Elevated T-Wave Alternans Predicts Nonsustained Ventricular Tachycardia in Association with Percutaneous Coronary Intervention in ST-Segment Elevation Myocardial Infarction (STEMI) Patients

Novel application of convolutional neural networks for artificial intelligence-enabled modified moving average analysis of P-, R-, and T-wave alternans for detection of risk for atrial and ventricular arrhythmias

BACKGROUND

T-wave alternans (TWA) analysis was shown in >14,000 individuals studied worldwide over the past two decades to be a useful tool to assess risk for cardiovascular mortality and sudden arrhythmic death. TWA analysis by the modified moving average (MMA) method is FDA-cleared and CMS-reimbursed (CAG-00293R2).

OBJECTIVE

Because the MMA technique is inherently suitable for dynamic tracking of alternans levels, it was selected for development of artificial intelligence (AI)-enabled algorithms using convolutional neural networks (CNN) to achieve rapid, efficient, and accurate assessment of P-wave alternans (PWA), R-wave alternans (RWA), and TWA.

METHODS

The novel application of CNN algorithms to enhance MMA analysis generated efficient and powerful pattern-recognition algorithms for highly accurate alternans quantification. Algorithm reliability and accuracy were verified using simulated ECGs achieving R2 ≥ 0.99 (p < 0.01) in response to noise inputs and artifacts that emulate real-life conditions.

RESULTS

Accuracy of the new AI-MMA algorithms in TWA analysis (n = 5) was significantly improved over unsupervised, automated MMA output (p = 0.036) and did not differ from conventional MMA analysis with expert overreading (p = 0.21). Accuracy of AI-MMA in PWA analysis (n = 45) was significantly improved over unsupervised, automated MMA output (p < 0.005) and did not differ from conventional MMA analysis with expert overreading (p = 0.89). TWA and PWA by AI-MMA were correlated with conventional MMA output over-read by an expert reader (R2 = 0.7765, R2 = 0.9504, respectively).

CONCLUSION

This novel technique for AI-MMA analysis could be suitable for use in diverse in-hospital and out-of-hospital monitoring systems, including cardiac implantable electronic devices and smartwatches, for tracking atrial and ventricular arrhythmia risk.

Continuous multi-day tracking of post-myocardial infarction recovery of cardiac electrical stability and autonomic tone using electrocardiogram patch monitors

BACKGROUND

Sudden cardiac death (SCD) risk is elevated following acute myocardial infarction (MI). The time course of SCD susceptibility post-MI requires further investigation.

METHODS

In this observational cohort study, we employed state-of-the-art noninvasive ECG techniques to track the daily time course of cardiac electrical instability and autonomic function following ST-segment elevation myocardial infarction (STEMI) and non-STEMI (NSTEMI). Preventice BodyGuardian MINI-EL Holters continuously recorded ECGs for 7 days at hospital discharge and at 40 days for STEMI (N = 5) or at 90 days for NSTEMI patients (N = 5). Cardiac electrical instability was assessed by T-wave alternans (TWA) and T-wave heterogeneity (TWH); autonomic tone was determined by rMSSD-heart rate variability (HRV).

RESULTS

TWA was severely elevated (≥60 μV) in STEMI patients (80 ± 10.3 μV) at discharge and throughout the first recording period but declined by 50% to 40 ± 2.3 μV (p = .03) by Day 40 and remained in the normal range (<47 μV). TWH, a related phenomenon analyzed from 12-lead ECGs, was reduced by 63% in the five STEMI patients from discharge to normal (<80 μV) at follow-up (105 ± 27.3 to 39 ± 3.3 μV, p < .04) but increased by 65% in a STEMI case (89 to 147 μV), who received a wearable defibrillator vest and later implantable cardioverter defibrillator. In NSTEMI patients, TWA was borderline abnormal (47 ± 3.3 μV) at discharge and declined by 19% to normal (38 ± 1.2 μV) by Day 90 (p = .05). An overall reciprocal increase in rMSSD-HRV suggested recovery of vagal tone.

CONCLUSIONS

This study provides proof-of-principle for tracking post-MI SCD risk in individual patients with implications for personalized therapy.

