Introduction: The diagnosis of myocardial infarction (MI) needs to be swift and accurate, but definitively diagnosing it based on the first test encountered in clinical practice, the electrocardiogram (ECG), is not an easy task. The purpose of the study was to develop a deep learning (DL) algorithm using multitask learning method to differentiate patients experiencing MI from those without coronary artery disease using image-based ECG data. Methods: A DL model was developed based on 11,227 ECG images. We developed a new ECG interpretation algorithm through signal-guided multitask learning, building on a previously published single-task algorithm. The utility of this model was evaluated by testing 51 physicians in interpreting ECGs with and without the assistance of the DL model. Results: The proposed model demonstrated superior performance, achieving 90.56% accuracy, 83.82% sensitivity, 93.02% specificity, 81.44% precision, and an F1 score of 82.61% in discriminating MI ECG. Overall, the median accuracy of ECG interpretation improved from 62% to 68% with the DL algorithm. Trainees and specialists in internal medicine experienced significant accuracy increases (60–66% for trainees, 72–80% for specialists). In the MI group, NSTEMI accuracy was notably lower than STEMI (33% vs. 80%, p < 0.001), but the DL algorithm improved interpretative capabilities in both NSTEMI and STEMI. Conclusions: Signal-guided multitask DL algorithm demonstrated superior performance compared with previous single-task algorithm. The DL algorithm supports the physicians’ decision discriminating MI ECGs from non-MI ECGs. The improvement was consistent in subgroups of STEMI and NSTEMI.

Introduction: To date, there is no specific evidence or criteria for the selection of patients with PH and severe tricuspid insufficiency that can be initiated into correction of tricuspid valvulopathy. Tricuspid regurgitation is a risk marker independent of mortality in patients with pulmonary hypertension. The critical factor for the procedure’s success is to find the parameters to select patients so that they do not become just a futile act. Method: From the initial group of 271 patients, a final group of 123 patients were selected, all diagnosed with precapillary PH confirmed by catheterization and with tricuspid regurgitation by echocardiography. Patients were in groups 1 and 2 according to the 2022 Pulmonary Hypertension Guidelines. Patients with right to left shunt were not excluded. Results: In patients with severe precapillary PH, the sPAP/PAAT ratio was close to 1 (0.89 ± 0.43), while in patients with mild precapillary PH or in the postcapillary group, the sPAP/PAAT ratio was considerably lower (0.47 ± 0.20, p < 0.001). The average sPAP/PAAT of deceased patients was 0.76. Among the 68 deceased patients, 42 (61.70%) had severe tricuspid regurgitation. Conclusion: In our study, the average sPAP/PAAT ratio of the deceased patients with severe FTR was 0.76 mm Hg/ms; nevertheless, this knowledge could have a potential use but is not sufficient for full-informed qualification or disqualification for valve intervention, which requires specific TTVR-related data.

Introduction: Exercise-based cardiac rehabilitation programs (CRPs) represent a multidisciplinary therapeutic approach tailored to improve cardiorespiratory fitness and overall health-related quality of life, while reducing cardiovascular hospitalizations and mortality. Our aim was to determine changes in exercise capacity and cardiac events according to the left ventricular ejection fraction (LVEF) following a CRP. Methods: This single-center cohort study included 131 patients who completed a center-based CRP from 2015 to 2022. Cardiopulmonary exercise testing (CPET), transthoracic echocardiography, and laboratory analysis were performed before and after program completion. Results: Of 232 enrolled patients, 131 were included in our cohort analysis, with 36.6% having a reduced LVEF. These patients exhibited higher rates of atrial fibrillation (p = 0.019) and NT-proBNP values (p < 0.001). Patients with reduced LVEF had lower peak VO2 (pVO2) (p = 0.046), percentage of predicted oxygen consumption (ppVO2) (p < 0.001), and VO2 at anaerobic threshold (p = 0.015) during baseline CPET. Completion of the CRP led to comparable improvements in exercise performance between groups, assessed through Δ pVO2 (p = 0.990), Δ ppVO2 (p = 0.610), and Δ VE/VCO2 slope (p = 0.200). In the study population, male gender and hypertension independently predicted variation in VE/VCO2 slope (p = 0.017) and oxygen uptake efficiency slope (p = 0.048), respectively, post-CRP. During follow-up, 9.2% of all patients had hospital admissions for cardiac events, and 3.8% died, with no significant differences between groups. Conclusion: Patients with reduced LVEF admitted to our CRP showed lower functional status based on specific CPET parameters. After CRP, both reduced and preserved LVEF groups experienced similar functional improvements. Therefore, strategies to increase participation of patients with reduced LVEF in such programs are warranted.

