EA patients and 21,430 age- and sex-matched controls (1:20 ratio) between 1987 - 2020. A CNN was trained, validated and tested using digital 12-lead ECGs in 70%, 10% and 20% of the cohort respectively, divided by random shuffling. Results: The mean age was 26.68±19.23 years (41.8% aged < 18 years, 56.41% female) for the EA cohort and 28.36±18.65 years (37.2% aged <18 years, 55.67% female) for the control cohort. Following training and validation the area under the curve (AUC) for detecting EA in the testing cohort was 0.976 (95% CI 0.962 - 0.989) with prediction accuracy of 0.959. In subgroup analysis, the AUC was comparable in males (0.970, 95% CI 0.946-0.995) and females (0.979, 95% CI 0.964-0.994), those <18 (0.980, 95% CI 0.958-1.000) and ≥18 years old (0.972, 95% CI 0.954-0.989), as well as in those with right bundle branch block (RBBB) (0.969, 95% CI 0.950-0.989) and without RBBB (0.950, 95% CI 0.922-0.978) on ECG. Conclusion: Artificial intelligence enhanced ECG based detection of EA is possible with high diagnostic accuracy. Following further refinement and external validation, this algorithm can facilitate early detection of EA using ECGs, a widely available and inexpensive tool. Figure: Receiver operating characteristic (ROC) curve for the performance of the artificial intelligence CNN in detecting EA in the test cohort stratified by (A) age (< 18 and ≥ 18 years) and sex and (B) presence of RBBB.

Abstract 14178: Artificial Intelligence (AI)-ECG Derived Biological Age Identifies Accelerated Aging in the Lamin A/C Gene ( LMNA ) Mutations

Introduction: LMNA encodes the nuclear envelope proteins lamin A/C that are involved in ageing processes. Mutations in LMNA cause a spectrum of phenotypes which include muscular dystrophy, arrhythmogenic dilated cardiomyopathy, peripheral neuropathy, progeroid syndrome and overlap syndrome. Early diagnosis is crucial to prevent sudden cardiac death. We hypothesized that LMNA mutations result in an ECG biological age older than chronological age. Methods: In this study, we used AI-enabled ECG network to investigate standard digital 12-lead ECGs in genetically characterized LMNA patients. Results: We identified 33 LMNA patients with 331, 12-lead ECG available for analysis. Median age at symptoms onset was 23.5 (range: <1 to 61 years;15 females). Mean follow up time was 86 months (range: <1 to 367 months). Symptoms at onset included muscle weakness (n=10) or cardiac symptoms (n=18) and asymptomatic (n=5). We included a total of 1532 age sex matched controls. We found significant differences in positive age-gap (biologically older than chronological age) comparing LMNA patients (median of 17.5) versus controls (4.9 years; p<0.001, Figure 1 ). AI-ECG age with 10 years or higher age-gap was observed in 73% of LMNA patients compared to 27% in the controls (p<0.001). Consecutive serial analysis for all ECG’s in the study versus predicted age showed faster aging when comparing the rates of age-gap overtime in the LMNA group compared to controls (p<0.0001, Figure 2 ). Moreover, the subset analysis of 23 ECG in the 5 asymptomatic patients versus controls also showed significantly higher AI-ECG age-gap (p=0.004) Conclusions: An AI-ECG algorithm to detect biological age effectively demonstrates accelerated aging in patients with LMNA mutations related to physiological aging. Our findings further validate the role of AI-ECG to detect accelerated aging, expanding previous studies demonstrating higher mortality among individuals with in increased positive age-gap.

B-PO03-162 MICROVOLT QRS ALTERNANS WITHOUT MICROVOLT T WAVE ALTERNANS IN HYPERTROPHIC CARDIOMYOPATHY: A NOVEL RISK MARKER OF LATE VENTRICULAR ARRHYTHMIAS

