Multichannel seismocardiography: an imaging modality for investigating heart vibrations.

OBJECTIVE: Seismocardiography is the measurement of vibration waves caused by the beating heart with accelerometer(s) placed on the chest. Investigating the nature and the behavior of these vibration waves, by comparing measurements from multiple sites, would help to understand the heart's mechanical contraction activity. APPROACH: Using newly designed multichannel seismocardiogram equipment, it was possible to investigate the vibration waves with 16 three-axis sensors. The equipment performed well with highly precise synchronization rate over 10 min, linear frequency response and high signal quality. The vibration waves were analyzed using the sagittal axis, a single cardiac cycle and focusing on four fiducial points. Two of the fiducial point where the negative and positive peaks associated with aorta valve opening, along with peaks associated with aorta valve closing. MAIN RESULTS: The respective average centers of mass of the four fiducial points in 13 subjects were at (frontal axis: 35 mm, vertical axis: 5 mm), (31, 6), (26, 24), and (4, -2), relative to the Xiphoid Process. Similar patterns among the subjects were identified for the propagation of the waves across the chest for the four fiducial points. SIGNIFICANCE: The multichannel seismocardiogram equipment successfully revealed a general pattern present in chest surface vibration maps.

Author Correction: Definition of Fiducial Points in the Normal Seismocardiogram.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Contractile Fronts In The Interventricular Septum: A Case For High Frame Rate Echocardiographic Imaging.

The real time high frame rate (HFR) 2-dimensional ultrasound system, T5, at Duke University is capable of imaging at up to 1000 images per second for adult cardiac imaging. A method for detecting and visualizing the mechanical contraction fronts using HFR echocardioagraphy-derived Strain Rate Image (SRI) was described in 26 patients. The Tissue Shortening Onset front durations for echocardiographic normal patients were significantly shorter than conduction disorder patients with left bundle branch block (LBBB) with intrinsic conduction and conduction disorder patients without LBBB (non-LBBB) with simulated LBBB (sLBBB). Echocardiographic normal patients had significantly higher correlation coefficients between their SRIs and spatially inverted versions of themselves compared to non-LBBB patients with intrinsic conduction and sLBBB. In conclusion, SRIs could spatially resolve contractile event fronts in patients.

Long-Term Prognostic Value of Less-Stringent Electrocardiographic Q Waves and Fourth Universal Definition of Myocardial Infarction Q Waves.

BACKGROUND: The Fourth Universal Definition of Myocardial Infarction defines electrocardiographic Q waves as duration ≥30 ms and amplitude ≥1 mm or QS complex in 2 contiguous leads. However, current taskforce criteria may be overly restrictive. Therefore, we investigated the association of isolated, lenient, or strict Q waves with long-term outcome. METHODS: From 2001 to 2015, we included Danish primary care patients with digital electrocardiograms (ECGs) that were evaluated for Q waves. If none occurred, patients had no Q waves. If no other contiguous Q wave occurred, patients had isolated Q waves. If another contiguous Q wave occurred meeting only 1 criterion (≥30 ms and <1 mm or <30 ms and ≥1 mm), patients had lenient Q waves. If another contiguous Q wave occurred, patients had strict Q waves. RESULTS: Of 365,206 patients, 87,957 had isolated, lenient, or strict Q waves (24%; median age, 61 years; male, 48%), and 277,249 had no Q waves (76%; median age, 53 years; male, 42%). Mortality risk was increased with isolated (all-cause adjusted hazard ratio [aHR], 1.33; 95% confidence interval [CI], 1.29-1.37; cardiovascular-cause aHR, 1.78; 95% CI, 1.70-1.87), lenient (all-cause aHR, 1.41; 95% CI, 1.33-1.50; cardiovascular-cause aHR, 1.78; 95% CI, 1.63-1.94), or strict (all-cause aHR, 1.64; 95% CI, 1.57-1.72; cardiovascular-cause aHR, 2.70; 95% CI, 2.52-2.89) Q waves compared with no Q waves. Highest mortality risk was associated with lenient or strict Q waves in anteroseptal leads. CONCLUSIONS: This large contemporary analysis suggests that less-stringent Q-wave criteria carry prognostic value in predicting adverse outcome among primary care patients.

Comparison of Different Methods for Estimating Cardiac Timings: A Comprehensive Multimodal Echocardiography Investigation.