Introduction: Heart failure (HF) may induce bowel hypoperfusion, leading to hypoxia of the villa of the bowel wall and the occurrence of Clostridioides difficile infection (CDI). However, the risk factors for the development of CDI in HF patients have yet to be fully illustrated, especially because of a lack of evidence from real-world data. Methods: Clinical data and survival situations of HF patients with CDI admitted to ICU were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database. For developing a model that can predict 28-day all-cause mortality in HF patients with CDI, the Recursive Feature Elimination with Cross-Validation (RFE-CV) method was used for feature selection. And nine machine learning (ML) algorithms, including logistic regression (LR), decision tree, Bayesian, adaptive boosting, random forest (RF), gradient boosting decision tree, XGBoost, light gradient boosting machine, and categorical boosting, were applied for model construction. After training and hyperparameter optimization of the models through grid search 5-fold cross-validation, the performance of models was evaluated by the area under curve (AUC), accuracy, sensitivity, specificity, precision, negative predictive value, and F1 score. Furthermore, the SHapley Additive exPlanations (SHAP) method was used to interpret the optimal model. Results: A total of 526 HF patients with CDI were included in the study, of whom 99 cases (18.8%) experienced death within 28 days. Eighteen of the 57 variables were selected for the model construction algorithm for model construction. Among the ML models considered, the RF model emerged as the optimal model achieving the accuracy, F1-score, and AUC values of 0.821, 0.596, and 0.864, respectively. The net benefit of the model surpassed other models at 16%–22% threshold probabilities based on decision curve analysis. According to the importance of features in the RF model, red blood cell distribution width, blood urea nitrogen, Simplified Acute Physiology Score II, Sequential Organ Failure Assessment, and white blood cell count were highlighted as the five most influential variables. Conclusions: We developed ML models to predict 28-day all-cause mortality in HF patients associated with CDI in the ICU, which are more effective than the conventional LR model. The RF model has the best performance among all the ML models employed. It may be useful to help clinicians identify high-risk HF patients with CDI.

Plain Language SummaryElectric shock treatment is safe and well tolerated in the most common form of heart arrhythmia. The most common form of heart arrhythmia in humans is called AF. This affects up to 1/3 of the general population in their lifetime. Many people with this condition develop symptoms such as breathlessness and chest pain. Even worse, this arrhythmia increases the risk of developing stroke dramatically. Common treatments include medicines that prevents the blood from clotting and thereby forming strokes and small, timed electric impulses to restore a regular rhythm. However, the transition from arrhythmia to a regular heart rhythm increases the risk for stroke as well, albeit only for a short time. Moreover, not all small strokes are necessarily immediately symptomatic. In this work, researchers used advanced imaging diagnostic to find out how often this may happen. They found out, that most people really felt better, though 1/3 of the patients may develop a relapse of their arrhythmia. Small strokes after the procedure were not a regular finding. However they may still develop, as seen in one patient in this study despite appropriate medicine to prevent blood clots.

Introduction: Muscular ventricular septal defect occluders (MVSDOs) have been attempted as an option in low-weight patients with patent ductus arteriosus (PDA). However, few studies have assessed the safety of transcatheter patent ductus arteriosus closure (TCPC) using MVSDO. Therefore, we compared the outcomes in low-weight patients who used MVSDO and mushroom-shaped occluder (MSO). Methods: Medical records of children under 10 kg (n = 417) who underwent TCPC from 2015 to 2021 at a Chinese health center were reviewed. They were divided into MSO (n = 372) and MVSDO (n = 45) groups. A 1:1 propensity score matching (PSM) was done considering gender, height, weight, body surface area (BSA), PDA diameter, and BSA-corrected PDA diameter. Results: All 45 children in the MVSDO group (mean weight: 5.92 ± 1.32 kg) achieved successful immediate occlusion. One case in the MVSDO group experienced device migration within 24 h requiring unplanned surgery. MVSDO significantly ameliorated pulmonary artery hypertension. After PSM, each group comprised 41 children. The MVSDO group had a smaller effect on platelet counts (MVSDO vs. MSO = 259.85 ± 114.82 vs. 356.12 ± 134.37, p < 0.001), a reduced incidence of thrombocytopenia (MVSDO vs. MSO = 2/41 vs. 7/41, p = 0.001), and a higher rate of residual shunting (MVSDO vs. MSO = 16/41 vs. 5/41, p = 0.005), compared with the MSO group. Thrombocytopenia resolved during hospitalization and micro-shunts disappeared by 6 months. No pulmonary artery or descending aortic secondary stenosis was observed in 1-year follow-up. Conclusions: MVSDO used in low-weight children is feasible, with high success and satisfactory postoperative and short-term follow-up outcomes, including lower thrombocytopenia incidence, compared to MSO. Further long-term studies with larger samples are recommended.

