MixEHR-SurG: A joint proportional hazard and guided topic model for inferring mortality-associated topics from electronic health records.

Survival models can help medical practitioners to evaluate the prognostic importance of clinical variables to patient outcomes such as mortality or hospital readmission and subsequently design personalized treatment regimes. Electronic Health Records (EHRs) hold the promise for large-scale survival analysis based on systematically recorded clinical features for each patient. However, existing survival models either do not scale to high dimensional and multi-modal EHR data or are difficult to interpret. In this study, we present a supervised topic model called MixEHR-SurG to simultaneously integrate heterogeneous EHR data and model survival hazard. Our contributions are three-folds: (1) integrating EHR topic inference with Cox proportional hazards likelihood; (2) integrating patient-specific topic hyperparameters using the PheCode concepts such that each topic can be identified with exactly one PheCode-associated phenotype; (3) multi-modal survival topic inference. This leads to a highly interpretable survival topic model that can infer PheCode-specific phenotype topics associated with patient mortality. We evaluated MixEHR-SurG using a simulated dataset and two real-world EHR datasets: the Quebec Congenital Heart Disease (CHD) data consisting of 8211 subjects with 75,187 outpatient claim records of 1767 unique ICD codes; the MIMIC-III consisting of 1458 subjects with multi-modal EHR records. Compared to the baselines, MixEHR-SurG achieved a superior dynamic AUROC for mortality prediction, with a mean AUROC score of 0.89 in the simulation dataset and a mean AUROC of 0.645 on the CHD dataset. Qualitatively, MixEHR-SurG associates severe cardiac conditions with high mortality risk among the CHD patients after the first heart failure hospitalization and critical brain injuries with increased mortality among the MIMIC-III patients after their ICU discharge. Together, the integration of the Cox proportional hazards model and EHR topic inference in MixEHR-SurG not only leads to competitive mortality prediction but also meaningful phenotype topics for in-depth survival analysis. The software is available at GitHub: https://github.com/li-lab-mcgill/MixEHR-SurG.

hART: Deep learning-informed lifespan heart failure risk trajectories.

BACKGROUND: Heart failure (HF) results in persistent risk and long-term comorbidities. This is particularly true for patients with lifelong HF sequelae of cardiovascular disease such as patients with congenital heart disease (CHD). PURPOSE: We developed hART (heart failure Attentive Risk Trajectory), a deep-learning model to predict HF trajectories in CHD patients. METHODS: hART is designed to capture the contextual relationships between medical events within a patient's history. It is trained to predict future HF risk by using the masked self-attention mechanism that forces it to focus only on the most relevant segments of the past medical events. RESULTS: To demonstrate the utility of hART, we used a large cohort containing healthcare administrative data from the Quebec CHD database (137,493 patients, 35-year follow-up). hART achieves an area under the precision-recall of 28% for HF risk prediction, which is 33% improvement over existing methods. Patients with severe CHD lesion showed a consistently elevated predicted HF risks throughout their lifespan, and patients with genetic syndromes exhibited elevated HF risks until the age of 50. The impact of the birth condition decreases on long-term HF risk. The timing of interventions such as arrhythmia surgery had varying impacts on the lifespan HF risk among the individuals. Arrhythmic surgery performed at a younger age had minimal long-term effects on HF risk, while surgeries during adulthood had a significant lasting impact. CONCLUSION: Together, we show that hART can detect meaningful lifelong HF risk in CHD patients by capturing both long and short-range dependencies in their past medical events.

Outcomes After Pediatric Pulmonary Valve Replacement in Patients With Tetralogy of Fallot.