T wave alternans (TWA) is associated with ventricular arrhythmia (VA) but QRS alternans (QRSA) has not been characterized in hypertrophic cardiomyopathy (HCM). To assess microvolt QRSA/TWA in relation to traditional risk factors (RF) and late VA outcomes in HCM. Prospectively enrolled HCM patients (n=130, age 53±14yrs) with prophylactic ICD had digital 12-lead ECGs recorded during ventricular pacing at 100-120bpm. QRSA/TWA was quantified by the spectral method. Patients were categorized as QRSA+ and/or TWA+ based on ≥2 precordial leads having an alternans signal:noise >3 in ≥3 consecutive 128-beat segments. The VA endpoint was appropriate ICD therapy. QRSA+ and TWA+ occurred together in 28% and alone in 7% and 7% of patients, respectively. QRSA and TWA magnitude (1.9±0.6 vs 6.2±2.0μV, p<0.01; 1.3±0.3 vs 3.0±0.7μV, p=0.01) increased with pacing rate. LV thickness was greater in QRSA+ than QRSA- patients (22±7 vs 20±6mm, p=0.04). At 5 years follow up, 17% of patients had VA. QRSA+ patients had more VA than QRSA-, with QRSA+/TWA- patients having the highest VA rate (Figure). In patients with <2 RF, the annual VA rate was only 0.6% in QRSA- patients. Separate multivariable Cox models revealed QRSA+ (HR[95%CI]: 2.9[1.2-7.0], p=0.02) and QRSA+/TWA- (7.9[2.9-21.7], p<0.01) as the most significant predictors of VA. TWA and traditional RF did not predict VA. In HCM, microvolt QRSA is a novel rate-dependent phenomenon that can exist without TWA and is associated with greater LV thickness. QRSA in the absence of TWA increases VA risk 8-fold in all patients, while the absence of QRSA confers low VA risk in patients with <2 RF, which may inform ICD therapy.

Association of QRS-T angle and Late Gadolinium Enhancement in patients with a Clinical Suspicion of Myocarditis.

Objective: To investigate the association of a wide QRS-T angle on the surface ECG and late gadolinium enhancement on contrast-enhanced cardiovascular magnetic (CMR) imaging in patients with clinically suspected myocarditis.Background: Diagnosis and risk stratification in patients with suspected myocarditis is particularly challenging due to a great spectrum of clinical presentations. Late gadolinium enhancement (LGE) visualizes myocardial necrosis and fibrosis in patients with biopsy-proven myocarditis. The presence or absence of late gadolinium enhancements in these patients is prognostically meaningful. The QRS-T angle from the surface ECG, on the other hand, may serve as a simple and easily available risk marker in suspected myocarditis.Methods: We enrolled 97 consecutive patients that were referred to CMR imaging for a clinical suspicion of myocarditis. All patients obtained a standardized digital 12-lead ECG for the calculation of the QRS-T angle and underwent contrast-enhanced CMR imaging. Patients were divided into two groups according to the absence or presence of LGE on CMR.Results: 78 of 97 patients with suspected myocarditis had LGE on CMR. Patients with LGE had wider QRS-T angles as compared to the patient group without LGE (53.95-47.5 vs. 26.2-21.2; p<0.001). The sensivity, specificity, negative predictive value and positive predictive value for a QRS-T angle above 90 degrees for LGE positive myocarditis were 16.5%, 100%, 24.7%, and 100%, respectively.Conclusion: A wide QRS-T angle of 90 degrees or more is linked to myocardial fibrosis or necrosis (late gadolinium enhancement) in patients with suspected myocarditis.

Quantification of abnormal QRS peaks predicts response to cardiac resynchronization therapy and tracks structural remodeling.

