Citations

Background: Sex and age have long been known to affect the ECG. Several biologic variables and anatomic factors may contribute to sex and age-related differences on the ECG. We hypothesized that a convolutional neural network (CNN) could be trained through a process called deep learning to predict a person’s age and self-reported sex using only 12-lead ECG signals. We further hypothesized that discrepancies between CNN-predicted age and chronological age may serve as a physiological measure of health. Methods: We trained CNNs using 10-second samples of 12-lead ECG signals from 499 727 patients to predict sex and age. The networks were tested on a separate cohort of 275 056 patients. Subsequently, 100 randomly selected patients with multiple ECGs over the course of decades were identified to assess within-individual accuracy of CNN age estimation. Results: Of 275 056 patients tested, 52% were males and mean age was 58.6±16.2 years. For sex classification, the model obtained 90.4% classification accuracy with an area under the curve of 0.97 in the independent test data. Age was estimated as a continuous variable with an average error of 6.9±5.6 years (R-squared =0.7). Among 100 patients with multiple ECGs over the course of at least 2 decades of life, most patients (51%) had an average error between real age and CNN-predicted age of <7 years. Major factors seen among patients with a CNN-predicted age that exceeded chronologic age by >7 years included: low ejection fraction, hypertension, and coronary disease ( P <0.01). In the 27% of patients where correlation was >0.8 between CNN-predicted and chronologic age, no incident events occurred over follow-up (33±12 years). Conclusions: Applying artificial intelligence to the ECG allows prediction of patient sex and estimation of age. The ability of an artificial intelligence algorithm to determine physiological age, with further validation, may serve as a measure of overall health.

AbstractBackgroundThere are few data regarding the outcome of a combined cryo‐ and radiofrequency (RF)‐catheter ablation of various types of supraventricular tachycardias (SVTs) originating from near the normal conduction system.METHODSWe analyzed all patients undergoing combined cryo‐ and RF‐ catheter ablation at Mayo Clinic, Rochester, MN as part of the ablation of SVTs with potential risks of injury to the normal conduction system. This study aimed to assess the outcome of a combined cryo‐ and RF‐catheter ablation of various types of SVTs.RESULTSThe study population consisted of 54 patients (38 ± 17 years, 32 men). A combined cryo‐ and RF‐catheter ablation was attempted for septal accessary pathways (APs) in 26, atrioventricular nodal reentrant tachycardia (AVNRT) in 14, atrial tachycardia (AT) in 7, and junctional ectopic tachycardia in 7 patients. Forty‐one patients (76%) were successfully ablated with cryoablation, and RF ablation after an unsuccessful cryoablation ablated the SVTs successfully at the same location in 6 patients (11%). Complication occurred in 1 patient (deep vein thrombosis). The cumulative freedom from SVT rate at 30 days after the procedure was 78% and there was no significant difference in the recurrence rate among the SVTs.CONCLUSIONA combined cryo‐ and RF‐catheter ablation is clinically effective in patients undergoing SVT ablation with the potential risk of injury to the normal conduction system.

AbstractIntroductionAutonomic modulation has been used as a therapy to control recurrent ventricular arrhythmia (VA). This study was to explore stellate ganglion block (SGB) effect on cardiac electrophysiologic properties and evaluate the long‐term outcome of cardiac sympathetic denervation (CSD) for patients with recurrent VA and structural heart disease (SHD).Materials and MethodsPatients who had recurrent VA due to SHD were enrolled prospectively. Electrophysiologic study and ventricular tachycardia (VT) induction were performed before and after left and right SGB. VA burden and long‐term outcomes were assessed for a separate patient group who underwent left or bilateral CSD for drug‐refractory VA due to SHD.ResultsElectrophysiologic study of nine patients showed that baseline mean (SD) corrected sinus node recovery time (cSNRT) increased from 320.4 (73.3) ms to 402.9 (114.2) ms after left and 482.4 (95.7) ms after bilateral SGB (P = .03). SGB did not significantly change P‐R, QRS, and Q‐T intervals and ventricular effective refractory period, nor did the inducibility of VA. Nineteen patients underwent left (n = 14) or bilateral (n = 5) CSD. CSD reduced VA burden and appropriate ICD therapies from a median (interquartile range) of 2.5 (0.4‐11.6) episodes weekly to 0.1 (0.0‐2.4) episodes weekly at 6‐month follow‐up (P = .002). Three‐year freedom from orthotopic heart transplant (OHT) and death was 52.6%. New York Heart Association functional class III/IV and VT rate less than 160 beats per minute were predictors of recurrent VA, OHT, and death.ConclusionSGB increased cSNRT without changing heart rate. CSD was more beneficial for patients with mild‐to‐moderate heart failure and faster VA.