BACKGROUND: People with tetralogy of Fallot might require a transannular patch during primary repair, which leads to pulmonary valve regurgitation. Pulmonary valve replacement (PVR) is performed to prevent complications of chronic pulmonary regurgitation, but the optimal timing of PVR remains a matter of debate. In this study we aimed to assess the association of PVR performed in children younger than 18 years of age with the rate of hospitalizations, interventions, and mortality. METHODS: This was a retrospective observational cohort of people with tetralogy of Fallot born in Québec between 1982 and 2015, using a combination of clinical and administrative data. Marginal means and rates models and survival curves were used to compare outcomes of patients with pediatric PVR (younger than 18 years) and those without. Outcomes of interest were rates of cardiac hospitalizations, all-cause hospitalizations, cardiac interventions, and mortality. Groups were balanced using models weighed on the inverse probability of receiving pediatric PVR. RESULTS: Of the 316 eligible patients, 58 (18.4%) received a pediatric PVR. Compared with patients who did not receive pediatric PVR, they were at increased risk of cardiac hospitalizations, although the rates of cardiac hospitalization were low: 0.50 vs 0.09 hospitalizations per 20 years (hazard ratio [HR] = 4.71 [95% confidence interval (CI), 2.22-9.96]). Patients who received a pediatric PVR had a comparable risk of all-cause hospitalizations (HR = 0.95 [95% CI, 0.71-1.26]) and of cardiac interventions (HR = 1.13 [95% CI, 0.72-1.77]). CONCLUSIONS: Patients who underwent pediatric PVR had higher rates of cardiac hospitalizations, but similar rates of all-cause hospitalizations, cardiac procedures, and mortality. In this observational cohort, pediatric PVR was not associated with an improved outcome.

CanCHD Study of Hematopoietic Cancers in Children With and Without Genetic Syndromes.

BACKGROUND: Individuals with genetic syndromes can manifest both congenital heart disease (CHD) and cancer attributable to possible common underlying pathways. To date, reliable risk estimates of hematopoietic cancer (HC) among children with CHD based on large population-based data remain scant. This study sought to quantify the risk of HC by the presence of genetic syndrome among children with CHD. METHODS AND RESULTS: Data sources were the Canadian CHD database, a nationwide database on CHD (1999-2017), and the CCR (Canadian Cancer Registry). Standardized incidence ratios were calculated for comparing HC incidences in children with CHD with the general pediatric population. A modified Kaplan-Meier curve was used to estimate the cumulative incidence of HC with death as a competing risk. A total of 143 794 children (aged 0-17 years) with CHD were followed up from birth to age 18 years for 1 314 603 person-years. Of them, 8.6% had genetic syndromes, and 898 HC cases were observed. Children with known syndromes had a substantially higher risk of incident HC than the general pediatric population (standardized incidence ratio, 13.4 [95% CI, 11.7-15.1]). The cumulative incidence of HC was 2.44% (95% CI, 2.11-2.76) among children with a syndrome and 0.79% (95% CI, 0.72-0.87) among children without a syndrome. Acute myeloid leukemia had a higher cumulative incidence during early childhood than acute lymphoblastic leukemia. CONCLUSIONS: This is the first large population-based analysis documenting that known genetic syndromes in children with CHD are a significant predictor of HC. The finding could be essential in informing risk-stratified policy recommendations for cancer surveillance in children with CHD.

Machine Learning Informed Diagnosis for Congenital Heart Disease in Large Claims Data Source.

Background: With an increasing interest in using large claims databases in medical practice and research, it is a meaningful and essential step to efficiently identify patients with the disease of interest. Objectives: This study aims to establish a machine learning (ML) approach to identify patients with congenital heart disease (CHD) in large claims databases. Methods: We harnessed data from the Quebec claims and hospitalization databases from 1983 to 2000. The study included 19,187 patients. Of them, 3,784 were labeled as true CHD patients using a clinician developed algorithm with manual audits considered as the gold standards. To establish an accurate ML-empowered automated CHD classification system, we evaluated ML methods including Gradient Boosting Decision Tree, Support Vector Machine, Decision tree, and compared them to regularized logistic regression. The Area Under the Precision Recall Curve was used as the evaluation metric. External validation was conducted with an updated data set to 2010 with different subjects. Results: Among the ML methods we evaluated, Gradient Boosting Decision Tree led the performance in identifying true CHD patients with 99.3% Area Under the Precision Recall Curve, 98.0% for sensitivity, and 99.7% for specificity. External validation returned similar statistics on model performance. Conclusions: This study shows that a tedious and time-consuming clinical inspection for CHD patient identification can be replaced by an extremely efficient ML algorithm in large claims database. Our findings demonstrate that ML methods can be used to automate complicated algorithms to identify patients with complex diseases.

Clinical Risk Assessment and Prediction in Congenital Heart Disease Across the Lifespan: JACC Scientific Statement.