Electrical alternans induced by a brief period of myocardial ischemia during percutaneous coronary intervention: The characteristic ECG morphology and relationship to mechanical alternans

BACKGROUND

Electrical alternans (EA) has not been fully studied in the current percutaneous coronary intervention (PCI) procedure.

OBJECTIVE

The purpose of this study was to evaluate visible EA and the morphology of ST segment during PCI.

METHODS

The incidence of visible EA and ST-segment morphology were studied while the coronary artery was occluded for 20 seconds. When data were available, the relationship between EA and blood pressure was analyzed. The clinical and electrocardiographic data were compared with those of the age- and sex-matched controls.

RESULTS

During balloon inflation, visible EA was observed in 5 of 306 patients (1.6%) in the last 2 years. EA was limited to PCI in the proximal left anterior descending artery. The ST segment elevated to 10.1 ± 3.2 mm, followed by an alternating QRS complex with a lower ST segment (5.6 ± 1.9 mm; P = .0047) with characteristic ST-segment morphology, which is known as lambda pattern. The mean age of the 5 patients was 68 ± 20 years, and 4(80%). were men. After the release of inflation, the ST-segment level returned rapidly to baseline, followed by normalization of J point. Compared with controls, the maximal elevated ST segment was significantly higher in patients with EA (5.7 ± 2.7 mm; P = .0028). The occlusion of the proximal left anterior descending artery with more severe ischemia seemed to be a prerequisite for developing EA. A higher ST segment was associated with a lower blood pressure and vice versa.

CONCLUSION

A short period of ischemia during PCI may induce visible EA and alternating QRS complexes with a characteristic ST-segment morphology. A higher ST segment was associated with a lower blood pressure and vice versa.

Utility of T-wave alternans during night time as a predictor for ventricular fibrillation in patients with Brugada syndrome

The prognostic value of T-wave alternans (TWA) during the night time in patients with Brugada syndrome (Br-S) remains unknown. We assessed TWA for risk stratification using 24-h multichannel Holter electrocardiogram (24-M-ECG) in Br-S. We enrolled 129 patients with Br-S [grouped according to histories of ventricular fibrillation (VF), n = 16; syncope, n = 10; or no symptoms (asymptomatic), n = 103] and 11 controls. Precordial electrodes were attached to the third (3L-V1, 3L-V2) and fourth (4L-V1, 4L-V2 and 4L-V5) intercostal spaces. We measured the values of maximum TWA (max-TWA) during the night time (12 a.m.-6 a.m.) and the day time (12 p.m.-6 p.m.) and calculated parameters of heart rate variability. Compared to the asymptomatic and control groups, the VF and syncope groups showed significantly greater 3L-V2 max-TWA during the night time. The cutoff value for the 3L-V2 max-TWA during the night time was determined as 20 µV (sensitivity 94 % and specificity 48 %; p = 0.01). Multivariate analysis revealed that 3L-V2 max-TWA during the night time ≥20 µV and previous VF episodes were independent predictors of future VF episodes. During a mean follow-up period of 68 ± 37 months, 16 patients experienced VF episodes. The incidence of VF episodes was the highest during the night time (p < 0.001). The 3L-V2 max-TWA during the night time may be a useful predictor for VF episodes in patients with Br-S.

Evaluation of T-wave alternans in high-resolution ECG maps recorded during the stress test in patients after myocardial infarction

INTRODUCTION

Recent studies point to analysis of T-wave alternans as a promising indicator of an increased risk of life-threatening ventricular arrhythmias. In this study the occurrence of T-wave alternans in the high-resolution ECGs recorded during the exercise stress test and scintigraphic tests (SPECT) in patients with ischemic heart disease was examined.

MATERIAL AND METHODS

The study group consisted of 33 patients after myocardial infarction. In the group of patients after myocardial infarction and with low left ventricular ejection fraction correlations of 70% between the test results of T-wave alternans and SPECT and 60% between the test results of T-wave alternans and stress test were found.