AbstractIntroductionIt is unknown if ANO‐1 is expressed in the heart, though the presence of a calcium‐activated chloride current has been proposed to mediate some cardiac dysrhythmias. Furthermore, a specific cell type termed telocytes, morphologically mimicking Cajal cells which use ANO‐1 to modulate their pacemaker activity in the gut, have been described in the heart. We therefore sought to determine whether this channel is expressed in the canine heart.MethodsMyocardium was sampled from the ventricles of five canines. Sections were labeled with anti‐Kit and anti‐ANO‐1 antibodies. Slides were reviewed by four investigators looking at cell morphology, distribution, and co‐localization. Identification of telocytes was based on criteria including morphology, Kit positivity (+), and ANO‐1 positivity (+).ResultsClusters of cells meeting criteria for telocytes were seen in the epicardium, sub‐epicardium, and mid‐myocardium. A small subset of cells that were morphologically similar to myocytes was ANO‐1 (+) but Kit (−). In total, three different cell classes were found: (i) Kit (+), ANO‐1 (+) cells with the appearance of telocytes; (ii) Kit (+), ANO‐1 (−) cells; and (iii) Kit (−), ANO‐1 (+) cells with the morphologic appearance of cardiac myocytes.ConclusionsTelocytes are present in the canine ventricle and express ANO‐1. These data merit further study to elucidate the functional expression of these channels in the heart and whether they may be targets for cardiac arrhythmias.

Background: Reentrant ventricular arrhythmias are a major cause of sudden death in patients with structural heart disease. Current treatments focus on electrically homogenizing regions of scar contributing to ventricular arrhythmia with ablation or altering conductive properties using antiarrhythmic drugs. The high conductivity of carbon nanotubes may allow restoration of conduction in regions where impaired electrical conduction results in functional abnormalities. We propose a new concept for arrhythmia treatment using a stretchable, flexible biopatch with conductive properties to attempt to restore conduction across regions in which activation is disrupted. Methods: Carbon nanotube patches composed of nanofibrillated cellulose/single-walled carbon nanotube ink 3-dimensionally printed in conductive patterns onto bacterial nanocellulose were developed and evaluated for conductivity, flexibility, and mechanical properties. The patches were applied on 6 canines to epicardium before and after surgical disruption. Electroanatomic mapping was performed on normal epicardium, then repeated over surgically disrupted epicardium, and then finally with the patch applied passively. Results: We developed a 3-dimensional printable carbon nanotube ink complexed on bacterial nanocellulose that was (1) expressable through 3-dimensional printer nozzles, (2) electrically conductive, (3) flexible, and (4) stretchable. Six canines underwent thoracotomy, and, during epicardial ventricular pacing, mapping was performed. We demonstrated disruption of conduction after surgical incision in all 6 canines based on activation mapping. The patch resulted in restored conduction based on mapping and assessment of conduction direction and velocities in all canines. Conclusions: We have demonstrated 3-dimensional custom-printed electrically conductive carbon nanotube patches can be surgically manipulated to improve cardiac conduction when passively applied to surgically disrupted epicardial myocardium in canines.