Congenital heart disease (CHD) comprises a range of structural anomalies, each with a unique natural history, evolving treatment strategies, and distinct long-term consequences. Current prediction models are challenged by generalizability, limited validation, and questionable application to extended follow-up periods. In this JACC Scientific Statement, we tackle the difficulty of risk measurement across the lifespan. We appraise current and future risk measurement frameworks and describe domains of risk specific to CHD. Risk of adverse outcomes varies with age, sex, genetics, era, socioeconomic status, behavior, and comorbidities as they evolve through the lifespan and across care settings. Emerging technologies and approaches promise to improve risk assessment, but there is also need for large, longitudinal, representative, prospective CHD cohorts with multidimensional data and consensus-driven methodologies to provide insight into time-varying risk. Communication of risk, particularly with patients and their families, poses a separate and equally important challenge, and best practices are reviewed.

Navigating meaningful engagement: lessons from partnering with youth and families in brain-based disability research.

BACKGROUND: While patient and family engagement in research has become a widespread practice, meaningful and authentic engagement remains a challenge. In the READYorNot™ Brain-Based Disabilities Study, we developed the MyREADY Transition™ Brain-Based Disabilities App to promote education, empowerment, and navigation for the transition from pediatric to adult care among youth with brain-based disabilities, aged 15-17 years old. Our research team created a Patient and Family Advisory Council (PFAC) to engage adolescents, young adults, and parent caregivers as partners throughout our multi-year and multi-stage project. MAIN BODY: This commentary, initiated and co-authored by members of our PFAC, researchers, staff, and a trainee, describes how we corrected the course of our partnership in response to critical feedback from partners. We begin by highlighting an email testimonial from a young adult PFAC member, which constituted a "critical turning point," that unveiled feelings of unclear expectations, lack of appreciation, and imbalanced relationships among PFAC members. As a team, we reflected on our partnership experiences and reviewed documentation of PFAC activities. This process allowed us to set three intentions to create a collective goal of authentic and meaningful engagement and to chart the course to get us there: (1) offering clarity and flexibility around participation; (2) valuing and acknowledging partners and their contributions; and (3) providing choice and leveraging individual interests and strengths. Our key recommendations include: (1) charting the course with a plan to guide our work; (2) learning the ropes by developing capacity for patient-oriented research; (3) all hands on deck by building a community of engagement; and (4) making course corrections and being prepared to weather the storms by remaining open to reflection, re-evaluation, and adjustment as necessary. CONCLUSIONS: We share key recommendations and lessons learned from our experiences alongside examples from the literature to offer guidance for multi-stage research projects partnering with adolescents, young adults, and family partners. We hope that by sharing challenges and lessons learned, we can help advance patient and family engagement in research.

Tetralogy of Fallot Across the Lifespan: A Focus on the Right Ventricle.

Tetralogy of Fallot is a cyanotic congenital heart disease, for which various surgical techniques allow patients to survive to adulthood. Currently, the natural history of corrected tetralogy of Fallot is underlined by progressive right ventricular (RV) failure due to pulmonic regurgitation and other residual lesions. The underlying cellular mechanisms that lead to RV failure from chronic volume overload are characterized by microvascular and mitochondrial dysfunction through various regulatory molecules. On a clinical level, these cardiac alterations are commonly manifested as exercise intolerance. The degree of exercise intolerance can be objectified and aid in prognostication through cardiopulmonary exercise testing. The timing for reintervention on residual lesions contributing to RV volume overload remains controversial; however, interval assessment of cardiac function and volumes by echocardiography and magnetic resonance imaging may be helpful. In patients who develop clinically important RV failure, clinicians should aim to maintain a euvolemic state through the use of diuretics while paying particular attention to preload and kidney function. In patients who develop signs of cardiogenic shock from right heart failure, stabilization through the use of inotropes and pressor is indicated. In special circumstances, the use of mechanical support may be appropriate. However, cardiologists should pay particular attention to residual lesions that may impact the efficacy of the selected device.

