Machine Learning in Electrocardiography and Echocardiography: Technological Advances in Clinical Cardiology.

PURPOSE OF REVIEW: Electrocardiography (ECG) and echocardiography are the most widely used diagnostic tools in clinical cardiology. This review focuses on recent advancements in applying machine learning (ML) in ECG and echocardiography and potential synergistic ML integration of ECG and echocardiography. RECENT FINDINGS: ML algorithms have been used in ECG for technical quality assurance, arrhythmia identification, and prognostic predictions, and in echocardiography to recognize image views, quantify measurements, and identify pathologic patterns. Synergistic application of ML in ECG and echocardiograph has demonstrated the potential to optimize therapeutic response, improve risk stratification, and generate new disease classification. There is mounting evidence that ML potentially outperforms in disease diagnoses and outcome prediction with ECG and echocardiography when compared with trained healthcare professionals. The applications of ML in ECG and echocardiography are playing increasingly greater roles in medical research and clinical practice, particularly for their contributions to developing novel diagnostic/prognostic prediction models. The automation in data acquisition, processing, and interpretation help streamline the workflows of ECG and echocardiography in contemporary cardiology practice.

Erratum: Health disparities and treatment approaches in portopulmonary hypertension and idiopathic pulmonary arterial hypertension: An analysis of the Pulmonary Hypertension Association Registry.

[This corrects the article DOI: 10.1177/20458940211020913.].

Mortality Correlates in Patients With Takotsubo Syndrome During the COVID-19 Pandemic.

We completed a systematic review of Takotsubo syndrome (TTS) cases reported during the coronavirus disease 2019 (COVID-19) pandemic and performed clustering and feature importance analysis and statistical testing for independence on the demographic, clinical, and imaging parameters. Compared with the data before the COVID-19 pandemic, TTS was increasingly diagnosed in physical stress (mostly COVID-19 pneumonia)-triggered male patients without psychiatric/neurologic disorders, warranting further investigation to establish new reference criteria to improve diagnostic specificity. In clustering analysis, sex and inpatient mortality primarily contributed to the automated classification of the TTS. Both sex and inpatient mortality had essential correlations with COVID-19 infection/pneumonia. There is effect modification of sex on outcomes in patients with COVID-19 infection and TTS, with male patients having significantly worse inpatient mortality. Meanwhile, significantly more male patients with TTS were classified as high risk according to International Takotsubo Registry prognostic scores, suggesting that male COVID-19/TTS survivors will likely have worse long-term outcome.

Spatio-temporal hybrid neural networks reduce erroneous human "judgement calls" in the diagnosis of Takotsubo syndrome.

BACKGROUND: We investigate whether deep learning (DL) neural networks can reduce erroneous human "judgment calls" on bedside echocardiograms and help distinguish Takotsubo syndrome (TTS) from anterior wall ST segment elevation myocardial infarction (STEMI). METHODS: We developed a single-channel (DCNN[2D SCI]), a multi-channel (DCNN[2D MCI]), and a 3-dimensional (DCNN[2D+t]) deep convolution neural network, and a recurrent neural network (RNN) based on 17,280 still-frame images and 540 videos from 2-dimensional echocardiograms in 10 years (1 January 2008 to 1 January 2018) retrospective cohort in University of Iowa (UI) and eight other medical centers. Echocardiograms from 450 UI patients were randomly divided into training and testing sets for internal training, testing, and model construction. Echocardiograms of 90 patients from the other medical centers were used for external validation to evaluate the model generalizability. A total of 49 board-certified human readers performed human-side classification on the same echocardiography dataset to compare the diagnostic performance and help data visualization. FINDINGS: The DCNN (2D SCI), DCNN (2D MCI), DCNN(2D+t), and RNN models established based on UI dataset for TTS versus STEMI prediction showed mean diagnostic accuracy 73%, 75%, 80%, and 75% respectively, and mean diagnostic accuracy of 74%, 74%, 77%, and 73%, respectively, on the external validation. DCNN(2D+t) (area under the curve [AUC] 0·787 vs. 0·699, P = 0·015) and RNN models (AUC 0·774 vs. 0·699, P = 0·033) outperformed human readers in differentiating TTS and STEMI by reducing human erroneous judgement calls on TTS. INTERPRETATION: Spatio-temporal hybrid DL neural networks reduce erroneous human "judgement calls" in distinguishing TTS from anterior wall STEMI based on bedside echocardiographic videos. FUNDING: University of Iowa Obermann Center for Advanced Studies Interdisciplinary Research Grant, and Institute for Clinical and Translational Science Grant. National Institutes of Health Award (1R01EB025018-01).

