Rana, Jamal S.

Background/Objectives: Transcatheter aortic valve replacement (TAVR) has been introduced as an optimal treatment for patients with severe aortic stenosis, offering a minimally invasive alternative to surgical aortic valve replacement. Predicting these outcomes following TAVR is crucial. Artificial intelligence (AI) has emerged as a promising tool for improving post-TAVR outcome prediction. In this systematic review and meta-analysis, we aim to summarize the current evidence on utilizing AI in predicting post-TAVR outcomes. Methods: A comprehensive search was conducted to evaluate the studies focused on TAVR that applied AI methods for risk stratification. We assessed various ML algorithms, including random forests, neural networks, extreme gradient boosting, and support vector machines. Model performance metrics—recall, area under the curve (AUC), and accuracy—were collected with 95% confidence intervals (CIs). A random-effects meta-analysis was conducted to pool effect estimates. Results: We included 43 studies evaluating 366,269 patients (mean age 80 ± 8.25; 52.9% men) following TAVR. Meta-analyses for AI model performances demonstrated the following results: all-cause mortality (AUC = 0.78 (0.74–0.82), accuracy = 0.81 (0.69–0.89), and recall = 0.90 (0.70–0.97); permanent pacemaker implantation or new left bundle branch block (AUC = 0.75 (0.68–0.82), accuracy = 0.73 (0.59–0.84), and recall = 0.87 (0.50–0.98)); valve-related dysfunction (AUC = 0.73 (0.62–0.84), accuracy = 0.79 (0.57–0.91), and recall = 0.54 (0.26–0.80)); and major adverse cardiovascular events (AUC = 0.79 (0.67–0.92)). Subgroup analyses based on the model development approaches indicated that models incorporating baseline clinical data, imaging, and biomarker information enhanced predictive performance. Conclusions: AI-based risk prediction for TAVR complications has demonstrated promising performance. However, it is necessary to evaluate the efficiency of the aforementioned models in external validation datasets.

Background Variation in outcomes by race/ethnicity in adults with heart failure (HF) has been previously observed. Identifying factors contributing to these variations could help target interventions. We evaluated the association of race/ethnicity with HF outcomes and potentially contributing factors within a contemporary HF cohort. Methods and Results We identified members of Kaiser Permanente Northern California, a large integrated healthcare delivery system, who were diagnosed with HF between 2012 and 2016 and had at least 1 year of prior continuous membership and left ventricular ejection fraction data. We used Cox regression with time‐dependent covariates to evaluate the association of self‐identified race/ethnicity with HF or all‐cause hospitalization and all‐cause death, with backward selection for potential explanatory variables. Among 34 621 patients with HF, compared with White patients, Black patients had a higher rate of HF hospitalization (adjusted hazard ratio [HR], 1.28; 95% CI, 1.18–1.38) but a lower rate of death (adjusted HR, 0.78; 95% CI, 0.72–0.85). In contrast, Asian/Pacific Islander patients had similar rates of HF hospitalization, but lower rates of all‐cause hospitalization (adjusted HR, 0.89; 95% CI, 0.85–0.93) and death (adjusted HR, 0.75; 95% CI, 0.69–0.80). Hispanic patients also had a lower rate of death (adjusted HR, 0.85; 95% CI, 0.80–0.91). Sensitivity analyses showed that effect sizes for Black patients were larger among patients with reduced ejection fraction. Conclusions In a contemporary and diverse population with HF, Black patients experienced a higher rate of HF hospitalization and a lower rate of death compared with White patients. In contrast, selected outcomes for Asian/Pacific Islander and Hispanic patients were more favorable compared with White patients. The observed differences were not explained by measured potentially modifiable factors, including pharmacological treatment. Future research is needed to identify explanatory mechanisms underlying ongoing racial/ethnic variation to target potential interventions.

Introduction: Alcohol use is common in the United States. Most studies describe a J-shaped relationship between alcohol consumption and cognitive function: frequent heavy consumption alters brain functions and decreases cognitive performance; regular light and moderate consumption may have protective impact. We are unaware of any reports utilizing longitudinally collected data to derive a cumulative exposure of alcohol use and assess its association with cognitive function. Hypothesis: Cumulative 30-year exposure to alcohol has a J-shaped association with mid-life cognitive function. Methods: We included 3,068 participants aged 18-30 years at the baseline examination (1985-86) and having up to 8 follow-up exams at 2- to 5-year intervals including at 30 years (2015-16). Cumulative “drink-years” exposure to alcohol was estimated at each exam by multiplying the usual number of drinks/day reported by the number of years since last exam, and then summing across exams. We used multivariable adjusted linear regression models to assess the independent associations of number of drink-years with 6 measures of cognitive function studied at Year 30: Rey Auditory Verbal Learning Test (RAVLT), Digit Symbol Substitution Test (DSST), Stroop Test, Montreal Cognitive Assessment (MoCA), Category Fluency Test, and Letter Fluency Test. We used inverse probability of censoring weighting to account for potential informative censoring. Linear regression models estimated standardized mean test scores for the cumulative alcohol use categories (drink-years: >0 and <15, 15-29.9, 30+). Too few participants were available for assessment of higher categories of use. Results: Compared with never drinking, the 3 categories of cumulative exposure were unassociated with any of the cognitive function measures. Conclusions: Cumulative 30-year alcohol use is not associated with cognitive function in middle-aged adults in the CARDIA study.