BackgroundAlthough QRS duration (QRSd) is an important determinant of cardiac resynchronization therapy (CRT) response, non-responder rates remain high. QRS fragmentation can also reflect electrical dyssynchrony. We hypothesized that quantification of abnormal QRS peaks (QRSp) would predict CRT response.MethodsForty-seven CRT patients (left ventricular ejection fraction = 23±7%) were prospectively studied. Digital 12-lead ECGs were recorded during native rhythm at baseline and 6 months post-CRT. For each precordial lead, QRSp was defined as the total number of peaks detected on the unfiltered QRS minus those detected on a smoothed moving average template QRS. CRT response was defined as >5% increase in left ventricular ejection fraction post-CRT.ResultsSixty-percent of patients responded to CRT. Baseline QRSd was similar in CRT responders and non-responders, and did not change post-CRT regardless of response. Baseline QRSp was greater in responders than non-responders (9.1±3.5 vs. 5.9±2.2, p = 0.001) and decreased in responders (9.2±3.6 vs. 7.9±2.8, p = 0.03) but increased in non-responders (5.5±2.3 vs. 7.5±2.8, p = 0.049) post-CRT. In multivariable analysis, QRSp was the only independent predictor of CRT response (Odds Ratio [95% Confidence Interval]: 1.5 [1.1–2.1], p = 0.01). ROC analysis revealed QRSp (area under curve = 0.80) to better discriminate response than QRSd (area under curve = 0.67). Compared to QRSd ≥150ms, QRSp ≥7 identified response with similar sensitivity but greater specificity (74 vs. 32%, p<0.05). Amongst patients with QRSd <150ms, more patients with QRSp ≥7 responded than those with QRSp <7 (75 vs. 0%, p<0.05).ConclusionsOur novel automated QRSp metric independently predicts CRT response and decreases in responders. Electrical dyssynchrony assessed by QRSp may improve CRT selection and track structural remodeling, especially in those with QRSd <150ms.

Repolarization Heterogeneity Measured With T-Wave Area Dispersion in Standard 12-Lead ECG Predicts Sudden Cardiac Death in General Population.

We developed a novel electrocardiographic marker, T-wave area dispersion (TW-Ad), which measures repolarization heterogeneity by assessing interlead T-wave areas during a single cardiac cycle and tested whether it can identify patients at risk for sudden cardiac death (SCD) in the general population. TW-Ad was measured from standard digital 12-lead ECG in 5618 adults (46% men; age, 50.9±12.5 years) participating in the Health 2000 Study-an epidemiological survey representative of the Finnish adult population. Independent replication was performed in 3831 participants of the KORA S4 Study (Cooperative Health Research in the Region of Augsburg; 49% men; age, 48.7±13.7 years; mean follow-up, 8.8±1.1 years). During follow-up (7.7±1.4 years), 72 SCDs occurred in the Health 2000 Survey. Lower TW-Ad was univariately associated with SCD (0.32±0.36 versus 0.60±0.19; P<0.001); it had an area under the receiver operating characteristic curve of 0.809. TW-Ad (≤0.46) conferred a hazard ratio of 10.8 (95% confidence interval, 6.8-17.4; P<0.001) for SCD; it remained independently predictive of SCD after multivariable adjustment for clinical risk markers (hazard ratio, 4.6; 95% confidence interval, 2.7-7.4; P<0.001). Replication analyses performed in the KORA S4 Study confirmed an increased risk for cardiac death (unadjusted hazard ratio, 5.5; 95% confidence interval, 3.2-9.5; P<0.001; multivariable adjusted hazard ratio, 1.9; 95% confidence interval, 1.1-3.5; P<0.05). Low TW-Ad, reflecting increased heterogeneity of repolarization, in standard 12-lead resting ECGs is a powerful and independent predictor of SCD in the adult general population.

Abstract 21148: Stress-Induced Cardiac Repolarization Changes Are Associated With Mental Stress-Induced Mycoardial Ischemia

Introduction: Mental Stress-Induced Myocardial Ischemia (MSIMI) can occur in patients with coronary artery disease (CAD) during acute mental stress challenge and is associated with an increased risk of adverse cardiovascular events. Adverse dynamic changes in repolarization with stress may also be particularly important when assessing risk of MSIMI. Hypothesis: In subjects with CAD, dynamic ECG repolarization changes during stress and recovery are associated with MSIMI. Methods: We studied 419 individuals with CAD who underwent mental stress challenge via a standardized speech stressor. Digital 12-lead ECGs were collected at baseline during resting , 1 minute after speech started, immediately after speech, and during recovery (5 minutes after speech ended). Repolarization metrics, including QRS-T angle, T-axis, ST depression, and T wave area were quantified in all 12 leads. Heart rate was also included. Subjects with baseline artifact were excluded. Changes amongst baseline, stress, and recovery were analyzed. Forward selection with alpha=0.01 was used to help guide the final model. MSIMI was assessed via Tc99m myocardial perfusion imaging (read by experienced clinicians). Results: The mean (SD) age was 56 (10), 38% were women, and 17% had MSIMI. The most significant predictors of MSIMI, based on the multivariate model, are summarized in the table. The C-statistic was 0.73, but reduced to 0.71 after leave-one-out cross validation. Goodness of fit was adequate. No significant race or gender differences were noted. Conclusion: Baseline and dynamic ECG repolarization changes in recovery (but not peak stress) contribute to MSIMI classification, and may be useful for risk stratification of arrhythmia risk due to emotional triggers. While the mechanisms regarding the relationship of MSIMI and ECG metrics during recovery are unknown, they could be due to hysteresis phenomena and/or known autonomic changes during recovery from stress.