Health disparities and treatment approaches in portopulmonary hypertension and idiopathic pulmonary arterial hypertension: an analysis of the Pulmonary Hypertension Association Registry.

Compared to idiopathic pulmonary arterial hypertension (IPAH), patients with portopulmonary hypertension (POPH) have worse survival. Health disparities may contribute to these differences but have not been studied. We sought to compare socioeconomic factors in patients with POPH and IPAH and to determine whether socioeconomic status and/or POPH diagnosis were associated with treatment and health-care utilization. We performed a cross-sectional study of adults enrolled in the Pulmonary Hypertension Association Registry. Patients with IPAH (n = 344) and POPH (n = 57) were compared. Compared with IPAH, patients with POPH were less likely to be college graduates (19.6% vs. 34.9%, p = 0.02) and more likely to be unemployed (54.7% vs. 30.5%, p < 0.001) and have an annual household income below poverty level (45.7% vs. 19.0%, p < 0.001). Patients with POPH had similar functional class, quality of life, 6-min walk distance, and mean pulmonary arterial pressure with a higher cardiac index. Compared with IPAH, patients with POPH were less likely to receive combination therapy (46.4% vs. 62.2%, p = 0.03) and endothelin receptor antagonists (28.6% vs. 55.1%, p < 0.001) at enrollment with similar treatment at follow-up. Patients with POPH had more emergency department visits (1.7 ± 2.1 vs. 0.9 ± 1.2, p = 0.009) and hospitalizations in the six months preceding enrollment (1.5 ± 2.1 vs. 0.8 ± 1.1, p = 0.02). Both POPH diagnosis and lower education level were independently associated with a higher number of emergency department visits. Compared to IPAH, patients with POPH have lower socioeconomic status, are less likely to receive initial combination therapy and endothelin receptor antagonists but have similar treatment at follow-up, and have increased health-care utilization.

Pharmacological vs Exercise Stress Echocardiography for Detection of Cardiac Allograft Vasculopathy.

OBJECTIVE: To test the hypothesis that exercise and dobutamine would provide levels of cardiac stress that are comparable to those achieved in a general stress test population, and to one another, in heart transplant recipients. PATIENTS AND METHODS: From February 10, 2015, to December 31, 2017, 81 patients underwent exercise stress (N=45) or dobutamine stress (N=36) echocardiography at a mean ± SD of 11±14 years (range, 1-29 years) after heart transplant. Hemodynamic and inotropic responses were compared between groups, and to a prior test, longitudinally. The primary outcome was peak heart rate (HR) × systolic blood pressure (SBP). RESULTS: Peak exercise HR × SBP × 10-3 was a mean ± SD of 24.9±4.9 mm Hg/min for exercise stress vs 21.2±3.4 mm Hg/min during dobutamine stress (P<.001). In 35 patients who underwent a dobutamine stress test followed later by another dobutamine stress test, peak HR × SBP changed by 4.2%±16% (P=.05). In 25 patients who underwent a dobutamine stress test followed later by an exercise stress test, peak HR × SBP increased by 12%±23% (P=.002 vs serial dobutamine stress tests). Peak exercise HR did not correlate with time since heart transplant, patient age, or graft age. Peak dobutamine HR correlated modestly with patient age (r 2 =0.28). Inotropic responses were similar in both groups. Overall, patients preferred exercise stress testing to dobutamine stress tests. Dobutamine stress testing was more expensive than exercise stress tests. CONCLUSION: Exercise induces a level of cardiac stress that is equal to or greater than dobutamine-induced stress, at lower cost, in heart transplant recipients who express preference for exercise stress testing.

Biventricular assist device utilization for patients with morbid congestive heart failure: a justifiable strategy.

