Can supervised deep learning architecture outperform autoencoders in building propensity score models for matching?

PURPOSE: Propensity score matching is vital in epidemiological studies using observational data, yet its estimates relies on correct model-specification. This study assesses supervised deep learning models and unsupervised autoencoders for propensity score estimation, comparing them with traditional methods for bias and variance accuracy in treatment effect estimations. METHODS: Utilizing a plasmode simulation based on the Right Heart Catheterization dataset, under a variety of settings, we evaluated (1) a supervised deep learning architecture and (2) an unsupervised autoencoder, alongside two traditional methods: logistic regression and a spline-based method in estimating propensity scores for matching. Performance metrics included bias, standard errors, and coverage probability. The analysis was also extended to real-world data, with estimates compared to those obtained via a double robust approach. RESULTS: The analysis revealed that supervised deep learning models outperformed unsupervised autoencoders in variance estimation while maintaining comparable levels of bias. These results were supported by analyses of real-world data, where the supervised model's estimates closely matched those derived from conventional methods. Additionally, deep learning models performed well compared to traditional methods in settings where exposure was rare. CONCLUSION: Supervised deep learning models hold promise in refining propensity score estimations in epidemiological research, offering nuanced confounder adjustment, especially in complex datasets. We endorse integrating supervised deep learning into epidemiological research and share reproducible codes for widespread use and methodological transparency.

Impact of Intraprocedural Mitral Regurgitation and Gradient Following Transcatheter Edge-to-Edge Repair for Primary Mitral Regurgitation.

BACKGROUND: The impact of intraprocedural results following transcatheter edge-to-edge repair (TEER) in primary mitral regurgitation (MR) is controversial. OBJECTIVES: This study sought to investigate the prognostic impact of intraprocedural residual mitral regurgitation (rMR) and mean mitral valve gradient (MPG) in patients with primary MR undergoing TEER. METHODS: The PRIME-MR (Outcomes of Patients Treated With Mitral Transcatheter Edge-to-Edge Repair for Primary Mitral Regurgitation) registry included consecutive patients with primary MR undergoing TEER from 2008 to 2022 at 27 international sites. Clinical outcomes were assessed according to intraprocedural rMR and mean MPG. Patients were categorized according to rMR (optimal result: ≤1+, suboptimal result: ≥2+) and MPG (low gradient: ≤5 mm Hg, high gradient: > 5 mm Hg). The prognostic impact of rMR and MPG was evaluated in a Cox regression analysis. The primary endpoint was 2-year all-cause mortality or heart failure hospitalization. RESULTS: Intraprocedural rMR and mean MPG were available in 1,509 patients (median age = 82 years [Q1-Q3: 76.0-86.0 years], 55.1% male). Kaplan-Meier analysis according to rMR severity showed significant differences for the primary endpoint between rMR ≤1+ (29.1%), 2+ (41.7%), and ≥3+ (58.0%; P < 0.001), whereas there was no difference between patients with a low (32.4%) and high gradient (42.1%; P = 0.12). An optimal result/low gradient was achieved in most patients (n = 1,039). The worst outcomes were observed in patients with a suboptimal result/high gradient. After adjustment, rMR ≥2+ was independently linked to the primary endpoint (HR: 1.87; 95% CI: 1.32-2.65; P < 0.001), whereas MPG >5 mm Hg was not (HR: 0.78; 95% CI: 0.47-1.31; P = 0.35). CONCLUSIONS: Intraprocedural rMR but not MPG independently predicted clinical outcomes following TEER for primary MR. When performing TEER in primary MR, optimal MR reduction seems to outweigh the impact of high transvalvular gradients.

Clinical outcomes and predictors of transapical transcatheter mitral valve replacement: the Tendyne Expanded Clinical Study.

BACKGROUND: Transcatheter mitral valve replacement (TMVR) is a therapeutic option for patients with severe mitral regurgitation (MR) who are ineligible for conventional surgery. There are limited data on the outcomes of large patient cohorts treated with TMVR. AIMS: This study aimed to investigate the outcomes and predictors of mortality for patients treated with transapical TMVR. METHODS: This analysis represents the clinical experience of all patients enrolled in the Tendyne Expanded Clinical Study. Patients with symptomatic MR underwent transapical TMVR with the Tendyne system between November 2014 and June 2020. Outcomes and adverse events up to 2 years, as well as predictors of short-term mortality, were assessed. RESULTS: A total of 191 patients were treated (74.1±8.0 years, 62.8% male, Society of Thoracic Surgeons Predicted Risk of Mortality 7.7±6.6%). Technical success was achieved in 96.9% (185/191), and there were no intraprocedural deaths. At 30-day, 1- and 2-year follow-up, the rates of all-cause mortality were 7.9%, 30.8% and 40.5%, respectively. Complete MR elimination (MR <1+) was observed in 99.3%, 99.1% and 96.3% of patients, respectively. TMVR treatment resulted in consistent improvement of New York Heart Association Functional Class and quality of life up to 2 years (both p<0.001). Independent predictors of early mortality were age (odds ratio [OR] 1.11; p=0.003), pulmonary hypertension (OR 3.83; p=0.007), and institutional experience (OR 0.40; p=0.047). CONCLUSIONS: This study investigated clinical outcomes in the full cohort of patients included in the Tendyne Expanded Clinical Study. The Tendyne TMVR system successfully eliminated MR with no intraprocedural deaths, resulting in an improvement in symptoms and quality of life. Continued refinement of clinical and echocardiographic risks will be important to optimise longitudinal outcomes.

