Agasthi, Pradyumna

Abstract Aims : Direct oral anticoagulants (DOACs) are widely used for stroke prevention in people with non-valvular atrial fibrillation/flutter (NVAF). Anticoagulants have different bleeding profiles. Apixaban is associated with the lowest bleeding risk. This study evaluates the interaction between DOAC used and right ventricular systolic pressure (RVSP) on echocardiogram and bleeding rates. It was hypothesized that pulmonary hypertension may be associated with excess bleeding based on anticoagulant-specific pharmacologic profiles. Methods & Results : A retrospective analysis of a database was conducted. Multivariable regression models for bleeding rates were developed using a causative analysis approach among participants who had undergone an echocardiogram, and interaction terms for DOAC choice were incorporated. Stratum specific estimates were calculated using linear computation. Patients taking Apixaban had lower bleeding rates than those taking Rivaroxaban and Dabigatran. Among those without pulmonary hypertension, there was no difference in bleeding rates. Where RVSP was ≥45mmHg, there was a 90% increase in bleeding rates with use of Dabigatran (HR 1.99, p=0.024) and Rivaroxaban (HR 1.86, p=0.042) compared to Apixaban (Figure 1). Conclusions : Elevated RVSP is associated with excess bleeding in patients taking Rivaroxaban and Dabigatran, but not Apixaban, after controlling for confounding variables.

Introduction: In the current study, we examined the temporal trends in mortality across gender, race, and 10-year age groups over last two-decades. Methods: We conducted a retrospective population-based study using Centers for Disease Control and Prevention database. Subjects with underlying cause of death from AMI were identified using validated ICD10 codes. Age adjusted mortality rates were calculated for gender and race whereas crude rates were calculated for 10-year age group. Using Poisson regression, we estimated the rate ratio between race/ethnicity pairs adjusting for age and sex in order to analyze temporal trend. Results: We identified 2,709,173 subjects with reported underlying cause of death as AMI between January-2000 to December-2019 of which 54.7% were males (n=1,483,163). The age adjusted mortality rate per 100,000 was higher in males compared to females (52.4% vs 30.7%, OR: 1.66 [1.45-1.89]). Among ethnicities African American (AA) population had the highest age adjusted mortality rate per 100,000 [Table 1]. As expected, age group >85 years had the highest mortality rate per 100,000 constituting 30% of total deaths (78% of total deaths were noted in >65yr age group). AMI age adjusted mortality rates/100,000 showed significant decline across gender and race with highest decline in female population (66.5% decline). Highest percentage decline in mortality was noted in AA female population (68.2%) whereas in males it was the white population (62%). In females, highest percentage decline in mortality was noted in 75-84 age-group (70%) while it was >85 age-group (67%) in males. Conclusion: Over last two decades there has been a significant decline in AMI mortality rates across gender and race. AA continue to have higher age adjusted mortality rate among both male and female population (38% and 22% respectively). Further studies are needed to identify and provide targeted prevention and intervention strategies in high-risk populations.

Abstract Percutaneous Coronary Intervention (PCI) is a widely utilized standard ofcare and efficacious intervention in the management of coronary artery disease,improving prognosis and symptoms for many patients. Prediction of post-PCIoutcomes contributes to effective patient management and quality improvementinitiatives in healthcare. Our study focuses on comparative performance analysisfor state-of-the-art vision transformer (ViT) and convolutional neural network(CNN) on multimodal data analysis in a joint fusion framework. We integrateimages of electrocardiogram (ECG) data and tabular electronic health records(EHR) to predict 3 clinically relevant post-PCI (6 months) endpoints - heart failure hospitalization, all-cause mortality, and ischemic stroke. To design a comparativemodel for ViT, we proposed a new joint fusion architecture that consistsof a convolutional neural network (CNN) with a convolutional block attentionmodule (CBAM). The learned representations are combined at an intermediatelayer followed by processing these representations using a fully connectedlayer. The proposed model demonstrates excellent performance with the highestAUROC scores of 0.849, 0.913, and 0.794 for predicting heart failure hospitalization,all-cause mortality, and stroke, respectively. Surpassing the baseline EHRmodel and ViT, the proposed CNN+CBAM fusion model showcases superiorpredictive capabilities for heart failure prediction.

