Abstract
BACKGROUND AND OBJECTIVE: Congenital heart disease (CHD) is the most common congenital anomaly and a leading cause of infant mortality. Fetal echocardiography (FE) serves as the cornerstone of prenatal screening, yet significant heterogeneity exists in its diagnostic performance. This review aims to critically appraise the current status, technological advancements, and persistent challenges in FE for diagnosing CHD. Focusing on the central issue of improving detection rates, it outlines a future direction toward an integrated, intelligent screening ecosystem. METHODS: Following a narrative review methodology, a systematic literature search was conducted in PubMed, Web of Science for articles and CNKI (China National Knowledge Infrastructure) published from January 2019 to May 2025. Search terms incorporated fetal echocardiography, CHD, screening/diagnosis, and related advanced technologies. Inclusion criteria covered relevant original studies, reviews, and guidelines, while editorials and articles with unavailable full texts were excluded. Literature screening, data extraction, and bias risk assessment were performed independently by two researchers. KEY CONTENT AND FINDINGS: Standardized acquisition of key views (e.g., four-chamber, outflow tracts) forms the foundation of screening. Technology-enhanced modalities such as three-dimensional/four-dimensional spatiotemporal image correlation (3D/4D STIC) and speckle tracking offer incremental diagnostic value for specific defects. Artificial intelligence (AI) demonstrates transformative potential in automating view identification, anomaly detection, and even community-based screening. However, diagnostic efficacy remains significantly hampered by operator dependency, limitations of screening protocols, and the inherent complexity of certain CHD types, resulting in a wide variation in detection rates ranging from 60% to 90%. CONCLUSIONS: FE, particularly comprehensive FE, is indispensable for the prenatal diagnosis of CHD. Future success hinges on constructing a tiered, integrated intelligent ecosystem. This involves leveraging AI tools to standardize basic screening, combining specialized training with targeted use of advanced modalities, and precisely directing complex cases to regional diagnostic and care centers.