Circulation: Arrhythmia and Electrophysiology March 2020 Issue

Paul J. Wang:

Welcome to the monthly podcast On the Beat for circulation, arrhythmia, and electrophysiology. I'm Dr. Paul Wang, Editor In Chief, with some of the key highlights from this month's issue.

 

Elizabeth Wang and Associates examined the relationship between acute precipitants of atrial fibrillation and long-term recurrence of atrial fibrillation, AF, from a multi-institutional, longitudinal electronic medical record database. Among 10,723 patients with newly diagnosed Afib, age 67.9 years, 41% women, the authors found that 19% had an acute AF precipitant, the most common of which were cardiac surgery in 22%, pneumonia in 20% and non-cardiothoracic surgery in 15%. The cumulative incidence of AF recurrence at five years was 41% among individuals with a precipitant, compared to 52% in those without a precipitant. Adjusted hazard ratio 0.75 P < 0.001. The lowest risk of recurrence among those with precipitants with postoperative atrial fibrillation, five-year incidence 32% in cardiac surgery and 39% in non-cardiothoracic surgery. Regardless of the initial precipitant, recurrent atrial fibrillation was associated with an increased adjusted risk of heart failure, hazard ratio of 2.74 P < 0.001, Stroke, hazard ratio 1.57 P < 0.001 and mortality, hazard ratio 2.96 P < 0.001. Thus, the authors found that atrial fibrillation after acute precipitant frequently recurs and the recurrence is associated with substantial long-term morbidity and mortality.

 

In the next paper, Jacob Koruth and associates examine the effect of pulse field ablation on the esophagus in a novel in-vivo porcine esophageal injury model. The authors studied 10 animals under general anesthesia while the lower esophagus was deflected towards the inferior vena cava using an esophageal deviation balloon and ablation was formed from within the inferior vena cava at areas of esophageal contact. Six animals received eight pulse field ablation applications per site and four animals received six clusters of irrigated radio frequency ablation applications at 30 Watts for 30 seconds. All animals survived to 25 days, sacrificed, and the esophagus was submitted for a pathological examination including 10 discreet histological sections of the esophagus.

The authors found that zero out of six pulse field ablation animals demonstrated esophageal lesions while esophageal injury occurred in all four radio frequency ablation animals, P = 0.005. A mean of 1.5 mucosal lesions per animal, length 21.8 millimeters with 4.9 millimeters were observed, including one esophageal pulmonary fistula, and deep esophageal ulcers in the other animals. Histological examination demonstrated tissue necrosis surrounded by an acute and chronic inflammation and fibrosis. The necrotic radio frequency ablation lesions involved multiple esophageal tissue layers with evidence of arteriolar medial thickening and fibrosis of peri-esophageal nerves, abscess formation and full thickness esophageal wall disruption were seen in the areas of perforation or fistula.

 

In our next paper, Peter Noseworthy and associates examine whether the ability of deep learning algorithms to detect low left ventricular ejection fraction using the 12 lead electrocardiogram varies by race or ethnicity. The authors used a retrospective cohort analysis and included 97,829 patients with paired electrocardiograms and echocardiograms and used a convolutional neural network to identify patients with a left ventricular ejection fraction less than or equal to 35% from the 12 lead electrocardiogram. The convolutional neural network was previously derived in a homogeneous population, 96.2% non Hispanic white, N = 44,959 which demonstrated consistent performance to detect low left ventricular ejection fraction across a range of racial ethnic subgroups in a separate cohort of 52,870 patients (Non-Hispanic white 44,524 patients with an AUC of 0.93; Asian 557 with an AUC of 0.96; Black/African American N = 651 with an AUC of 0.937; in Hispanic/Latino N = 331 AUC of 0.937; in Native American/Alaskan N = 223 AUC of 0.938).

In secondary analysis, a separate neural network was able to discern racial subgroup category, Black/African American AUC 0.84 and white non-Hispanic AUC 0.75 in a five-class classifier. In a network trained only in non-Hispanic whites, from the original derivation cohort, performed similarly well across a range of racial ethnic subgroups in the testing cohort with at least an AUC of 0.93 in all racial ethnic subgroups. The authors concluded that while ECG characteristics vary by race, this did not impact the ability of a convolutional neural network to predict low left ventricular ejection fraction from the ECGs. They recommend reporting of performance against diverse ethnic, racial, age, and gender groups for all new artificial intelligent tools.

 

In our next paper, Benjamin Shoemaker and associates examine the association