RESUMO
Kawasaki disease (KD) and Multisystem Inflammatory Syndrome in Children (MIS-C) associated with COVID-19 show clinical overlap and both lack definitive diagnostic testing, making differentiation challenging. We sought to determine how cardiac biomarkers might differentiate KD from MIS-C. The International Kawasaki Disease Registry enrolled contemporaneous KD and MIS-C pediatric patients from 42 sites from January 2020 through June 2022. The study population included 118 KD patients who met American Heart Association KD criteria and compared them to 946 MIS-C patients who met 2020 Centers for Disease Control and Prevention case definition. All included patients had at least one measurement of amino-terminal prohormone brain natriuretic peptide (NTproBNP) or cardiac troponin I (TnI), and echocardiography. Regression analyses were used to determine associations between cardiac biomarker levels, diagnosis, and cardiac involvement. Higher NTproBNP (≥ 1500 ng/L) and TnI (≥ 20 ng/L) at presentation were associated with MIS-C versus KD with specificity of 77 and 89%, respectively. Higher biomarker levels were associated with shock and intensive care unit admission; higher NTproBNP was associated with longer hospital length of stay. Lower left ventricular ejection fraction, more pronounced for MIS-C, was also associated with higher biomarker levels. Coronary artery involvement was not associated with either biomarker. Higher NTproBNP and TnI levels are suggestive of MIS-C versus KD and may be clinically useful in their differentiation. Consideration might be given to their inclusion in the routine evaluation of both conditions.
RESUMO
Precision and personalized medicine, the process by which patient management is tailored to individual circumstances, are now terms that are familiar to cardiologists despite still being an emerging field. While precision medicine relies most often on the underlying biology/pathophysiology of a patient's condition, personalized medicine rely on digital biomarkers generated through algorithms. Given the complexity of the underlying data, these digital biomarkers are most often general through machine learning algorithms. There are a number of analytical considerations regarding the creation of digital biomarkers that are discussed in this review, including: data pre-processing, time dependency and gating, dimensionality reduction and novel methods both in the realm of supervised and unsupervised machine learning. Some of these considerations, such as sample size requirements and measurements of model performance are particularly challenging in small and heterogenous populations with rare outcomes like children with congenital heart disease. Finally, we review analytical considerations for the deployment of digital biomarkers in clinical settings, including the emerging field of clinical AI operations, computational needs for deployment, efforts to increase the explainability of AI, algorithmic drift and the needs for distributed surveillance and federated learning. We conclude this review by discussing a recent simulation study that shows that despite these analytical challenges and complications, the use of digital biomarkers in managing clinical care might have substantial benefits regarding individual patient outcomes.
RESUMO
IMPORTANCE AND OBJECTIVES: The current medical paradigm of evidence-based medicine relies on clinical guidelines derived from randomized clinical trials (RCTs), but these guidelines often overlook individual variations in treatment effects. Approaches have been proposed to develop models predicting the effects of individualized management, such as predictive allocation, individualizing treatment allocation. It is currently unknown whether widespread implementation of predictive allocation could result in better population-level outcomes over guideline-based therapy. We sought to simulate the potential effect of predictive allocation using data from previously conducted RCTs. METHODS AND RESULTS: Data from 3 RCTs (positive trial, negative trial, trial stopped for futility) in pediatric cardiology were used in a computational simulation study to quantify the potential benefits of a personalized approach based on predictive allocation. Outcomes were compared when using a universal approach vs predictive allocation where each patient was allocated to the treatment associated with the lowest predicted probability of negative outcome. Compared to results from RCTs, predictive allocation yielded absolute risk reductions of 13.8% (95% confidence interval [CI] -1.9 to 29.5), 13.9% (95% CI 4.5-23.2), and 15.6% (95% CI 1.5-29.6), respectively, corresponding to a number needed to treat of 7.3, 7.2, and 6.4. The net benefit of predictive allocation was directly proportional to the performance of the prediction models and disappeared as model performance degraded below an area under the curve of 0.55. DISCUSSION: These findings highlight that predictive allocation could result in improved group-level outcomes, particularly when highly predictive models are available. These findings will need to be confirmed in simulations of other trials with varying conditions and eventually in RCTs of predictive vs guideline-based treatment allocation.