Cardiac time intervals are important hemodynamic indices and provide information about left ventricular performance. Phonocardiography (PCG), impedance cardiography (ICG), and recently, seismocardiography (SCG) have been unobtrusive methods of choice for detection of cardiac time intervals and have potentials to be integrated into wearable devices. The main purpose of this study was to investigate the accuracy and precision of beat-to-beat extraction of cardiac timings from the PCG, ICG and SCG recordings in comparison to multimodal echocardiography (Doppler, TDI, and M-mode) as the gold clinical standard. Recordings were obtained from 86 healthy adults and in total 2,120 cardiac cycles were analyzed. For estimation of the pre-ejection period (PEP), 43% of ICG annotations fell in the corresponding echocardiography ranges while this was 86% for SCG. For estimation of the total systolic time (TST), these numbers were 43, 80, and 90% for ICG, PCG, and SCG, respectively. In summary, SCG and PCG signals provided an acceptable accuracy and precision in estimating cardiac timings, as compared to ICG.

Quantitative Parameters of High-Frame-Rate Strain in Patients with Echocardiographically Normal Function.

Recently, we developed a high-frame-rate echocardiographic imaging system capable of acquiring images at rates up to 2500 per second. High imaging rates were used to quantify longitudinal strain parameters in patients with echocardiographically normal function. These data can serve as a baseline for comparing strain parameters in disease states. The derived timing data also reveal the propagation of mechanical events in the left ventricle throughout the cardiac cycle. High-frame-rate echocardiographic images were acquired from 17 patients in the apical four-chamber view using Duke University's phased array ultrasound system, T5. B-Mode images were acquired at 500-1000 images per second by employing 16:1 or 32:1 parallel processing in receive, a scan depth ≤14 cm and an 80° field of view with a 3.5-MegaHertZ (MHz), 96-element linear array. The images were analyzed using a speckle tracking algorithm tailored for high-frame-rate echocardiographic images developed at Aalborg and Duke University. Four specific mechanical events were defined using strain curves from six regions along the myocardial contour of the left ventricle. The strain curves measure the local deformation events of the myocardium and are independent of the overall cardiac motion. We observed statistically significant differences in the temporal sequence among different myocardial segments for the first mechanical event described, myocardial tissue shortening onset (p < 0.01). We found that the spatial origin of tissue shortening was located near the middle of the interventricular septum in patients with echocardiographically normal function. The quantitative parameters defined here, based on high-speed strain measurements in patients with echocardiographically normal function, can serve as a means of assessing degree of contractile abnormality in the myocardium and enable the identification of contraction propagation. The relative timing pattern among specific events with respect to the Q wave may become an important new metric in assessing cardiac function and may, in turn, improve diagnosis and prognosis.

Definition of Fiducial Points in the Normal Seismocardiogram.

The purpose of this work is to define fiducial points in the seismocardiogram (SCG) and to correlate them with physiological events identified in ultrasound images. For 45 healthy subjects the SCG and the electrocardiogram (ECG) were recorded simultaneously at rest. Immediately following the SCG and ECG recordings ultrasound images of the heart were also obtained at rest. For all subjects a mean SCG signal was calculated and all fiducial points (peaks and valleys) were identified and labeled in the same way across all signals. Eight physiologic events, including the valve openings and closings, were annotated from ultrasound as well and the fiducial points were correlated with those physiologic events. A total of 42 SCG signals were used in the data analysis. The smallest mean differences (±SD) between the eight events found in the ultrasound images and the fiducial points, together with their correlation coefficients (r) were: atrial systolic onset: -2 (±16) ms, r = 0.75 (p < 0.001); peak atrial inflow: 13 (±19) ms, r = 0.63 (p < 0.001); mitral valve closure: 4 (±11) ms, r = 0.71 (p < 0.01); aortic valve opening: -3 (±11) ms, r = 0.60 (p < 0.001); peak systolic inflow: 13 (±23) ms, r = 0.42 (p < 0.01); aortic valve closure: -5 (±12) ms, r = 0.94 (p < 0.001); mitral valve opening: -7 (±19) ms, r = 0.87 (p < 0.001) and peak early ventricular filling: -18 (±28 ms), r = 0.79 (p < 0.001). In conclusion eight physiologic events characterizeing the cardiac cycle, are associated with reproducible, well-defined fiducial points in the SCG.

An Intraneural Electrode for Bioelectronic Medicines for Treatment of Hypertension.