Left Atrial Roof Enlargement Is a Distinct Feature of Heart Failure With Preserved Ejection Fraction.

BACKGROUND: It remains unknown to what extent intrinsic atrial cardiomyopathy or left ventricular diastolic dysfunction drive atrial remodeling and functional failure in heart failure with preserved ejection fraction (HFpEF). Computational 3-dimensional (3D) models fitted to cardiovascular magnetic resonance allow state-of-the-art anatomic and functional assessment, and we hypothesized to identify a phenotype linked to HFpEF. METHODS: Patients with exertional dyspnea and diastolic dysfunction on echocardiography (E/e', >8) were prospectively recruited and classified as HFpEF or noncardiac dyspnea based on right heart catheterization. All patients underwent rest and exercise-stress right heart catheterization and cardiovascular magnetic resonance. Computational 3D anatomic left atrial (LA) models were generated based on short-axis cine sequences. A fully automated pipeline was developed to segment cardiovascular magnetic resonance images and build 3D statistical models of LA shape and find the 3D patterns discriminant between HFpEF and noncardiac dyspnea. In addition, atrial morphology and function were quantified by conventional volumetric analyses and deformation imaging. A clinical follow-up was conducted after 24 months for the evaluation of cardiovascular hospitalization. RESULTS: Beyond atrial size, the 3D LA models revealed roof dilation as the main feature found in masked HFpEF (diagnosed during exercise-stress only) preceding a pattern shift to overall atrial size in overt HFpEF (diagnosed at rest). Characteristics of the 3D model were integrated into the LA HFpEF shape score, a biomarker to characterize the gradual remodeling between noncardiac dyspnea and HFpEF. The LA HFpEF shape score was able to discriminate HFpEF (n=34) to noncardiac dyspnea (n=34; area under the curve, 0.81) and was associated with a risk for atrial fibrillation occurrence (hazard ratio, 1.02 [95% CI, 1.01-1.04]; P=0.003), as well as cardiovascular hospitalization (hazard ratio, 1.02 [95% CI, 1.00-1.04]; P=0.043). CONCLUSIONS: LA roof dilation is an early remodeling pattern in masked HFpEF advancing to overall LA enlargement in overt HFpEF. These distinct features predict the occurrence of atrial fibrillation and cardiovascular hospitalization. REGISTRATION: URL: https://www.clinicaltrials.gov; Unique identifier: NCT03260621.

Graft Patency and Clinical Outcomes in Patients With Radial Artery Grafts Previously Instrumented for Cardiac Catheterization.

BACKGROUND: While transradial access is favored for cardiac catheterization, the radial artery (RA) is increasingly preferred for coronary artery bypass grafting. Whether the RA is suitable for use as a graft following instrumentation for transradial access remains uncertain. METHODS: Consecutive patients from 2015 to 2019 who underwent coronary artery bypass grafting using both the left and right RAs as grafts were included. Instrumented RAs underwent careful preoperative assessment for suitability. The clinical analysis was stratified by whether patients received an instrumented RA graft (instrumented versus noninstrumented groups). Eligible patients with both instrumented and noninstrumented RAs underwent computed tomography coronary angiography to evaluate graft patency. The primary outcome was a within-patient paired analysis of graft patency comparing instrumented to noninstrumented RA grafts. RESULTS: Of the 1123 patients who underwent coronary artery bypass grafting, 294 had both the left and right RAs used as grafts and were included. There were 126 and 168 patients in the instrumented and noninstrumented groups, respectively. Baseline characteristics and perioperative outcomes were comparable. The rate of major adverse cardiac events at 2 years following coronary artery bypass grafting was 2.4% in the instrumented group and 5.4% in the noninstrumented group (hazard ratio, 0.44 [95% CI, 0.12-1.61]; P=0.19). There were 50 patients included in the graft patency analysis. At a median follow-up of 4.3 (interquartile range, 3.7-4.5) years, 40/50 (80%) instrumented and 41/50 (82%) noninstrumented grafts were patent (odds ratio, 0.86 [95% CI, 0.29-2.52]; P>0.99). No significant differences were observed in the luminal diameter or cross-sectional area of the instrumented and noninstrumented RA grafts. CONCLUSIONS: There was no evidence found in this study that RA graft patency was affected by prior transradial access, and the use of an instrumented RA was not associated with worse outcomes in the exploratory clinical analysis. Although conduits must be carefully selected, prior transradial access should not be considered an absolute contraindication to the use of the RA as a bypass graft. REGISTRATION: URL: https://www.anzctr.org.au/; Unique identifier: ACTRN12621000257864.