Introduction: Patients undergoing left atrial appendage occlusion (LAAO) are at increased risk for bleeding or thromboembolic events. Concurrently, biomarkers are of growing importance in risk stratification for atrial fibrillation patients. We aimed to evaluate the association of hematological markers and clinical characteristics with the occurrence of thromboembolic and bleeding events following LAAO. Methods: Seven implanting centers retrospectively gathered data on hematological markers (i.e., platelet count [PC], mean platelet volume [MPV], and fibrinogen) prior to LAAO. Prespecified thromboembolic and major bleeding outcomes were collected and the association with pre-procedural hematological markers and clinical characteristics was evaluated using Cox regression analysis. Results: In total, 1,315 patients were included (74 ± 9 years, 36% female, CHA2DS2-VASc 4.3 ± 1.5, HAS-BLED 3.3 ± 1.1). Over a total follow-up duration of 2,682 patient years, 77 thromboembolic events and 107 major bleeding events occurred after LAAO. Baseline PC was the only biomarker showing a signal for a relation to thromboembolic events (HR 1.18, 95% CI: 1.00–1.39) per 50*109 increment, p = 0.056). Thrombotic event rates, including device-related thrombus, increased within higher PC quartiles. Thromboembolism was associated with age (HR 1.05, 95% CI: 1.00–1.10, per year increase) and prior thromboembolism (HR 2.08, 95% CI: 1.07–4.03), but with none of the biomarkers in multivariate analysis. No association of any of the hematological markers with major bleeding was observed. Major bleeding following LAAO was associated with prior major bleeding (HR 5.27, 95% CI: 2.71–10.22), renal disease (HR 1.93, 95% CI: 1.17–3.18), and discharge on dual antiplatelet therapy (DAPT) (HR 1.71, 95% CI: 1.05–2.77). Conclusion: Most thrombotic events occurred in the highest PC quartile, but no association of any of the hematological markers with thromboembolism or major bleeding was observed in our analysis. In multivariate analysis, older age and prior thromboembolism were associated with thromboembolism. Prior major bleeding, renal disease and discharge on DAPT were multivariate predictors of major bleeding after LAAO.

Introduction: Angiotensin-converting enzyme 2 (ACE2) of the renin-angiotensin-aldosterone system (RAAS) serves as a functional receptor to gain entry into the cells for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes coronavirus disease 2019 (COVID-19). The interaction between SARS-CoV-2 and ACE2 is a potential virulent factor in infectivity. Our study aimed to ascertain the association of RAAS inhibitors with adverse cardiovascular and other outcomes in hospitalized COVID-19 patients. Methods: This is a retrospective study of medical records of ≥18-year-old patients hospitalized for COVID-19 from March 2020 to October 2020. Primary outcomes were acute cardiovascular events (ST-elevation myocardial infarction, non-ST-elevation myocardial infarction type 1, acute congestive heart failure, acute stroke) and mortality. Secondary outcomes were respiratory failure, need for and duration of mechanical ventilation, acute deep vein thrombosis or pulmonary embolism (DVT/PE), and readmission rate. Results: Among 376 hospitalized COVID-19 patients, 149 were on RAAS inhibitors. No statistically significant differences were found between RAAS inhibitor and non-RAAS inhibitor groups with respect to acute cardiovascular events (6% vs. 6.2%, p = 0.94), acute DVT/PE (4.7% vs. 4.8%, p = 0.97), hypoxia (62.4% vs. 58.6%, p = 0.46), need for mechanical ventilation (18.1% vs. 16.7%, p = 0.72), mortality (19.5% vs. 22%, p = 0.56), and readmission rate (11.4% vs. 14.1%, p = 0.45). Some nuances discovered were a higher rate of hospitalizations among Native Americans receiving RAAS inhibitors (30.2% vs. 19.8%) and significantly lower levels of procalcitonin in patients on RAAS inhibitors. Conclusions: Among hospitalized patients with COVID-19, those on RAAS inhibitors showed no significant differences in acute cardiovascular events, acute DVT/PE, hypoxia, need for mechanical ventilation, readmission, or mortality rate compared to those not on them. However, further large-scale studies are needed to validate these findings.