RESULTS

In the group of patients after myocardial infarction but with high left ventricular ejection fraction correlations were respectively 39% and 48%. The analysis of the electrocardiographic maps showed a strong dependence of this correlation on the T-wave alternans amplitude and location of the ECG measuring electrode on the chest. The results might suggest that in patients after myocardial infarction and at increased risk for sudden cardiac death T-wave alternans may also provide information about cardiac electrical instability associated with ischemia.

CONCLUSIONS

It can also be assumed that the position of the electrode where the highest level of the T-wave alternans was detected can indicate the location of the ischemic region of the heart.

Quantitative T-wave alternans analysis for guiding medical therapy: an underexploited opportunity

Reducing the toll of sudden cardiac death (SCD) remains a major challenge in cardiology, as it is the leading cause of adult mortality in the industrially developed world, claiming 310,000 lives annually in the United States alone. The main contemporary noninvasive index of cardiovascular risk, left ventricular ejection fraction (LVEF), has not proved adequately reliable, as the majority of individuals who die suddenly have relatively preserved cardiac mechanical function. Monitoring of T-wave alternans (TWA), a beat-to-beat fluctuation in ST-segment or T-wave morphology, is an attractive approach to risk stratification on both scientific and clinical grounds, as this ECG phenomenon has been shown using the FDA-cleared Spectral and Modified Moving Average methods to assess risk for cardiovascular mortality including SCD in studies enrolling >12,000 individuals with depressed or preserved LVEF. The evidence supporting TWA as a therapeutic target is reviewed.

Relation of T-wave alternans to mortality and nonsustained ventricular tachycardia in patients with non-ST-segment elevation acute coronary syndrome from the MERLIN-TIMI 36 trial of ranolazine versus placebo

We explored the utility of T-wave alternans (TWA) in predicting mortality in patients with non-ST-segment elevation acute coronary syndrome (NSTEACS). Maximum TWA was calculated using Modified Moving Average method from continuous electrocardiographic recordings in patients with left ventricular ejection fraction <40% and ventricular tachycardia (VT) ≥4 beats during index hospitalization or sudden cardiac death during the follow-up year and age- and sex-matched controls in the Metabolic Efficiency with Ranolazine for Less Ischemia in Non-ST Elevation Acute Coronary Syndrome-Thrombolysis In Myocardial Infarction (MERLIN-TIMI) 36 trial. All patients received standard therapy for NSTEACS plus ranolazine (n = 109) or placebo (n = 101). Median follow-up was 1 year. Baseline clinical characteristics did not differ between patients with elevated TWA (≥47 μV) compared with lower levels. Patients with TWA ≥47 μV at admission had increased risk of total mortality (adjusted odds ratio [ORadj] 2.35, p = 0.04) during follow-up and VT ≥4 beats (ORadj 2.70, p = 0.01) during hospitalization with a trend toward increased cardiovascular death risk (ORadj 2.18, p = 0.07) during follow-up. In patients receiving placebo, TWA ≥47 μV on day 6 was associated with increased risk of total mortality (OR 4.12, 95% confidence interval 1.25 to 13.64, p = 0.02) and cardiovascular death (OR 4.73, p = 0.01) during follow-up. No deaths occurred among patients with TWA ≥47 μV assigned to ranolazine. In conclusion, in patients with NSTEACS and left ventricular ejection fraction <40%, TWA ≥47 μV early after admission is associated with increased risk of mortality at 1 year and with nonsustained VT during hospitalization. TWA may be useful in risk estimation in patients with NSTEACS. The possibility that TWA may serve as a therapeutic target deserves further exploration.

Time-domain T-wave alternans is strongly associated with a history of ventricular fibrillation in patients with Brugada syndrome

AIMS

T-wave alternans (TWA) is an indicator of vulnerability to ventricular arrhythmias and is useful for predicting sudden cardiac death (SCD) in patients with various structural heart diseases. We evaluated whether high levels of time-domain TWA on ambulatory ECG (AECG) are associated with a history of ventricular fibrillation (VF) in Brugada syndrome (BrS) patients.