Anesthetic Choice for Percutaneous Transcatheter Closure of the Left Atrial Appendage: A National Anesthesia Clinical Outcomes Registry Analysis.

CONTEXT: Left atrial appendage closure (LAAC) was developed as a novel stroke prevention alternative for patients with atrial fibrillation, particularly for those not suitable for long-term oral anticoagulant therapy. Traditionally, general anesthesia (GA) has been more commonly used primarily due to the necessity of transesophageal echocardiography. AIMS: Compare trends of monitored anesthesia care (MAC) versus GA for percutaneous transcatheter LAAC with endocardial implant and assess for independent variables associated with primary anesthetic choice. SETTINGS AND DESIGN: Multi-institutional data collected from across the United States using the National Anesthesia Clinical Outcomes Registry. MATERIAL AND METHODS: Retrospective data analysis from 2017-2021. STATISTICAL ANALYSIS USED: Independent-sample t tests or Mann-Whitney U tests were used for continuous variables and Chi-square tests or Fisher's exact test for categorical variables. Multivariate logistic regression was used to assess patient and hospital characteristics. RESULTS: A total of 19,395 patients underwent the procedure, and 352 patients (1.8%) received MAC. MAC usage trended upward from 2017-2021 (P < 0.0001). MAC patients were more likely to have an American Society of Anesthesiologists (ASA) physical status of≥ 4 (33.6% vs 22.89%) and to have been treated at centers in the South (67.7% vs 44.2%), in rural locations (71% vs 39.5%), and with lower median annual percutaneous transcatheter LAAC volume (102 vs 153 procedures) (all P < 0.0001). In multivariate analysis, patients treated in the West had 85% lower odds of receiving MAC compared to those in the Northeast (AOR: 0.15; 95% CI 0.03-0.80, P = 0.0261). CONCLUSIONS: While GA is the most common anesthetic technique for percutaneous transcatheter closure of the left atrial appendage, a small, statistically significant increase in MAC occurred from 2017-2021. Anesthetic management for LAAC varies with geographic location.

Transcatheter Mitral Valve Intervention: Current and Future Role of Multimodality Imaging for Device Selection and Periprocedural Guidance.

Mitral regurgitation (MR) is a broadly diffuse valvular heart disease (VHD) with a significant impact on the healthcare system and patient prognosis. Transcatheter mitral valve interventions (TMVI) are now well-established techniques included in the therapeutic armamentarium for managing patients with mitral regurgitation, either primary or functional MR. Even if the guidelines give indications regarding the correct management of this VHD, the wide heterogeneity of patients' clinical backgrounds and valvular and heart anatomies make each patient a unique case, in which the appropriate device's selection requires a multimodal imaging evaluation and a multidisciplinary discussion. Proper pre-procedural evaluation plays a pivotal role in judging the feasibility of TMVI, while a cooperative work between imagers and interventionalist is also crucial for procedural success. This manuscript aims to provide an exhaustive overview of the main parameters that need to be evaluated for appropriate device selection, pre-procedural planning, intra-procedural guidance and post-operative assessment in the setting of TMVI. In addition, it tries to give some insights about future perspectives for structural cardiovascular imaging.

Practical Guidance for Hemodynamic Assessment by Right Heart Catheterization in Management of Heart Failure.

Heart failure is a clinical syndrome characterized by the inability of the heart to meet the circulatory demands of the body without requiring an increase in intracardiac pressures at rest or with exertion. Hemodynamic parameters can be measured via right heart catheterization, which has an integral role in the full spectrum of heart failure: from ambulatory patients to those in cardiogenic shock, as well as patients being considered for left ventricular device therapy and heart transplantation. Hemodynamic data are critical for prompt recognition of clinical deterioration, assessment of prognosis, and guidance of treatment decisions. This review is a field guide for hemodynamic assessment, troubleshooting, and interpretation for clinicians treating patients with heart failure.