Automated Quantification of Low-Amplitude Abnormal QRS Peaks From High-Resolution ECG Recordings Predicts Arrhythmic Events in Patients With Cardiomyopathy.

Cardiomyopathy patients are at risk of sudden death, typically from scar-related abnormalities of electrical activation that promote ventricular tachyarrhythmias. Abnormal intra-QRS peaks may provide a measure of altered activation. We hypothesized that quantification of such QRS peaks (QRSp) in high-resolution ECGs would predict arrhythmic events in implantable cardioverter-defibrillator (ICD)-eligible cardiomyopathy patients. Ninety-nine patients with ischemic or non-ischemic dilated cardiomyopathy undergoing prophylactic ICD implantation were prospectively enrolled (age 62±11 years, left ventricular ejection fraction 27±7%). High-resolution (1024 Hz) digital 12-lead ECGs were recorded during intrinsic rhythm. QRSp was quantified for each precordial lead as the total number of low-amplitude deflections that deviated from their respective naive QRS template. The primary end point of arrhythmic events was defined as appropriate ICD therapy or sustained ventricular tachyarrhythmias. After a median follow-up of 24 (15-43) months, 20 (20%) patients had arrhythmic events. Both QRSp and QRS duration were greater in those with arrhythmic events (both P<0.001) and this was consistent for QRSp for both cardiomyopathy types. In a multivariable Cox regression model that included age, left ventricular ejection fraction, QRS duration, and QRSp, only QRSp was an independent predictor of arrhythmic events (hazard ratio, 2.1; P<0.001). Receiver operating characteristic analysis revealed that a QRSp ≥2.25 identified arrhythmic events with greater sensitivity (100% versus 70%, P<0.05) and negative predictive value (100% versus 89%, P<0.05) than QRS duration ≥120 ms. QRSp measured from high-resolution digital 12-lead ECGs independently predicts ventricular tachyarrhythmias in ICD-eligible cardiomyopathy patients. This novel QRS morphology index has the potential to improve sudden death risk stratification and patient selection for prophylactic ICD therapy.

ECG-derived spatial QRS-T angle is associated with ICD implantation, mortality and heart failure admissions in patients with LV systolic dysfunction.

BackgroundIncreased spatial QRS-T angle has been shown to predict appropriate implantable cardioverter defibrilIator (ICD) therapy in patients with left ventricular systolic dysfunction (LVSD). We performed a retrospective cohort study in patients with left ventricular ejection fraction (LVEF) 31–40% to assess the relationship between the spatial QRS-T angle and other advanced ECG (A-ECG) as well as echocardiographic metadata, with all-cause mortality or ICD implantation for secondary prevention.Methods534 patients ≤75 years of age with LVEF 31–40% were identified through an echocardiography reporting database. Digital 12-lead ECGs were retrospectively matched to 295 of these patients, for whom echocardiographic and A-ECG metadata were then generated. Data mining was applied to discover novel ECG and echocardiographic markers of risk. Machine learning was used to develop a model to predict possible outcomes.Results49 patients (17%) had events, defined as either mortality (n = 16) or ICD implantation for secondary prevention (n = 33). 72 parameters (58 A-ECG, 14 echocardiographic) were univariately different (p<0.05) in those with vs. without events. After adjustment for multiplicity, 24 A-ECG parameters and 3 echocardiographic parameters remained different (p<2x10-3). These included the posterior-to-leftward QRS loop ratio from the derived vectorcardiographic horizontal plane (previously associated with pulmonary artery pressure, p = 2x10-6); spatial mean QRS-T angle (134 vs. 112°, p = 1.6x10-4); various repolarisation vectors; and a previously described 5-parameter A-ECG score for LVSD (p = 4x10-6) that also correlated with echocardiographic global longitudinal strain (R2 = - 0.51, P < 0.0001). A spatial QRS-T angle >110° had an adjusted HR of 3.4 (95% CI 1.6 to 7.4) for secondary ICD implantation or all-cause death and adjusted HR of 4.1 (95% CI 1.2 to 13.9) for future heart failure admission. There was a loss of complexity between A-ECG and echocardiographic variables with an increasing degree of disease.ConclusionSpatial QRS-T angle >110° was strongly associated with arrhythmic events and all-cause death. Deep analysis of global ECG and echocardiographic metadata revealed underlying relationships, which otherwise would not have been appreciated. Delivered at scale such techniques may prove useful in clinical decision making in the future.

