Pediatric Cardiology最新文献

Multiparametric cardiac magnetic resonance (CMR) is increasingly used for rejection and coronary artery vasculopathy (CAV) surveillance in pediatric heart transplant recipients (PHTR). There is limited data regarding how graft assessment by multiparametric CMR may reflect functional capacity in PHTR. To explore the relationship between multiparametric CMR and markers of exercise capacity in PHTR. PHTR who underwent CMR within 1 year of cardiopulmonary exercise testing (CPET) were retrospectively reviewed. Those with submaximal effort on CPET (respiratory exchange ratio < 1.10), depressed function (left ventricular ejection fraction (LVEF) < 50% and/or right ventricular ejection fraction (RVEF) < 45%), or significant clinical events (rejection, new or worsening CAV, cardiac hospitalizations) between CMR and CPET were excluded. CMR variables included biventricular volumes, ejection fraction, cardiac index (CI), myocardial T2, T1/extracellular volume fraction (ECV), and myocardial perfusion reserve index (MPRI). CPET variables were VO_2peak, O₂ pulse, percent age-predicted maximum heart rate (APMHR), HR reserve, and exercise duration. Relationships between variables were studied using correlations and regression. Forty-seven PHTR were included. Time between CPET and CMR was 5.5 ± 3.4 months. CI correlated positively with O₂ pulse (R = 0.642, p < 0.001). Global T1 correlated negatively with both APMHR (R = - 0.538, p < 0.001) and HR reserve (R = - 0.598, p < 0.001). Global T2 correlated negatively with APMHR (R = - 0.335, p = 0.049), HR reserve (R = - 0.488, p = 0.003), and VO_2peak (R = - 0.327, p = 0.045). In PHTR with normal LV function, CMR-derived tissue characteristics have correlations with exercise capacity. Larger studies are needed to understand the role of multiparametric CMR in the functional surveillance of PHTR.

This letter comments on a recent study that explored the use of an artificial intelligence language model to generate educational materials for parents of children undergoing cardiac catheterization. The study demonstrated improved understanding and satisfaction among parents after reading the ChatGPT-produced information sheets. While the use of language models in patient education represents an innovative and promising approach, several methodological aspects merit attention. The lack of a control group makes it difficult to determine whether the improvement in understanding was specific to ChatGPT or could have been achieved through other written materials. Additionally, the use of parametric analysis on ordinal Likert data may limit the validity of statistical inferences. Despite these limitations, the study contributes valuable insight into how artificial intelligence can enhance the accessibility and efficiency of patient education. Creating readable materials aligned with the educational level of patients or caregivers has been shown to support better comprehension and engagement. Future research using controlled designs and appropriate statistical methods will help clarify the specific benefits of AI-generated educational tools and further advance their role in patient-centered communication.

Tetralogy of Fallot (TOF) is a common cyanotic congenital heart defect whose surgical repair has improved long-term survival. However, many patients develop right ventricular outflow tract (RVOT) dysfunction and severe pulmonary insufficiency (PI), often leading to right ventricular dilatation and arrhythmias. Pulmonary valve replacement (PVR) is the standard treatment. Open surgery remains the preferred option in many cases, especially when the RVOT is large or has a complex anatomy. However, in patients with significant comorbidities that increase surgical risk, percutaneous options may be a viable alternative. In the case we present, an innovative approach was used, as the patient was not a candidate for surgery due to his high comorbidity (bladder neoplasm, obesity and severe obstructive respiratory pattern) and his giant RVOT posed a challenge for percutaneous valve options. A large Myval^® 32 mm prosthesis was implanted, occupying the rest of the RVOT with a vascular stent in parallel occluded by a vascular plug device. This approach, although uncommon, has proven to be a viable palliative option in selected cases, providing an improvement in right ventricular function and reducing symptoms.