Iatrogenic Atrial Septal Defect After MitraClip Transcatheter Edge-to-Edge Repair: To Close or Not to Close?

The evolution of percutaneous procedures that use transseptal puncture to treat left-sided structural heart disease has led to the emergence of iatrogenic atrial septal defects as a potential complication. These defects can result in hemodynamic decompensation and worsening clinical outcomes. Some iatrogenic atrial septal defects require immediate closure, others do not. This case report presents 2 patients who underwent transcatheter edge-to-edge mitral valve repair with transseptal puncture and required iatrogenic atrial septal defect closure (1 immediate and 1 delayed). The goal of this report is to highlight iatrogenic atrial septal defect assessment and the possible need for closure after transseptal puncture.

Incidence and predictors of 2-year mortality following percutaneous left atrial appendage occlusion in the EWOLUTION trial.

AIMS: Sufficient survival time following left atrial appendage occlusion (LAAO) is essential for ensuring the efficacy and cost-effectiveness of this strategy for stroke prevention. Understanding prognostic factors for early mortality after LAAO could optimize patient selection. In the current study, we perform an in-depth analysis of 2-year mortality after LAAO, focusing particularly on potential predictors. METHODS AND RESULTS: The EWOLUTION registry is a real-world cohort comprising 1020 patients that underwent LAAO. Endpoint definitions were pre-specified, and death was categorized as cardiovascular, non-cardiovascular, or unknown origin. Mortality rates were calculated from Kaplan-Meier estimates. Baseline characteristics significantly associated with death in univariate Cox regression analysis were incorporated into the multivariate analysis. All multivariate predictors were included in a risk model. Two-year mortality rate was 16.4% [confidence interval (CI): 14.0-18.7%], with 50% of patients dying from a non-cardiovascular cause. Multivariate baseline predictors of 2-year mortality included age [hazard ratio (HR) 1.05, CI: 1.03-1.08, per year increase], heart failure (HR 1.73, CI: 1.24-2.41), vascular disease (HR 1.47, CI: 1.05-2.05), valvular disease (HR 1.63, CI: 1.15-2.33), abnormal liver function (HR 1.80, CI: 1.02-3.17), and abnormal renal function (HR 1.58, CI: 1.10-2.27). Mortality rate exhibited a gradual rise as the number of risk factors increased, reaching 46.1% in patients presenting with five or six risk factors. CONCLUSION: One in six patients died within 2 years after LAAO. We identified six independent predictors of mortality. When combined, this model showed a gradual increase in mortality rate with a growing number of risk factors, which may guide appropriate patient selection for LAAO. CLINICAL TRIAL REGISTRATION: The original EWOLUTION registry was registered at clinicaltrials.gov under identifier NCT01972282.

Transcatheter and Surgical Ductus Arteriosus Closure in Very Low Birth Weight Infants: 2018-2022.

BACKGROUND AND OBJECTIVE: The optimal patent ductus arteriosus (PDA) closure method in very low birth weight (VLBW) infants is uncertain. In 2019, the first transcatheter occlusion device was approved in the United States for infants ≥700 g. We described survival and short-term outcomes among VLBW infants who underwent transcatheter or surgical PDA closure (2018-2022). METHODS: Vermont Oxford Network members submitted data on infants born from 401 to 1500 g or 22 to 29 weeks' gestational age. Adjusted risk ratios (aRR) for survival, length of stay (LOS), prematurity complications, and discharge support were used to compare transcatheter versus surgical closure. Subgroup analyses were conducted for infants with birth weight ≥700 g and born in 2020-2022. RESULTS: Overall, 6410 of 216 267 infants at 726 hospitals received invasive PDA treatment. Transcatheter closure increased from 29.8% in 2018 to 71.7% in 2022. VLBW infants undergoing transcatheter closure had higher survival (adjusted rate ratio [aRR] 1.03; 1.02-1.04) with similar LOS (aRR 1.00; 0.97-1.03), neonatal complications (aRR 1.00; 0.98-1.01), and receipt of discharge support (aRR 0.94; 0.89-1.01). In subgroup analyses, survival (aRR 1.02; 1.00-1.04) and discharge support (aRR 0.90; 0.81-1.01) were similar between groups, whereas selected neonatal complications (aRR 0.95; 0.93-0.98) and LOS (aRR 0.95; 0.90-0.99) were lower after transcatheter closure. CONCLUSIONS: Transcatheter PDA closure in VLBW infants was increasingly used after 2018. Selected short-term outcomes for infants receiving transcatheter closure may be more favorable, compared with surgical, and warrants further clinical investigation.