BACKGROUND: The rationale for the use of a biventricular assist device (BiVAD) for morbid congestive heart failure (MCHF) has been questioned because of historically unacceptable rates of postimplant and post-transplant mortality as well as perceived barriers to their outpatient management. METHODS AND RESULTS: All patients who received a Thoratec BiVAD from January 1990 to December 2003 at the University of Pittsburgh were studied retrospectively. There were a total of 73 patients (32% ischemic, 21% idiopathic, and 47% other) who had a BiVAD implanted. Before implantation, 100% were on > or =1 inotropic agent, and 77% had an intra-aortic balloon pump. Overall survival was 69%; 42 patients (84%) received cardiac transplantation, 5 patients (10%) were weaned, and 3 (6%) remained supported on BiVAD. If the 14 patients with postcardiotomy failure and acute myocardial infarction with shock are excluded, the overall survival improves to 75%. Five-year actuarial survival after heart transplantation was 58%. Of the 29 patients implanted before 2000, the 4-month actuarial freedom from driveline infections, bloodstream infections, and neurological events was 10%, 54%, and 48%, respectively, whereas the rates of these events for the 44 patients implanted after 2000 improved to 70%, 79%, and 80%, respectively. Since 2000, 21 (48%) patients were discharged from the hospital, of whom 38% went to an outpatient residence, 33% to a skilled nursing facility, and 29% to home. Once discharged, > or =1 readmission occurred in 45% and > or =2 readmissions in 48%. CONCLUSIONS: BiVAD support for MCHF has an acceptable overall mortality and survival to transplantation. Morbidity has been significantly reduced in the past 4 years, and management as an outpatient is achievable.

The beta1-adrenergic receptor mediates the pharmacogenetic interaction of the ACE D allele and beta-blockers.

The role of beta-receptor selectivity for the interaction between the angiotensin-converting enzyme (ACE) insertion/deletion polymorphism and beta-blocker therapy was investigated in 479 subjects with left ventricular dysfunction. Subjects were separated into no beta-blocker, beta1 -selective, and nonselective beta-blocker treatment groups. The D allele adversely affected transplant-free survival for subjects not on beta-blockers (p= 0.004). Treatment with selective beta1-blockers eliminated the impact of the D allele (p= 0.51) in a manner similar to nonselective beta1,2-blockers (p= 0.80). Treatment with beta1-blockers was sufficient to eliminate the adverse impact of the ACE D allele, suggesting this pharmacogenetic interaction is mediated through the beta1-receptor.

Cerebrovascular accidents in patients with a ventricular assist device.

OBJECTIVE: A cerebrovascular accident is a devastating adverse event in a patient with a ventricular assist device. The goal was to clarify the risk factors for cerebrovascular accident. METHODS: Prospectively collected data, including medical history, ventricular assist device type, white blood cell count, thrombelastogram, and infection, were reviewed retrospectively in 124 patients. RESULTS: Thirty-one patients (25%) had 48 cerebrovascular accidents. The mean ventricular assist device support period was 228 and 89 days in patients with and without cerebrovascular accidents, respectively (P < .0001). Sixty-six percent of cerebrovascular accidents occurred within 4 months after implantation. Actuarial freedom from cerebrovascular accident at 6 months was 75%, 64%, 63%, and 33% with the HeartMate device (Thoratec Corp, Pleasanton, Calif), Thoratec biventricular ventricular assist device (Thoratec Corp), Thoratec left ventricular assist device (Thoratec), and Novacor device (WorldHeart, Oakland, Calif), respectively. Twenty cerebrovascular accidents (42%) occurred in patients with infections. The mean white blood cell count at the cerebrovascular accident was greater than the normal range in patients with infection (12,900/mm3) and without infection (9500/mm3). The mean maximum amplitude of the thrombelastogram in the presence of infection (63.6 mm) was higher than that in the absence of infection (60.7 mm) (P = .0309). CONCLUSIONS: The risk of cerebrovascular accident increases with a longer ventricular assist device support period. Infection may activate platelet function and predispose the patient to a cerebrovascular accident. An elevation of the white blood cell count may also exacerbate the risk of cerebrovascular accident even in patients without infection. Selection of device type, prevention of infection, and meticulous control of anticoagulation are key to preventing cerebrovascular accident.