The application of artificial intelligence (AI) to congenital heart disease (CHD) has become a rapidly expanding field, promising to transform diagnostics, prognostics, and management. However, a comprehensive map of the global research landscape, its key contributors, thematic structure, and evolutionary trends is currently lacking. This study provides a bibliometric analysis to delineate this dynamic domain. We conducted a systematic search of the Web of Science Core Collection for all publications related to AI in CHD up to December 31, 2024. Bibliometric analyses were performed using VOSviewer and the Bibliometrix R package to map publication trends, international collaboration networks, leading contributors, and the conceptual structure of the field through co-occurrence and thematic mapping. Our analysis of 500 publications revealed an exponential growth in research output since 2020. The United States, China, and the United Kingdom were the most productive countries, forming central hubs in a robust international collaboration network. Harvard University and the University of London were the leading institutions. Thematic analysis identified six dominant research clusters: (1) AI-enhanced surgical/interventional support, (2) prenatal diagnosis via imaging, (3) disease-specific outcome modeling (e.g., Tetralogy of Fallot), (4) advanced imaging analytics with deep learning, (5) comprehensive risk and outcome management, and (6) pediatric cardiology applications. Temporal analysis confirmed a decisive recent shift towards deep learning methodologies, and analysis of highly cited works underscored the impact of AI in imaging and prognostic modeling. The convergence of AI and CHD has matured into a vibrant, clinically-focused research field with a clear technological trajectory towards deep learning. This global research panorama highlights established strengths in imaging and diagnostics while pointing to emerging frontiers in personalized medicine and interventional support. Fostering greater collaboration and focusing on clinical translation, model explainability, and equity will be crucial for realizing the full potential of this AI-driven transformation in improving care for patients with CHD.

Although percutaneous atrial septal defect closure is widely used, the surgical method may be preferred in cases with conditions that complicate transcatheter closure (large defect, rim deficiency or floppy rim, presence of septal aneurysm, multifenestrated defects, low age, and weight). This retrospective study evaluated data from patients with complex atrial septal defects who underwent transcatheter closure between January 2021 and August 2024. The parameters assessed included age, weight, defect size, rim size, septum characteristics, success rates, and complication rates. Transcatheter atrial septal defect closure was performed in 51 patients with complex defects. Two patients had multifenestrated defects, five had multiple defects, and 15 had an aneurysmal septum. Six patients were under one year old, and 21 had defects > 20 mm. In 25 patients, the stretched balloon size/total septum ratio was > 50%, in 44 patients, the balloon diameter/body surface area ratio was > 15 mm/m², and one had a malalignment defect. Among the patients with rim deficiency, six had isolated aortic, six had isolated inferior vena cava, five had isolated superior vena cava, and one had isolated posterior rim deficiency. Three patients had combined aortic and inferior vena cava deficiencies, and one had combined aortic and superior vena cava deficiency. The study revealed no complications except in two cases of device embolization. The success rate was 96.08% across all procedures. Percutaneous closure can be successfully performed in patients with complex atrial septal defects with low complication rates, similar to other atrial septal defects, with good results in experienced centers.

The arterial switch operation (ASO) is the standard surgical intervention to correct transposition of the great arteries (TGA). The aim of this study is to describe long-term outcomes after the ASO in patients with TGA, and to explore anatomical and clinical factors potentially associated with adverse outcomes, comparing simple and complex TGA. Since December 2001 to October 2023, 233 patients underwent arterial switch operation (158 simple TGA and 75 complex TGA). Kaplan-Meier analysis assessed survival and reoperation risk, Chi-square test and T-student test have been performed in order to analyse pre-, intra- and post-operative data among the two groups. In-hospital mortality was 1.7%. Overall survival rates were 97.9%, 96.8% and 96.8% at 5, 10 and 15 years, respectively. Freedom from surgical reoperation at 15 years was 98.3% in simple TGA and 91% in complex TGA at 15 years (p = 0.039). Reinterventions were infrequent and mainly related to pacemaker implantation and right ventricular outflow tract obstruction. Coronary artery stenosis occurred in 5 patients (2.2%). Risk factors analysis suggested that anatomical complexity such as intramural coronary artery (p = 0.003), aortic arch hypoplasia (p = 0.004), aortic arch interruption (p = < 0.001) and coronary reimplantation with button technique (p = 0.018) are predictive factors for mortality. In addition, complex TGA (p = 0.022) and post-operative aortic regurgitation (p < 0.001) are identified as potential predictive factors for reintervention. Arterial switch operation provides excellent long-term survival and low reintervention rates in TGA. These results support tailored long-term surveillance, particularly for patients with complex anatomy or coronary anomalies.