Background Little is known about outcomes following heart failure (HF) hospitalization among adults with congenital heart disease (CHD) in the United States. We aim to compare the outcomes of HF versus non‐HF hospitalizations in adults with CHD. Methods and Results Using a national deidentified administrative claims data set, patients with adult congenital heart disease (ACHD) hospitalized with and without HF (ACHDHF+, ACHDHF−) were characterized to determine the predictors of 90‐day and 1‐year mortality and quantify the risk of mortality, major adverse cardiac and cerebrovascular events, and health resource use. Cox proportional hazard regression was used to compare ACHDHF+ versus ACHDHF− for risk of events and health resource use. Of 26 454 unique ACHD admissions between January 1, 2010 and December 31, 2020, 5826 (22%) were ACHDHF+ and 20 628 (78%) were ACHDHF−. The ACHD HF+ hospitalizations increased from 6.6% to 14.0% ( P <0.0001). Over a mean follow‐up period of 2.23 ± 2.19 years, patients with ACHDHF+ had a higher risk of mortality (hazard ratio [HR], 1.86 [95% CI, 1.67–2.07], P <0.001), major adverse cardiac and cerebrovascular events (HR, 1.73 [95% CI, 1.63–1.83], P <0.001) and health resource use including rehospitalization (HR, 1.09 [95% CI, 1.05–1.14], P <0.001) and increased postacute care service use (HR, 1.56 [95% CI, 1.32–1.85], P <0.001). Cardiology clinic visits within 30 days of hospital admission were associated with lower 90‐day and 1‐year all‐cause mortality (odds ratio [OR], 0.62 [95% CI, 0.49–0.78], P <0.001; OR, 0.69 [95% CI, 0.58–0.83], P <0.001, respectively). Conclusions HF hospitalization is associated with increased risk of mortality and morbidity with high health resource use in patients with ACHD. Recent cardiology clinic attendance appears to mitigate these risks.

Introduction: Post-transcatheter aortic valve replacement (TAVR) outcome of patients is an important research topic. To accurately assess post-TAVR mortality, we examined a family of new echo-parameters (augmented systolic blood pressure (AugSBP) and arterial mean pressure parameters (AugMAP)) derived from blood pressure and aortic valve gradient ( Figure 1 ). Methods: Patients in the Mayo Clinic National Cardiovascular Diseases Registry-TAVR database who underwent TAVR between January 1, 2012, and June 30, 2017, were identified to retrieve baseline clinical, echocardiographic, and mortality data. The median of AugSBP and AugMAP were used to stratify patients in Kaplan-Meier analysis. Results: The final cohort contained 974 patients with a mean age of 81.4±8.3 years old, and 56.6% were male. The mean Society of Thoracic Surgeons risk score was 8.2±5.2. The median follow-up duration was 354 days, and the one-year all-cause mortality rate was 14.2%. Detailed characteristics are summarized in Table 1. AugMAP1 < 102.5 mmHg (median) was associated with a 3-fold risk of all-cause mortality post-TAVR at 1 year (hazard ratio 3.0, 95%confidence interval 2.0-4.5, p<0.0001)( Figure 1 ). Conclusions: Augmented mean arterial pressure provides clinicians with a simple but effective approach to quickly identify patients at risk and potentially improve post-TAVR prognosis.

AbstractBackgroundAortic stenosis (AS) is associated with myocardial ischemia through different mechanisms and may impair coronary arterial flow. However, data on the impact of moderate AS in patients with acute myocardial infarction (MI) is limited.AimsThis study aimed to investigate the impact of moderate AS in patients presenting with acute myocardial infarction (MI).MethodsWe conducted a retrospective analysis of all patients who presented with acute MI to all Mayo Clinic hospitals, using the Enterprise Mayo PCI Database from 2005 to 2016. Patients were stratified into two groups: moderate AS and mild/no AS. The primary outcome was all cause mortality.ResultsThe moderate AS group included 183 (13.3%) patients, and the mild/no AS group included 1190 (86.7%) patients. During hospitalization, there was no difference between both groups in mortality. Patients with moderate AS had higher in‐hospital congestive heart failure (CHF) (8.2% vs. 4.4%, p = 0.025) compared with mild/no AS patients. At 1‐year follow‐up, patients with moderate AS had higher mortality (23.9% vs. 8.1%, p < 0.001) and higher CHF hospitalization (8.3% vs. 3.7%, p = 0.028). In multivariate analysis, moderate AS was associated with higher mortality at 1‐year (odds ratio 2.4, 95% confidence interval [1.4–4.1], p = 0.002). In subgroup analyses, moderate AS increased all‐cause mortality in STEMI and NSTEMI patients.ConclusionThe presence of moderate AS in acute MI patients was associated with worse clinical outcomes during hospitalization and at 1‐year follow‐up. These unfavorable outcomes highlight the need for a close follow‐up of these patients and for timely therapeutic strategies to best manage these coexisting conditions.