AbstractObjectivesWe sought to validate a deep learning algorithm designed to predict an ejection fraction (EF) less than or equal to 35% based on the 12‐lead electrocardiogram (ECG) in a large prospective cohort.BackgroundPatients undergoing routine ECG may have undetected left ventricular (LV) dysfunction that warrants further echocardiographic assessment. However, identification of these patients can be challenging.MethodsWe applied the algorithm to all ECGs interpreted by the Mayo Clinic ECG laboratory in September 2018. The performance of the algorithm was tested among patients with recent echocardiographic assessments of LV function. We also applied the algorithm in patients with no recent echocardiographic assessments of LV function to determine the rate of new “positive screens.”ResultsAmong 16 056 adult patients who underwent routine ECG, 8600 (age 67.1 ± 15.2 years, 45.6% male), had a transthoracic echocardiogram (TTE) and 3874 patients had a TTE and ECG less than 1 month apart. Among these patients, the algorithm was able to detect an EF less than or equal to 35% with 86.8% specificity, 82.5% sensitivity, and 86.5% accuracy, (area under the curve, 0.918). Among 474 “false‐positives screens,” 189 (39.8%) had an EF of 36% to 50%. Among patients with no prior TTE, the algorithm identified 3.5% of the patients with suspected EF less than or equal to 35%. Exploratory analysis suggests false positives could be reduced by assessing NT‐pro‐BNP after the initial “positive screen.”ConclusionsA deep learning algorithm detected depressed LV function with good accuracy in routine practice. Further studies are needed to validate the algorithm in patients with no prior echocardiogram and to assess the impact on echocardiography utilization, cost, and clinical outcomes.

AbstractBackgroundImplantable loop recorders (ILRs) are effective in achieving symptom‐rhythm correlation. However, diagnostic yield in routine clinical practice is not well established.MethodsPatients undergoing ILR implantation between April 2010 and May 2015 were included. All devices were enrolled in remote monitoring with automatic arrhythmia detection and P sense algorithms switched “ON.” Symptom‐rhythm correlation was assessed and changes in management were recorded.ResultsA total of 312 patients (57% male, age 53 ± 22 years; median CHADS2VaSc score = 1) were included in this study. ILRs were implanted for evaluation of syncope in 206 (66.0%), presyncope in 23 (7.4%), unexplained palpitations in 51 (16.3%), and cryptogenic stroke in 27 (8.7%) patients. ILR monitoring yielded a diagnosis that changed management strategy in 146 (46.8%) patients over a median of 12 (1‐42) months. Out of 163 (52.2%) patients with symptoms during the monitoring period, 100 (61.3%) had an arrhythmia. ILR was useful in ruling out an arrhythmic cause for symptoms in 63 (38.7%) patients. ILR results led to pacemaker implantation in 23 patients (7.4% overall and 11.2% of those with syncope) after median follow‐up of 3 months. A new diagnosis of atrial fibrillation was made in 38 (12.2%) patients, 11 of whom were initiated on oral anticoagulants. ILR results led to pacemaker implantation in 31 patients (9.9% overall and 19.0% of those with syncope) after median follow‐up of 3 months. A new diagnosis of atrial fibrillation was made in 38 (12.2%) patients, nine of whom were initiated on oral anticoagulants. Overall, ILR led to a change in management in 47% patients with a number needed to implant of 2.1 to change management.ConclusionILR monitoring is effective in achieving symptom‐rhythm correlation and results in changes in management in nearly half of implanted patients. Additional studies are needed to evaluate cost efficacy of ILR and the optimal monitoring duration.

AbstractIntroductionCatheter ablation (CA) has emerged as the preferred modality of treatment for many cardiac arrhythmias. Anatomical sites of ablation are often located in close proximity to coronary arteries. However, the incidence of CA‐related coronary injury has not been well studied. We sought to systematically evaluate all cases of CA‐related coronary injuries.Methods and ResultsA PubMed search was conducted from inception until May 1, 2017 using the keywords “coronary artery” and “ablation.” We identified 2817 published articles of which 43 articles met our inclusion criteria representing 61 cases of coronary artery injury attributed to CA procedures from 1992 to 2017. Posteroseptal accessory pathway ablation was associated with the highest incidence of coronary injury (35.6% of cases), followed by cavotricuspid isthmus‐dependent flutter (19.3%). The right coronary artery was the site of injury in over two‐thirds of all reported cases. Coronary injury was detected intraprocedurally in about half of the cases (43.1%), whereas it was a delayed presentation in the other half. Coronary intervention was performed in a third of all cases (32.7%). There were a total of three deaths attributed to coronary artery injury.ConclusionsMost (91.8%) coronary injuries are a result of anatomic proximity to the site of ablation. Awareness of the relation between coronary artery course and anatomical site of ablation could prevent myocardial damage and improve procedural safety.