Abstract MP081: Short-long-short Sequence of RR Intervals is Associated With Increased Risk of Sudden Cardiac Death: the Atherosclerosis Risk in Community Study

Introduction: A Short-long-short sequence of RR’ intervals frequently precedes ventricular tachyarrhythmia. However, an association of RR’ alternans with sudden cardiac death (SCD) in the general population has not been previously studied. Hypothesis: We hypothesized that RR’ alternans is associated with SCD. Methods: We analyzed baseline visit 10-second ECG in 14,250 participants of the Atherosclerosis Risk in Communities (ARIC) cohort (mean age 54.1±5.87y; 45% men; 74% white). Only normal sinus beats were included in the semi-automated analysis. Digital 12-lead ECG was transformed into orthogonal ECG, from which respiratory signal was derived. Joint symbolic analysis was used to assess increase or decrease in successive RR’ intervals and R-instant respiratory phases calculated using the Hilbert transform. Short-long-short or long-short-long RR’ sequence was defined as RR’ alternans. RR’ alternans occurring during respiratory phase transitions were not considered. SCD, non-SCD, and non-cardiac death served as competing outcomes. Three competing risks models were constructed. Model 1 was adjusted for age, sex, race, and study center. Model 2 was in addition adjusted for prevalent CVD. Model 3 was further adjusted for subclinical CVD, its risk factors, including traditional ECG predictors, and medications affecting SA or AV nodes. Results: Over median follow-up of 24.4 years, there were 497 SCDs (incidence 1.66 [95%CI 1.52-1.82], 742 non-sudden cardiac deaths (incidence 2.48 [95%CI 2.31-2.67], and 3,753 non-cardiac deaths (incidence 12.6 [95%CI 12.1-13.0]) per 1,000 person-years. RR alternans was present in 1,679(11.8%) participants. In competing risks analysis (Table) the presence of RR’ alternans, excluding those occurring due to respiratory phase changes, was independently associated with increased risk of SCD. Conclusion: In competing risks analysis the presence of RR’ alternans that are not due to respiratory phase transition is associated with SCD. Further study of RR’ alternans is warranted.

Diagnostic and prognostic values of the V-index, a novel ECG marker quantifying spatial heterogeneity of ventricular repolarization, in patients with symptoms suggestive of non-ST-elevation myocardial infarction

BackgroundThe V-index is an ECG marker quantifying spatial heterogeneity of ventricular repolarization. We prospectively assessed the diagnostic and prognostic values of the V-index in patients with suspected non-ST-elevation myocardial infarction (NSTEMI). MethodsWe prospectively enrolled 497 patients presenting with suspected NSTEMI to the emergency department (ED). Digital 12-lead ECGs of five-minute duration were recorded at presentation. The V-index was automatically calculated in a blinded fashion. Patients with a QRS duration >120ms were ruled out from analysis. The final diagnosis was adjudicated by two independent cardiologists. The prognostic endpoint was all-cause mortality during 24months of follow-up. ResultsNSTEMI was the final diagnosis in 14% of patients. V-index levels were higher in patients with AMI compared to other causes of chest pain (median 23ms vs. 18ms, p<0.001). The use of the V-index in addition to conventional ECG-criteria improved the diagnostic accuracy for the diagnosis of NSTEMI as quantified by area under the ROC curve from 0.66 to 0.73 (p=0.001) and the sensitivity of the ECG for AMI from 41% to 86% (p<0.001). Cumulative 24-month mortality rates were 99.4%, 98.4% and 88.3% according to tertiles of the V-index (p<0.001). After adjustment for age and important ECG and clinical parameters, the V-index remained an independent predictor of death. ConclusionsThe V-index, an ECG marker quantifying spatial heterogeneity of ventricular repolarization, significantly improves the accuracy and sensitivity of the ECG for the diagnosis of NSTEMI and independently predicts mortality during follow-up.