Background: Post-transcatheter aortic valve replacement (TAVR) patient outcome is an important research topic. To accurately assess post-TAVR mortality, we examined a family of new echo parameters (augmented systolic blood pressure (AugSBP) and arterial mean pressure (AugMAP)) derived from blood pressure and aortic valve gradients. Methods: Patients in the Mayo Clinic National Cardiovascular Diseases Registry-TAVR database who underwent TAVR between 1 January 2012 and 30 June 2017 were identified to retrieve baseline clinical, echocardiographic and mortality data. AugSBP, AugMAP and valvulo-arterial impedance (Zva) (Zva) were evaluated using Cox regression. Receiver operating characteristic curve analysis and the c-index were used to assess the model performance against the Society of Thoracic Surgeons (STS) risk score. Results: The final cohort contained 974 patients with a mean age of 81.4 ± 8.3 years old, and 56.6% were male. The mean STS risk score was 8.2 ± 5.2. The median follow-up duration was 354 days, and the one-year all-cause mortality rate was 14.2%. Both univariate and multivariate Cox regression showed that AugSBP and AugMAP parameters were independent predictors for intermediate-term post-TAVR mortality (all p < 0.0001). AugMAP1 < 102.5 mmHg was associated with a 3-fold-increased risk of all-cause mortality 1-year post-TAVR (hazard ratio 3.0, 95%confidence interval 2.0–4.5, p < 0.0001). A univariate model of AugMAP1 surpassed the STS score model in predicting intermediate-term post-TAVR mortality (area under the curve: 0.700 vs. 0.587, p = 0.005; c-index: 0.681 vs. 0.585, p = 0.001). Conclusions: Augmented mean arterial pressure provides clinicians with a simple but effective approach to quickly identify patients at risk and potentially improve post-TAVR prognosis.

Abstract Background. Clinical medicine relies heavily on the synthesis of information and data from multiple sources. However, often simple feature concatenation is used as a strategy for developing a multimodal machine learning model in the cardiovascular domain, and thus the models are often limited by pre-selected features and moderate accuracy. Method. We proposed a two-branched joint fusion model for fusing the 12-lead electrocardiogram (ECG) signal data with clinical variables from the electronic medical record (EMR) in an end-to-end deep learning architecture. The model follows the joint fusion scheme and learns complementary information from ECG and EMR. Retrospective data from the Mayo Clinic Health Systems across four sites for patients that underwent percutaneous coronary intervention (PCI) were obtained. Model performance was assessed by area under the receiver-operating characteristics (AUROC) and Delong’s test. Results. The final cohort included 17,356 unique patients with a mean age of 67.2 ± 12.6 year (mean ± std) and 9,163 (52.7%) were male. The joint fusion model outperformed the ECG time-domain model with statistical margin. The model with clinical data obtained the highest AUROC for all-cause mortality (0.91 at 6 months) but the joint fusion model outperformed for cardiovascular outcomes - heart failure hospitalization and ischemic stroke with a significant margin (Delong’s p < 0.05). Conclusion. To the best of our knowledge, this is the first study that developed a deep learning model with joint fusion architecture for the prediction of post-PCI prognosis and outperformed machine learning models developed using traditional single-source features (clinical variables or ECG features). Adding ECG data with clinical variables did not improve prediction of all-cause mortality as may be expected, but the improved performance of related cardiac outcomes shows that the fusion of ECG generates additional value.

Introduction: Multiple supervised machine learning (ML) models have been used to assess the outcomes in patients who underwent transcatheter aortic valve replacement (TAVR) procedure, however, supervised ML requires significant expert time for annotation and labeling. Unsupervised ML can provide insight into the dataset without this process. Hypothesis: We hypothesized that unsupervised k-means ML can identify distinct phenogroups in TAVR patients that can predict prognosis without a priori knowledge. Methods: A retrospective study of patients who underwent TAVR (Jan 2012-Dec 2017) at Mayo Clinic sites was performed. Baseline demographics, EKG, cardiac computed tomography, and echocardiographic data were abstracted. Variables with ≥ 20% missing data were excluded. Prior to the k-means clustering algorithm, principal component analysis was used for dimensionality reduction. Input variables were analyzed according to k-means determined clusters. Kaplan-Meier survival analysis was used to assess the survival of each cluster. Results: A total of 1071 consecutive patients were included in the final analysis. A total of 141 unique variables were used. The mean age was 80.7±8.6 years, 58.2% were male. Patients were divided into 2 distinct phenogroups. In the two phenogroups, cluster 0 (n=639) has significantly better survival than cluster 1 (n= 432) (log-rank p=0.0002) (Figure 1). Cluster 1 has features of low flow-low gradient (LFLG) hemodynamic profile, including lower systolic blood pressure, mean aortic valve gradient, peak aortic valve velocity, and worse LV ejection fraction. Conclusions: Unsupervised ML can identify distinct phenotype features with prognostic significance in TAVR patients. Our results suggest that patients with LFLG hemodynamic features are intrinsically different from other TAVR patients and require more attention to avoid adverse outcomes.