OBJECTIVES: As the left vagus nerve (LVN) mediates a baroreflex blood pressure (BP) decrease, LVN stimulation (LVNS) could be a therapy for hypertension. Moreover, LVNS could elegantly be adjusted to the patient's actual BP and physical activity by using the neural information about BP and respiration extractable from LVN. However, unselective LVNS will trigger undesirable side-effects and therefore we here investigated the feasibility of using an intraneural electrode for extracting BP and respiration markers from the LVN and for selective LVNS. MATERIALS AND METHODS: Experiments were performed on six anesthetized pigs from which the BP was recorded using arterial catheters and the respiratory cycles by recording the airway pressure. An electrode comprising four tripolar channels was inserted longitudinally in the LVN of the animals to extract BP and respiration markers from the LVN and for selective LVNS. RESULTS: BP-related and respiratory-related neural profiles (BPnPs and RnPs, respectively) were derived from at least two electrode channels in all pigs. The BPnPs accurately resembled the BP waves and the RnPs accurately resembled the respiratory cycles, which suggests that those profiles could serve as BP and respiration markers, respectively. The BP was decreased by intraneural LVNS in all pigs and in four of those pigs such an effect was induced without major cardiac changes through a channel-selective stimulation. CONCLUSION: This study shows that it is feasible to extract BP and respiratory markers from the LVN with the tested intraneural electrode and suggests that this electrode could also be used for selective LVNS.

A History of Drug-Induced Torsades de Pointes Is Associated With T-wave Morphological Abnormalities.

The hypothesis of the study is that Torsades de pointes (TdP) history can be better identified using T-wave morphology compared to Fridericia-corrected QT interval (QTcF) at baseline. ECGs were recorded at baseline and during sotalol challenge in 20 patients with a history of TdP (+TdP) and 16 patients without previous TdP (-TdP). The QTcF and T-wave morphology combination score (MCS) were calculated. At baseline, there was no significant difference in QTcF between the groups (+TdP: QTcF = 446 ± 9 ms; -TdP: QTcF = 431 ± 9 ms, P = 0.27). In contrast, MCS was significantly different between the groups at baseline (+TdP: MCS = 1.07 ± 0.095; -TdP: MCS = 0.74 ± 0.07, P = 0.012). Both QTcF and MCS could be used to discriminate between +TdP and -TdP after sotalol but only MCS reached statistical significance at baseline. Combining QTcF with MCS provided a significantly larger difference between groups than QTcF alone.

A Respiratory Marker Derived From Left Vagus Nerve Signals Recorded With Implantable Cuff Electrodes.

OBJECTIVE: Left vagus nerve (LVN) stimulation (LVNS) has been tested for lowering the blood pressure (BP) in patients with resistant hypertension (RH). Whereas, closed-loop LVNS (CL-LVNS) driven by a BP marker may be superior to open-loop LVNS, there are situations (e.g., exercising) when hypertension is normal. Therefore, an ideal anti-RH CL-LVNS system requires a variable to avoid stimulation in such conditions, for example, a respiratory marker ideally extracted from the LVN. As the LVN conducts respiratory signals, this study aimed to investigate if such signals can be recorded using implantable means and if a marker to monitor respiration could be derived from such recordings. MATERIALS AND METHODS: The experiments were performed in 14 anesthetized pigs. Five pigs were subjected to changes of the respiratory frequency and nine to changes of the respiratory volume. The LVN electroneurogram (VENG) was recorded using two cuff electrodes and the respiratory cycles (RC) using a pressure transducer. To separate the afferent and efferent VENGs, vagotomy was performed between the cuffs in the first group of pigs. The VENG was squared to derive respiration-related neural profiles (RnPs) and their correlation with the RCs was investigated in regard to timing and magnitude parameters derived from the two waveforms. RESULTS: The RnPs were morphologically similar with the RCs and the average RnPs represented accurate copies of the average RCs. Consequently, the lung inflation/deflation RC and RnP components had the same duration, the respiratory frequency changes affected in the same way both waveforms and the RnP amplitude increased linearly with the lung inflation in all tested pigs (R2 values between 0.85 and 0.99). CONCLUSIONS: The RnPs comprise information regarding the timing and magnitude of the respiratory parameters. As those LVN profiles were derived using implantable means, this study indicates that the RnPs could serve as respiratory markers in implantable systems.

Association between maternal vascular murmur and the small-for-gestational-age fetus with abnormal umbilical artery Doppler flow.