ECG-Derived Spatial QRS-T Angle Is Associated with ICD Implantation, Mortality and Heart Failure Admissions in Patients with LV Systolic Dysfunction

Background: Increased spatial QRS-T angle has been shown to predict appropriate implantable cardioverter defibrilIator (ICD) therapy in patients with left ventricular systolic dysfunction (LVSD). We performed a retrospective cohort study to assess the relationship between advanced ECG (A-ECG) as well as echocardiographic metadata, with all-cause mortality or ICD implantation for secondary prevention. Methods: 534 patients ≤75 years of age with LVEF 31-40% were identified through an echocardiography reporting database. Digital 12-lead ECGs were retrospectively matched to 295 of these patients. Data mining and machine learning was applied to discover novel ECG and echocardiographic markers of risk. Results: 49 patients (17%) had events, defined as either mortality (n = 16) or ICD implantation for secondary prevention (n = 33). 72 parameters (58 A-ECG, 14 echocardiographic) were univariately different (p < 0.05) in those with vs. without events. After adjustment for multiplicity, 24 A-ECG parameters and 3 echocardiographic parameters remained different (p < 2 × 10-3). A spatial QRS-T angle >110o had an adjusted HR of 3.4 (95% CI 1.6 to 7.4) for secondary ICD implantation or all-cause death and adjusted HR of 4.1 (95% CI 1.2 to 13.9) for future heart failure admission. There was a loss of complexity between A-ECG and echocardiographic variables with worsening disease states. Conclusion: Spatial QRS-T angle >110o was strongly associated with arrhythmic events and all-cause death. Deep analysis of global ECG and echocardiographic metadata revealed underlying relationships, which otherwise would not have been appreciated.

Changes in cardiac repolarisation during spontaneous nocturnal hypoglycaemia in subjects with type 1 diabetes: a preliminary report

Experimental studies have revealed that hypoglycaemia can result in morphological changes in electrocardiographic repolarisation in subjects with type 1 diabetes. However, the influence of spontaneous nocturnal hypoglycaemia on repolarisation morphology in a 'real life' situation is not clear. Adults with type 1 diabetes (n=11) underwent continuous glucose monitoring with a subcutaneous sensor and digital 12-lead ECG recording for three nights. T-wave morphology was analysed with custom-made software during both hypoglycaemia (glucose <3.5mmol/l at least 20min) from ten consecutive heart beats in the middle of the deepest hypoglycaemia and from a control nonhypoglycaemic period (glucose ≥5.0mmol/l) from the same recording. In the comparison of 10 hypoglycaemia-control pairs, heart rate (65±12 beats/min during normoglycaemia versus 85±19 beats/min during hypoglycaemia, p=0.028) increased and the QTc interval (439±5 vs. 373±5ms, respectively, p=0.025) decreased significantly during hypoglycaemia. The spatial QRS-T angle (TCRT) was reduced, and the roughness of the T-wave loop (T-E) increased significantly (p=0.037 for both) in the patients during hypoglycaemia. In adults with type 1 diabetes, spontaneous nocturnal hypoglycaemia results in morphological changes and increased heterogeneity of global cardiac repolarisation. These changes may contribute to the risk of 'dead in bed' syndrome encountered in young individuals with type 1 diabetes.

Comparison of the spatial QRS-T angle derived from digital ECGs recorded using conventional electrode placement with that derived from Mason-Likar electrode position

BackgroundThe spatial QRS-T angle is ideally derived from orthogonal leads. We compared the spatial QRS-T angle derived from orthogonal leads reconstructed from digital 12-lead ECGs and from digital Holter ECGs recorded with the Mason-Likar (M-L) electrode positions. Methods and resultsOrthogonal leads were constructed by the inverse Dower method and used to calculate spatial QRS-T angle by (1) a vector method and (2) a net amplitude method, in 100 volunteers.Spatial QRS-T angles from standard and M-L ECGs differed significantly (57°±18° vs 48°±20° respectively using net amplitude method and 53°±28° vs 48°±23° respectively by vector method; p<0.001). Difference in amplitudes in leads V4–V6 was also observed between Holter and standard ECGs, probably due to a difference in electrical potential at the central terminal. ConclusionMean spatial QRS-T angles derived from standard and M-L lead systems differed by 5°–9°. Though statistically significant, these differences may not be clinically significant.