METHODS AND RESULTS

We examined the associations among VF history, family history of SCD, spontaneous type 1 electrocardiogram (ECG), late potentials, VF induction by programmed electrical stimulation, and TWA in 45 BrS patients (44 males; mean age, 45 ± 15 years). TWA analyzed from 24-h AECG recordings using the modified moving average method was positive in 13 of 43 patients (30%). Patients with a history of VF had a significantly higher incidence of a positive TWA test (82% vs. 13%; P < 0.001) and spontaneous type 1 ECG (92% vs. 38%; P = 0.007) than those without VF history. Multivariate analysis indicated that positive TWA (OR 7.217; 95% CI 2.503-35.504; P = 0.002) and spontaneous type 1 ECG (OR 5.530; 95% CI 1.651-34.337; P = 0.020) were closely associated with VF history. Spontaneous type 1 ECG had high sensitivity (92%) but low specificity (63%). Positive TWA was a reliable marker with high sensitivity and specificity (82% and 88%, respectively).

CONCLUSION

Elevated time-domain TWA on AECG confirms arrhythmia risk in symptomatic BrS patients without the need for provocative stimuli.

T-wave alternans as an arrhythmic risk stratifier: state of the art

Microvolt level T-wave alternans (MTWA), a phenomenon of beat-to-beat variability in the repolarization phase of the ventricles, has been closely associated with an increased risk of ventricular tachyarrhythmic events (VTE) and sudden cardiac death (SCD) during medium- and long-term follow-up. Recent observations also suggest that heightened MTWA magnitude may be closely associated with short-term risk of impending VTE. At the subcellular and cellular level, perturbations in calcium transport processes likely play a primary role in the genesis of alternans, which then secondarily lead to alternans of action potential morphology and duration (APD). As such, MTWA may play a role not only in risk stratification but also more fundamentally in the pathogenesis of VTE. In this paper, we outline recent advances in understanding the pathogenesis of MTWA and also the utility of T-wave alternans testing for clinical risk stratification. We also highlight emerging clinical applications for MTWA.

Maximizing the value of mobile health monitoring by avoiding redundant patient reports: prediction of depression-related symptoms and adherence problems in automated health assessment services

Background

Interactive voice response (IVR) calls enhance health systems’ ability to identify health risk factors, thereby enabling targeted clinical follow-up. However, redundant assessments may increase patient dropout and represent a lost opportunity to collect more clinically useful data.

Objective

We determined the extent to which previous IVR assessments predicted subsequent responses among patients with depression diagnoses, potentially obviating the need to repeatedly collect the same information. We also evaluated whether frequent (ie, weekly) IVR assessment attempts were significantly more predictive of patients’ subsequent reports than information collected biweekly or monthly.

Methods

Using data from 1050 IVR assessments for 208 patients with depression diagnoses, we examined the predictability of four IVR-reported outcomes: moderate/severe depressive symptoms (score ≥10 on the PHQ-9), fair/poor general health, poor antidepressant adherence, and days in bed due to poor mental health. We used logistic models with training and test samples to predict patients’ IVR responses based on their five most recent weekly, biweekly, and monthly assessment attempts. The marginal benefit of more frequent assessments was evaluated based on Receiver Operator Characteristic (ROC) curves and statistical comparisons of the area under the curves (AUC).

Results

Patients’ reports about their depressive symptoms and perceived health status were highly predictable based on prior assessment responses. For models predicting moderate/severe depression, the AUC was 0.91 (95% CI 0.89-0.93) when assuming weekly assessment attempts and only slightly less when assuming biweekly assessments (AUC: 0.89; CI 0.87-0.91) or monthly attempts (AUC: 0.89; CI 0.86-0.91). The AUC for models predicting reports of fair/poor health status was similar when weekly assessments were compared with those occurring biweekly (P value for the difference=.11) or monthly (P=.81). Reports of medication adherence problems and days in bed were somewhat less predictable but also showed small differences between assessments attempted weekly, biweekly, and monthly.

Conclusions

The technical feasibility of gathering high frequency health data via IVR may in some instances exceed the clinical benefit of doing so. Predictive analytics could make data gathering more efficient with negligible loss in effectiveness. In particular, weekly or biweekly depressive symptom reports may provide little marginal information regarding how the person is doing relative to collecting that information monthly. The next generation of automated health assessment services should use data mining techniques to avoid redundant assessments and should gather data at the frequency that maximizes the value of the information collected.