Prenatal diagnosis of cardiac rhabdomyoma: implications for predicting tuberous sclerosis complex and guiding perinatal management.

objectiveTo identify prenatal predictors of tuberous sclerosis complex (TSC) in fetuses with one or more cardiac rhabdomyomas (CR), evaluate an integrated multimodal diagnostic workflow using fetal magnetic resonance imaging (MRI) and trio whole-exome sequencing (trio-WES) and characterize perinatal outcomes.

methodsThis was a retrospective cohort study of 80 fetuses that were diagnosed prenatally with one or more CR between February 2016 and December 2024 at a single tertiary center and that received a definitive TSC diagnosis either prenatally (via genomic or clinical criteria) or by postnatal pathological verification. All fetuses underwent routine high-resolution echocardiography. Those in which a CR was suspected were then offered the option of further investigation via fetal brain MRI and trio-WES, to identify pathogenic variants in TSC1/TSC2 genes. A definitive prenatal TSC diagnosis was assigned based on genetic testing or the presence of two major clinical features. Clinical, imaging, genetic and perinatal data were retrieved and compared between the TSC-positive and TSC-negative groups. Postnatal or postmortem verification was only required in cases that lacked a definitive prenatal diagnosis. Performance metrics, including sensitivity, specificity and positive (PPV) and negative (NPV) predictive values, were calculated to evaluate the diagnostic utility of tumor multiplicity and fetal brain MRI in prediction of TSC.

resultsForty-eight (60.0%) fetuses were diagnosed with TSC and 32 (40.0%) were TSC-negative. Tumor multiplicity had a sensitivity of 87.5% (95% CI, 74.8-95.3%) but low specificity (50.0%; 95% CI, 31.9-68.1%) for the prediction of TSC. Seventy fetuses underwent brain MRI. Among these, MRI achieved a specificity of 100.0% (95% CI, 89.1-100.0%) and identified central nervous system lesions in 89.4% of those with TSC that underwent the MRI examination, of which 81.0% were occult on ultrasound examination. Trio-WES was performed in 31 cases. Among these, pathogenic variants were identified in 20 cases, with 85.0% of these affecting the TSC2 gene. Trio-WES identified pathogenic variants in five TSC-positive cases that presented with normal fetal brain MRI, demonstrating its essential additive value for fetuses with occult imaging phenotypes, while MRI detected structural lesions in four trio-WES-negative cases. TSC2 variants were associated with more severe phenotypes, including subependymal giant cell astrocytoma, whereas TSC1 variants typically presented with milder manifestations. Sixty-two fetuses underwent termination of pregnancy (TOP); the TOP rate was 87.5% in the TSC-positive group and 62.5% in the TSC-negative group. Eighteen fetuses were liveborn (six TSC-positive and 12 TSC-negative). The TSC-positive infants had significantly lower birth weight compared with the TSC-negative infants (median, 3050 g vs 3500 g; P = 0.025), despite the two groups having a similar gestational age at delivery. All liveborn TSC-positive infants developed postnatal epilepsy.

conclusionMultiple CR at screening ultrasound examination and a major lesion at fetal brain MRI are key prenatal predictors of TSC, but accurate diagnosis requires not only MRI but also trio-WES, to overcome the developmental limitations imposed by the late appearance of some fetal anomalies associated with TSC and genetic limitations inherent in standard molecular analysis. Given the severe perinatal outcomes associated with TSC observed in this study, including high termination rates and postnatal epilepsy in all liveborn infants, definitive prenatal diagnosis is essential for accurate risk stratification, parental counseling and proactive neurological management. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.