OBJECTIVE: To investigate the association between maternal vascular murmurs (MVMs) and fetal growth restriction (defined as small-for-gestational-age [SGA] fetus) and abnormal Doppler pulsatility index (PI) of the uterine and/or umbilical arteries. METHODS: A cross-sectional study of women aged 18 years or older with a singleton pregnancy at 28-34 weeks was conducted at Regional Hospital Viborg, Denmark, between May 1 and August 1, 2013. Ultrasound fetal biometry was performed and the Doppler PI of the umbilical and uterine arteries was determined. An estimated fetal weight (EFW) at or below the 10th percentile was defined as SGA. Microphone recordings from the lower abdomen were divided into heart valve sounds and MVMs. RESULTS: The final analysis included 63 participants, with 25 classified as SGA and 38 as non-SGA. The mean pregnancy duration was 32.4 ± 1.4 weeks. In total, 17 participants had MVMs. There was a clear association between MVMs and a composite of SGA and an abnormal PI of the uterine and/or the umbilical artery (P<0.001), but not between MVMs and SGA only (P=0.154). CONCLUSION: Maternal vascular murmurs are significantly associated with fetal growth restriction, but not with SGA per se.

A neural blood pressure marker for bioelectronic medicines for treatment of hypertension.

A novel therapeutic approach for treating resistant hypertension could be the use of bioelectronic medicines to achieve blood pressure (BP) control in implanted patients by closed-loop stimulation of the left vagus nerve (LVN). However, such a technology would require an implantable BP marker, which is not available yet. As it is known that the LVN conducts afferent BP-related signals, this study aimed to investigate if such signals could be recorded with implantable means and used to derive BP markers. The present experiments were performed in anesthetized pigs subjected to a transient or stable BP increase induced by adrenaline administration. The LVN signals recorded with cuff electrodes and the BP waves recorded with carotid catheters were ensemble averaged relative to the R-peaks of the electrocardiogram. Through this procedure, afferent BP-related neural profiles (BPnPs) were derived from the LVN signals. As the BPnPs represented accurate copies of the BP waves, the timing parameters of both waveforms were almost the same and the BPnP amplitude increased linearly with the diastolic, systolic and mean BP. These results indicate that the BPnPs comprise accurate BP information and that the BPnP amplitude could serve as a BP marker in implantable systems.

Separation and characterization of maternal cardiac and vascular sounds in the third trimester of pregnancy.

OBJECTIVE: To characterize the vascular sounds of the uteroplacental blood flow obtained by microphones. METHODS: The present retrospective study took place in an anechoic chamber facility at Aalborg University, Aalborg, Denmark, in 2012, and included pregnant participants aged 18-40 years with a singleton pregnancy at 32-36 weeks and a parity of 0-2. Abdominal Doppler ultrasonography was performed bilaterally on the uterine arteries. Subsequently, in the same positions, sound recordings were performed with microphones. The derived raw sound signal was separated into two frequency ranges, and characterized accordingly. RESULTS: The mean pregnancy length among 25 participants was 33.6 ± 2.0 weeks. The pulsatility index of the uterine artery was 0.67 ± 0.24. All 50 recordings displayed the first and second maternal heart sounds (frequency 25-100 Hz), and in 17 of 50 recordings, maternal vascular murmurs (frequency 200-800 Hz) were present. The average pulse wave velocity between the maternal aortic valve and the uterine artery was estimated to be 6.6 ± 1.5 m/s. CONCLUSION: Maternal vascular murmurs in the frequency range of 200-800 Hz were identified as a possible marker of abnormal uteroplacental blood flow, and provide a means to measure the arterial pulse wave velocity.

Minimizing a Wireless Passive LC-Tank Sensor to Monitor Bladder Pressure: A Simulation Study.

In this simulation study, a wireless passive LC-tank sensor system was characterized. Given the application of continuous bladder monitoring, a specific system was proposed in terms of coil geometries and electronic circuitry. Coupling coefficients were spatially mapped by simulation, as a function of both coil distance, and longitudinal and transverse translation of the sensor relative to the antenna. Further, two interrogation schemes were outlined. One was an auto-balancing bridge for computing the sensor-system impedance. In this case, the theoretical noise limit of the analogue part of the system was found by simulations. As the full system is not necessary for obtaining a pressure reading from the sensor, a simplified circuit more suited for an implantable system was deduced. For this system, both the analogue and digital parts were simulated. First, the required ADC resolution for operating the system at a given coupling was found by simulations in the noise-free case. Then, for one selected typical operational point, noise was added gradually, and through Monte-Carlo type simulations, the system performance was obtained. Combining these results, it was found that it at least is possible to operate the proposed system for distances up to 12 mm, or equivalently for coupling coefficients above 0.005. In this case a 14 bit ADC is required, and a carrier SNR of 27 dB can be tolerated.