Assuntos
Doenças Cardiovasculares , Simulação por Computador , Medicina de Precisão , Ensaios Clínicos Controlados Aleatórios como Assunto , Humanos , Doenças Cardiovasculares/terapia , Criança , Resultado do TratamentoRESUMO
Aims: Mathematical models previously developed to predict outcomes in patients with heart failure (HF) generally have limited performance and have yet to integrate complex data derived from cardiopulmonary exercise testing (CPET), including breath-by-breath data. We aimed to develop and validate a time-to-event prediction model using a deep learning framework using the DeepSurv algorithm to predict outcomes of HF. Methods and results: Inception cohort of 2490 adult patients with high-risk cardiac conditions or HF underwent CPET with breath-by-breath measurements. Potential predictive features included known clinical indicators, standard summary statistics from CPETs, and mathematical features extracted from the breath-by-breath time series of 13 measurements. The primary outcome was a composite of death, heart transplant, or mechanical circulatory support treated as a time-to-event outcomes. Predictive features ranked as most important included many of the features engineered from the breath-by-breath data in addition to traditional clinical risk factors. The prediction model showed excellent performance in predicting the composite outcome with an area under the curve of 0.93 in the training and 0.87 in the validation data sets. Both the predicted vs. actual freedom from the composite outcome and the calibration of the prediction model were excellent. Model performance remained stable in multiple subgroups of patients. Conclusion: Using a combined deep learning and survival algorithm, integrating breath-by-breath data from CPETs resulted in improved predictive accuracy for long-term (up to 10 years) outcomes in HF. DeepSurv opens the door for future prediction models that are both highly performing and can more fully use the large and complex quantity of data generated during the care of patients with HF.
RESUMO
Background: Diaphragm atrophy can contribute to dyspnea in patients with heart failure (HF) with its link to central neurohormonal overactivation. HF medications that cross the blood-brain barrier could act centrally and improve respiratory function, potentially alleviating diaphragmatic atrophy. Therefore, we compared the benefit of central- vs peripheral-acting HF drugs on respiratory function, as assessed by a single cardiopulmonary exercise test (CPET) and outcomes in HF patients. Methods: A retrospective study was conducted of 624 ambulatory adult HF patients (80% male) with reduced left ventricular ejection fraction ≤ 40% and a complete CPET, followed at a single institution between 2001 and 2017. CPET parameters, and the outcomes all-cause death, a composite endpoint (all-cause death, need for left ventricular assist device, heart transplantation), and all-cause and/or HF hospitalizations, were compared in patients receiving central-acting (n = 550) vs peripheral-acting (n = 74) drugs. Results: Compared to patients who receive peripheral-acting drugs, patients who receive central-acting drugs had better respiratory function (peak breath-by breath oxygen uptake [VO2], P = 0.020; forced expiratory volume in 1 second [FEV1], P = 0.007), and ventilatory efficiency (minute ventilation / carbon dioxide production [VE/VCO2], P < 0.001; end-tidal carbon dioxide tension [PETCO2], P = 0.015; and trend for forced vital capacity [FVC], P = 0.056). Many of the associations between the CPET parameters and drug type remained significant after multivariate adjustment. Moreover, patients receiving central-acting drugs had fewer composite events (P = 0.023), and HF hospitalizations (P = 0.044), although significance after multivariant correction was not achieved, despite the hazard ratio being 0.664 and 0.757, respectively. Conclusions: Central-acting drugs were associated with better respiratory function as measured by CPET parameters in HF patients. This could extend to clinically meaningful composite outcomes and hospitalizations but required more power to be definitive in linking to drug effect. Central-acting HF drugs show a role in mitigating diaphragm weakness.