De nombreuses techniques chirurgicales permettent aux patients présentant une tétralogie de Fallot (TF), une forme de cardiopathie congénitale, de survivre jusqu'à l'âge adulte. À l'heure actuelle, l'évolution naturelle de la TF corrigée est caractérisée par une insuffisance ventriculaire droite (VD) progressive attribuable à une régurgitation pulmonaire et à d'autres lésions résiduelles. Les mécanismes cellulaires sous-jacents qui mènent à l'insuffisance VD due à une surcharge volumique chronique sont caractérisés par une dysfonction microvasculaire et mitochondriale faisant intervenir diverses molécules régulatrices. Sur le plan clinique, ces atteintes cardiaques se manifestent par une intolérance à l'effort qui peut être évaluée au moyen d'une épreuve d'effort cardiorespiratoire, ce qui permet de faciliter l'établissement d'un pronostic. Le moment propice pour une réintervention en cas de lésions résiduelles contribuant à la surcharge volumique du ventricule droit demeure controversé; toutefois, il peut être utile d'évaluer régulièrement la fonction et les volumes cardiaques au moyen d'une échocardiographie et de tests d'imagerie par résonance magnétique. En présence d'une insuffisance VD cliniquement importante, les cliniciens doivent tenter de maintenir les patients dans un état euvolémique en utilisant des diurétiques, tout en accordant une attention particulière à la précharge et à la fonction rénale. Si les patients manifestent des signes de choc cardiogénique associé à une insuffisance cardiaque droite, il convient de leur administrer des inotropes et des vasopresseurs pour stabiliser leur état. Dans certains cas, l'utilisation d'un dispositif d'assistance mécanique peut être appropriée. Cependant, les cardiologues doivent être attentifs aux lésions résiduelles, car elles peuvent influencer l'efficacité de ce dispositif.

A commentary on the healthcare transition policy landscape for youth with disabilities or chronic health conditions, the need for an inclusive and equitable approach, and recommendations for change in Canada.

There is a growing number of youth with healthcare needs such as disabilities or chronic health conditions who require lifelong care. In Canada, transfer to the adult healthcare system typically occurs at age 18 and is set by policy regardless of whether youth and their families are ready. When the transition to adult services is suboptimal, youth may experience detrimental gaps in healthcare resulting in increased visits to the emergency department and poor healthcare outcomes. Despite the critical need to support youth with disabilities and their families to transition to the adult healthcare system, there is limited legislation to ensure a successful transfer or to mandate transition preparation in Canada. This advocacy and policy planning work was conducted in partnership with the Patient and Family Advisory Council (PFAC) within the CHILD-BRIGHT READYorNot™ Brain-Based Disabilities (BBD) Project and the CHILD-BRIGHT Policy Hub. Together, we identified the need to synthesize and better understand existing policies about transition from pediatric to adult healthcare, and to recommend solutions to improve healthcare access and equity as Canadian youth with disabilities become adults. In this perspective paper, we will report on a dialogue with key informants and make recommendations for change in healthcare transition policies at the healthcare/community, provincial and/or territorial, and/or national levels.

Born to Age: When Adult Congenital Heart Disease Converges With Geroscience.

Advances in imaging, catheter-based interventions, congenital heart disease surgery, and clinical management of congenital heart disease (CHD) have yielded a dramatic change in age distribution of the CHD population. This implores clinicians and researchers to gain a better understanding of aging, as this will be the cornerstone to how we plan and manage this rapidly evolving group of patients. In this article, we first review the demographic changes in the CHD population and then describe the systemic complications of disease observed in young patients with CHD, following which we discuss general concepts in aging that may be transferable to the CHD population. Finally, we review inflammation and its potential impact on aging. We provide a new lens on aging in CHD and its functional consequences in CHD, with the goal of stimulating an exchange of knowledge between geroscientists and CHD.

Artificial Intelligence in Congenital Heart Disease: Current State and Prospects.

The current era of big data offers a wealth of new opportunities for clinicians to leverage artificial intelligence to optimize care for pediatric and adult patients with a congenital heart disease. At present, there is a significant underutilization of artificial intelligence in the clinical setting for the diagnosis, prognosis, and management of congenital heart disease patients. This document is a call to action and will describe the current state of artificial intelligence in congenital heart disease, review challenges, discuss opportunities, and focus on the top priorities of artificial intelligence-based deployment in congenital heart disease.