Minimal T-wave representation and its use in the assessment of drug arrhythmogenicity.

BACKGROUND: Recently, numerous models and techniques have been developed for analyzing and extracting features from the T wave which could be used as biomarkers for drug-induced abnormalities. The majority of these techniques and algorithms use features that determine readily apparent characteristics of the T wave, such as duration, area, amplitude, and slopes. METHODS: In the present work the T wave was down-sampled to a minimal rate, such that a good reconstruction was still possible. The entire T wave was then used as a feature vector to assess drug-induced repolarization effects. The ability of the samples or combinations of samples obtained from the minimal T-wave representation to correctly classify a group of subjects before and after receiving d,l-sotalol 160 mg and 320 mg was evaluated using a linear discriminant analysis (LDA). RESULTS: The results showed that a combination of eight samples from the minimal T-wave representation can be used to identify normal from abnormal repolarization significantly better compared to the heart rate-corrected QT interval (QTc). It was further indicated that the interval from the peak of the T wave to the end of the T wave (Tpe) becomes relatively shorter after IKr inhibition by d,l-sotalol and that the most pronounced repolarization changes were present in the ascending segment of the minimal T-wave representation. CONCLUSIONS: The minimal T-wave representation can potentially be used as a new tool to identify normal from abnormal repolarization in drug safety studies.

Changes in vagus nerve activity associated with ictal tachycardia in pigs.

OBJECTIVE: Ictal tachycardia (IT) is common and may pave the way towards cardiac conditions with high risk potential. However, the mechanisms of IT remain obscure and therefore difficult to control. For example, whereas IT is associated with a sympathetic surge, it is unclear why the IT effects are not opposed by baroreflex cardiac inhibition during seizures. As the vagus nerves (VN) are main mediators for such baroreflexes, this study was performed to investigate the VN activity in IT. METHODS: The present experiments were performed in ten pigs where IT seizures were induced by controlled infusion of pentylenetetrazole. The electrocorticogram was recorded using a cranial electrode, the electrocardiogram (ECG) using surface electrodes and the blood pressure (BP) using a catheter inserted in the right carotid artery. The VN activity was recorded from both nerves using cuff electrodes and further analyzed in correlation with the cortical seizures and the associated heart rate (HR), BP and HR variability (HRV) changes. RESULTS: The cortical seizures progressed from spike-and-wave (SW) to tonic-clonic (TC) discharges associated with ECG, HR and BP changes proportional with this progression and comparable to the IT effects reported in humans. Those IT effects were accompanied by parasympathetic HRV changes, a 20% VN activation (p=0.004) before the onset of TC seizures, a suppression of this VN activation during the TC episode and a rebound VN activation by 79% (left VN, p=0.02) and 57% (right VN, p=0.03) after the TC offset. Further analysis of an afferent BP-related VN component and a mixed VN component showed normal BP-related afferent input and a suppressed efferent output through both nerves during the TC episode. CONCLUSIONS: This study indicates a suppressed ictal VN activation and a rebound postictal VN activation, which may account for the absence of baroreflexes during seizures and the postictal cardiac inhibition, respectively.

Association Between Heart Rate at Rest and Incident Atrial Fibrillation (from the Copenhagen Electrocardiographic Study).