Contexte: L'atrophie du diaphragme peut contribuer à la dyspnée chez les personnes atteintes d'insuffisance cardiaque (IC), compte tenu de son lien avec la suractivation neuro-hormonale centrale. Or, les médicaments contre l'IC qui franchissent la barrière hématoencéphalique pourraient exercer une action centrale, améliorer la respiration et ainsi éventuellement atténuer l'atrophie du diaphragme. C'est pourquoi nous avons voulu comparer, au moyen d'une seule épreuve d'effort cardiopulmonaire (EECP), les effets bénéfiques exercés par des médicaments à action périphérique et des médicaments à action centrale sur la fonction respiratoire, de même que l'issue des patients atteints d'IC auxquels ils ont été administrés. Méthodologie: Nous avons réalisé une étude rétrospective auprès de 624 adultes ambulatoires atteints d'IC (80 % d'hommes) dont la fraction d'éjection ventriculaire gauche était réduite (≤ 40 %), qui se sont prêtés à une EECP complète et qui ont été suivis dans le même établissement entre 2001 et 2017. Les paramètres de l'EECP et la mortalité toutes causes confondues, un critère d'évaluation composé (décès toutes causes confondues, nécessité de recourir à un dispositif d'assistance ventriculaire gauche, transplantation cardiaque), et les hospitalisations toutes causes confondues et/ou liées à l'IC ont été comparés entre les patients qui recevaient des médicaments à action centrale (n = 550) et ceux qui recevaient des médicaments à action périphérique (n = 74). Résultats: Comparativement aux patients ayant reçu des médicaments à action périphérique, ceux qui ont reçu des médicaments à action centrale ont bénéficié d'une meilleure fonction respiratoire (consommation maximale d'oxygène [VO2], p = 0,020; volume expiratoire maximal par seconde [VEMS], p = 0,007) et d'une meilleure efficacité ventilatoire (ventilation minute/production de dioxyde de carbone [VE/VCO2], p < 0,001; pression partielle de dioxyde de carbone en fin d'expiration [PETCO2], p = 0,015; et tendance de la capacité vitale forcée [CVF], p = 0,056). De plus, bon nombre des associations entre les paramètres de l'EECP et le type de médicament sont demeurées significatives après ajustement multivarié. Les patients qui ont reçu des médicaments à action centrale ont également présenté moins d'événements faisant partie du critère d'évaluation composé (p = 0,023) et moins d'hospitalisations liées à l'IC (p = 0,044), même si la différence après correction multivariée n'a pas été significative et que les rapports de risques étaient respectivement de 0,664 et de 0,757. Conclusions: Les médicaments à action centrale ont été associés à une meilleure fonction respiratoire, mesurée à l'aide des paramètres d'une EECP, chez les patients atteints d'IC. Ce résultat pourrait également s'appliquer au critère d'évaluation composé et aux hospitalisations, mais une étude plus puissante est nécessaire pour établir un lien cliniquement significatif avec l'effet des médicaments. Les médicaments à action centrale contre l'IC ont donc un rôle à jouer dans la correction de la faiblesse du diaphragme.
RESUMO
BACKGROUND: Risk assessment in pulmonary arterial hypertension (PAH) is fundamental to guiding treatment and improved outcomes. Clinical models are excellent at identifying high-risk patients but leave uncertainty amongst moderate risk patients. RESEARCH QUESTION: Can a multiple blood biomarker model of PAH, using previously described biomarkers, improve risk discrimination over current models? STUDY DESIGN AND METHODS: Using multiplex ELISA, we measured NT-proBNP, ST2, IL-6, Endostatin, Galectin-3, HDGF, and IGF binding proteins (IGFBP1-7) in train (n=1623), test (n=696) and validation (n=237) cohorts. Clinical variables, biomarkers were evaluated by principal component analysis. NT-proBNP was not included to develop an NT-proBNP independent model. Unsupervised k-means clustering classified subjects into clusters. Transplant-free survival by cluster was examined using Kaplan-Meier and Cox proportional hazard regressions. Hazard by cluster was compared to NT-proBNP, REVEAL, and ESC/ERS Risk models alone, and combined clinical and biomarker models. RESULTS: The algorithm generated 5 clusters with good risk discrimination using 6 biomarkers, weight, height, and age at PAH diagnosis. In the test and validation cohorts the biomarker model alone performed equivalent to REVEAL (AUC 0.74). Adding the biomarker model to the ESC/ERS, and REVEAL scores improved the ESC/ERS and REVEAL scores. The best overall model was the biomarker model adjusted for NT-proBNP with the best C-statistic, AIC, and calibration for the adjusted model compared to either the biomarker or NT-proBNP model alone. INTERPRETATION: A multi-biomarker model alone was equivalent to current PAH clinical mortality risk prediction models and improved performance when combined, and added to NT-proBNP. Clinical risk scores offer excellent predictive models but require multiple tests; adding blood biomarkers to models can improve prediction or enable more frequent, non-invasive monitoring of risk in PAH to support therapeutic decision making.