Genetic testing plays an expanding role in pediatric cardiology, particularly in congenital heart disease (CHD). Despite advancements in genomic medicine, no universal guidelines exist for genetic evaluation in CHD, resulting in varied practices, especially within Cardiac Intensive Care Units (CICUs). This study examines genetic evaluation practices for neonates with CHD cared for in pediatric CICUs across the United States. A two-tier survey of CICU medical directors and genetic specialists was distributed via RedCap to 118 centers contributing to the Society of Thoracic Surgeons Congenital Heart Surgery Database. Surveys assessed CICU demographics, genetic evaluation practices, and institutional protocols. Descriptive statistics were analyzed using R. Fifty unique institutions (42%) completed the first-tier survey; 46/50 completed the second-tier. Most (78%) respondents reported dedicated CICUs, but only 32% had formal genetic testing protocols. Tier 1 showed genetic testing was reported in 58% of neonatal CHD cases, the remainder evaluated on a case-by-case basis. Substantial variability existed in both cardiac substrates prompting genetic evaluation and first-line test selection. Only 28% reported consultation with genetic providers for neonatal CHD cases. Tier 2 showed genetic consultation increased to 54% for patients with syndromic CHD, and genetic testing was obtained in 46% of isolated and 61% of syndromic CHD. Access remained limited as only 42% had cardiac-specific genetic providers. Fewer than 12% of institutions used exome or genome as first-line testing. Genetic evaluation in CICUs is highly variable, with underutilization of genetic providers and limited access to advanced testing. Standardized guidelines and consensus statements are needed to improve access, streamline care, optimize outcomes, and enable genome-informed management in CHD.

Pulmonary vein stenosis (PVS) is a challenging diagnosis, especially in premature infants. Echocardiography is often used as the primary screening tool in this population, but its accuracy in identifying PVS requiring intervention remains incompletely defined. We aimed to evaluate echocardiographic parameters associated with the need for pulmonary vein intervention in premature infants with suspected PVS. We performed a retrospective, single-center study evaluating premature infants with PVS requiring intervention (balloon angioplasty or stent placement) from 2012 to 2022. Patients with significant congenital heart disease, anomalous pulmonary venous return, or prior pulmonary vein stenting were excluded. The last echocardiogram prior to catheterization was reviewed, and pulmonary veins were analyzed individually. Parameters assessed included aliasing, subjective narrowing, pulmonary vein mean gradient, pulmonary vein variability index (PVVI), and Doppler tracing returning to baseline. 16 unique patients and 31 pre-catheterization echocardiograms (150 pulmonary veins) were included. Doppler tracing not returning to baseline was sensitive but not specific in identifying veins requiring intervention. In contrast, narrowing and aliasing were more specific but less sensitive. When defining a pulmonary vein as abnormal based on any of the three qualitative parameters, sensitivity and negative predictive values were highest, though specificity remained low. Pulmonary vein gradients and PVVI demonstrated decreasing sensitivity and increasing specificity with increasing thresholds. Echocardiography is a valuable tool to identify pulmonary veins requiring intervention, though no single parameter demonstrated optimal sensitivity and specificity. Pulmonary vein gradients and PVVI may further support identification of abnormal veins requiring intervention.

Congenital heart disease (CHD) is the most common congenital malformation, with most cases exhibiting a multifactorial etiology involving genetic and environmental factors. Congenital anomalies of the atrioventricular valve or septum (CAAVAS) and functionally univentricular heart (FUH) are complex subtypes of CHD, where disruptions in key molecular pathways are implicated. This study investigates the genetic burden contributing to these anomalies. This case-control study included 48 participants: 24 patients diagnosed with CAAVAS or FUH and 24 healthy controls. Whole-exome sequencing (WES) was conducted to assess genetic burden by evaluating minor allele frequencies (MAF) using gnomAD and predicting functional impact of variants with REVEL scores. A secondary filtration was performed, focusing on 349 genes associated with abnormal heart valve morphology (HP:0001654) as defined by the Human Phenotype Ontology (HPO) database, to identify pathogenic variants exclusive to the case group. Genetic burden risk (GBR) analysis revealed a significantly higher median number of common variants in the case group compared to controls (p = 0.035). Genetic analysis identified variants in genes involved in contractile cardiac and cytoskeletal proteins (MYH3, ACTC1), extracellular matrix proteins (FBN1, FREM1, HSPG2), ciliary proteins (EVC2, PKD1L1), enzymes (POLG, DNASE1L3), Cell-signaling proteins (TGFB2, CCDC22) and transcription factors (NKX2-5, NONO). This study highlights the significant role of genetic burden and gene variants associated with congenital mitral and tricuspid valve anomalies. Our findings reinforce the strong genetic predisposition underlying these malformations, as evidenced by the increased genetic burden in affected individuals compared to controls without CHD.