Heart rate (HR) at rest is a well-known marker of cardiovascular morbidity and mortality. Results on the association between HR and incident atrial fibrillation (AF) have, however, been conflicting. Using digital electrocardiograms from 281,451 primary care patients, we aimed to describe the association between HR at rest and the hazards of incident AF. Secondary end points were death from all causes and pacemaker implantation. Data on drug use, co-morbidity, and outcomes were collected from nationwide administrative health care registries. During a median follow-up time of 8.4 years, 15,666 subjects were observed to develop AF, of which 1,631 were lone AF. A HR at rest from 30 to 51 beats/min was associated with an adjusted hazard ratio of 1.16 (95% CI 1.06 to 1.27) for AF compared with the reference group (66 to 72 beats/min). From 72 beats/min and upward, the hazard ratio of AF increased in a dose-response manner, reaching an adjusted hazard ratio of 1.36 (95% CI 1.26 to 1.46) for HR between 95 and 120 beats/min. Both for low and high HR, the associations were accentuated for the outcome lone AF (adjusted hazard ratios of 1.48, 95% CI 1.19 to 1.84 and 1.84, 95% CI 1.47 to 2.30 for HR between 30 to 51 and 95 to 120 beats/min, respectively). For death from all causes, the hazard increased almost linearly with increasing HR. A HR at rest from 30 to 51 beats/min was associated with an adjusted hazard ratio of 1.80 (95% CI 1.46 to 2.21) for pacemaker implantation. In conclusion, a U-shaped association was found between HR at rest and incident AF, and this association was strongest for the outcome lone AF.

Major rapid weight loss induces changes in cardiac repolarization.

INTRODUCTION: Obesity is associated with increased all-cause mortality, but weight loss may not decrease cardiovascular events. In fact, very low calorie diets have been linked to arrhythmias and sudden death. The QT interval is the standard marker for cardiac repolarization, but T-wave morphology analysis has been suggested as a more sensitive method to identify changes in cardiac repolarization. We examined the effect of a major and rapid weight loss on T-wave morphology. METHODS AND RESULTS: Twenty-six individuals had electrocardiograms (ECG) taken before and after eight weeks of weight loss intervention along with plasma measurements of fasting glucose, HbA1c, and potassium. For assessment of cardiac repolarization changes, T-wave Morphology Combination Score (MCS) and ECG intervals: RR, PR, QT, QTcF (Fridericia-corrected QT-interval), and QRS duration were derived. The participants lost on average 13.4% of their bodyweight. MCS, QRS, and RR intervals increased at week 8 (p<0.01), while QTcF and PR intervals were unaffected. Fasting plasma glucose (p<0.001) and HbA1c both decreased at week 8 (p<10(-5)), while plasma potassium was unchanged. MCS but not QTcF was negatively correlated with HbA1c (p<0.001) and fasting plasma glucose (p<0.01). CONCLUSION: Rapid weight loss induces changes in cardiac repolarization. Monitoring of MCS during calorie restriction makes it possible to detect repolarization changes with higher discriminative power than the QT-interval during major rapid weight loss interventions. MCS was correlated with decreased HbA1c. Thus, sustained low blood glucose levels may contribute to repolarization changes.

Electrocardiographic Tpeak-Tend interval and risk of cardiovascular morbidity and mortality: Results from the Copenhagen ECG study.

BACKGROUND: The electrocardiographic Tpeak-Tend interval is considered a novel risk marker of cardiac arrhythmias and cardiovascular death; however, results to date have been conflicting. OBJECTIVE: The purpose of this study was to investigate the association between this interval and the risk of all-cause and cardiovascular mortality, atrial fibrillation, and heart failure, allowing for nonlinear relationships. METHODS: From primary care, 138,404 individuals were included and categorized into seven groups based on Tpeak-Tend interval. Cox regression models were used to describe the association between these groups and the risk of the selected outcomes. RESULTS: Compared with the reference groups (104-115 ms for all-cause mortality and 98-103 ms for all other outcomes), individuals with a Tpeak-Tend interval in lead V5 <5th percentile (58-77 ms) had hazard ratios of 1.29 (95% confidence interval [CI] 1.21-1.38, P <.001) for all-cause mortality, 1.31 (95% CI 1.15-1.50, P <.001) for cardiovascular death, 1.18 (95% CI 1.06-1.32, P = .003) for atrial fibrillation, and 1.52 (95% CI 1.33-1.74, P <.001) for heart failure. Individuals with a Tpeak-Tend interval ≥95th percentile (116-140 ms) had hazard ratios of 1.15 (95% CI 1.08-1.23, P <.001) for all-cause mortality, 1.30 (95% CI 1.15-1.47, P <.001) for cardiovascular death, 1.09 (95% CI 0.99-1.22, P = .088) for atrial fibrillation, and 1.28 (95% CI 1.12-1.46, P <.001) for heart failure. Similar results were obtained for leads II and V2. CONCLUSION: We observed U-shaped associations between the Tpeak-Tend interval and risk of all-cause and cardiovascular mortality, atrial fibrillation, and heart failure.