RESUMO
Medical advancements in the diagnosis, surgical techniques, perioperative care, and continued care throughout childhood have transformed the outlook for individuals with tetralogy of Fallot (TOF), improving survival and shifting the perspective towards lifelong care. However, with a growing population of survivors, longstanding challenges have been accentuated, and new challenges have surfaced, necessitating a re-evaluation of TOF care. Availability of prenatal diagnostics, insufficient information from traditional imaging techniques, previously unforeseen medical complications, and debates surrounding optimal timing and indications for reintervention are among the emerging issues. To address these challenges, the integration of artificial intelligence and machine learning holds great promise as they have the potential to revolutionize patient management and positively impact lifelong outcomes for individuals with TOF. Innovative applications of artificial intelligence and machine learning have spanned across multiple domains of TOF care, including screening and diagnosis, automated image processing and interpretation, clinical risk stratification, and planning and performing cardiac interventions. By embracing these advancements and incorporating them into routine clinical practice, personalized medicine could be delivered, leading to the best possible outcomes for patients. In this review, we provide an overview of these evolving applications and emphasize the challenges, limitations, and future potential for integrating them into clinical care.
De grandes avancées médicales touchant le diagnostic de la tétralogie de Fallot (TF), les techniques chirurgicales, les soins périopératoires ainsi que les soins continus au cours de l'enfance ont transformé le pronostic de cette maladie et prolongé la survie des patients, d'où la nécessité d'adopter une approche thérapeutique à long terme. Compte tenu du nombre croissant de survivants, certains défis prennent une plus grande ampleur et de nouvelles difficultés s'y ajoutent. Il convient donc de réévaluer les soins pour les patients atteints de TF. L'accès limité au diagnostic prénatal, les informations fragmentaires obtenues avec les techniques d'imagerie traditionnelles, les complications médicales inattendues et les débats sur les indications et le moment approprié pour les interventions chirurgicales subséquentes sont de nouveaux enjeux. Pour y faire face, l'intégration des outils d'intelligence artificielle (IA) et d'apprentissage automatique (AA) est prometteuse et pourrait réinventer la prise en charge des patients atteints de TF en plus d'améliorer leurs résultats à long terme. L'utilisation innovante de l'IA et de l'AA touche de nombreux aspects des soins offerts à ces patients, par exemple le dépistage et le diagnostic, l'analyse et l'interprétation automatiques d'images, la stratification du risque clinique de même que la planification et la réalisation d'interventions cardiaques. L'adoption de ces avancées technologiques et leur intégration dans la pratique clinique courante ouvrent la voie à une approche de médecine personnalisée dans l'espoir d'obtenir les meilleurs résultats possibles pour les patients. Notre article de synthèse présente ces applications en pleine évolution et met en évidence leurs perspectives d'intégration aux soins cliniques, mais aussi les défis et les limites qui accompagnent cette approche.
RESUMO
Background: The incidence of hospitalizations for cardiovascular events has been associated with specific weather conditions and air pollution. A comprehensive model including the interactions between various environmental factors remains to be developed. Objectives: The purpose of this study was to develop a comprehensive model of the association between weather patterns and the incidence of cardiovascular events and use this model to forecast near-term spatiotemporal risk. Methods: We present a spatiotemporal analysis of the association between atmospheric data and the incidence rate of hospital admissions related to heart failure (922,132 episodes), myocardial infarction (521,988 episodes), and ischemic stroke (263,529 episodes) in â¼24 million people in Canada between 2007 and 2017. Our hierarchical Bayesian model captured the spatiotemporal distribution of hospitalizations and identified weather and air pollution-related factors that could partially explain fluctuations in incidence. Results: Models that included weather and air pollution variables outperformed models without those covariates for most event types. Our results suggest that environmental factors may interact in complex ways on human physiology. The impact of environmental factors was magnified with increasing age. The weather and air pollution variables included in our models were predictive of the future incidence of heart failure, myocardial infarction, and ischemic strokes. Conclusions: The increasing importance of environmental factors on cardiovascular events with increasing age raises the need for the development of educational materials for older patients to recognize environmental conditions where exacerbations are more likely. This model could be the basis of a forecasting system used for local, short-term clinical resource planning based on the anticipated incidence of events.