89-02: Vectorcardiographic QRS area identifies delayed left ventricular lateral wall activation determined by electroanatomic mapping in candidates for cardiac resynchronization therapy

Introduction: Delayed left ventricular lateral wall (LVLW) activation is considered the electrical substrate underlying LV dysfunction amenable to cardiac resynchronization therapy (CRT). The purpose of this study was to assess LVLW activation in CRT candidates using coronary venous electroanatomic mapping (EAM) and to investigate whether the QRS area (QRSAREA) on the vectorcardiogram (VCG) can identify delayed LVLW activation. Methods: 51 consecutive CRT candidates (29 LBBB, 15 IVCD, 7 RBBB) underwent intraprocedural coronary venous EAM using EnSite NavX. VCGs were constructed from preprocedural digital 12-lead ECGs using the Kors method. QRSAREA was assessed and compared to QRS duration and 5 different LBBB definitions. Results: Delayed LVLW activation (activation time >75% of QRS duration, example in figure) occurred in 38 of 51 patients (29/29 LBBB, 8/15 IVCD, 1/7 RBBB). QRSAREA was larger in patients with than in patients without delayed LVLW activation (108 ± 42 µVs vs 51 ± 27 µVs, P < .001), and identified delayed LVLW activation better than QRS duration (area under the curve 0.89 [95% confidence interval 0.79–0.99] vs 0.49 [95% confidence interval 0.33-0.65] ). QRSAREA >69 µVs diagnosed delayed LVLW activation with a higher sum of sensitivity (87%) and specificity (92%) than any of the LBBB definitions. Of the different LBBB definitions, the European Society of Cardiology textbook definition performed best with sensitivity of 76% and specificity of 100%. Conclusion: Coronary venous EAM can be used during CRT implantation to determine the presence of delayed LVLW activation. QRSAREA is a noninvasive alternative for intracardiac measurements of electrical activation, which identifies delayed LVLW activation better than QRS duration and LBBB morphology.

Vectorcardiographic QRS area identifies delayed left ventricular lateral wall activation determined by electroanatomic mapping in candidates for cardiac resynchronization therapy

Delayed left ventricular (LV) lateral wall (LVLW) activation is considered the electrical substrate underlying LV dysfunction amenable to cardiac resynchronization therapy (CRT). The purpose of this study was to assess LVLW activation in CRT candidates using coronary venous electroanatomic mapping (EAM) and to investigate whether the QRS area (QRSAREA) on the vectorcardiogram (VCG) can identify delayed LVLW activation. Fifty-one consecutive CRT candidates (29 left bundle branch block [LBBB], 15 intraventricular conduction delay [IVCD], 7 right bundle branch block [RBBB]) underwent intraprocedural coronary venous EAM using EnSite NavX. VCGs were constructed from preprocedural digital 12-lead ECGs using the Kors method. QRSAREA was assessed and compared to QRS duration and 5 different LBBB definitions. Delayed LVLW activation (activation time >75% of QRS duration) occurred in 38 of 51 patients (29/29 LBBB, 8/15 IVCD, 1/7 RBBB). QRSAREA was larger in patients with than in patients without delayed LVLW activation (108 ± 42 µVs vs 51 ± 27 µVs, P < .001), and identified delayed LVLW activation better than QRS duration (area under the curve 0.89 [95% confidence interval 0.79-0.99] vs 0.49 [95% confidence interval 0.33-0.65]). QRSAREA >69 µVs diagnosed delayed LVLW activation with a higher sum of sensitivity (87%) and specificity (92%) than any of the LBBB definitions. Of the different LBBB definitions, the European Society of Cardiology textbook definition performed best with sensitivity of 76% and specificity of 100%. Coronary venous EAM can be used during CRT implantation to determine the presence of delayed LVLW activation. QRSAREA is a noninvasive alternative for intracardiac measurements of electrical activation, which identifies delayed LVLW activation better than QRS duration and LBBB morphology.