Abstract: Background：There is limited research on long-term follow-up studies of isolated ventricular septal defect (VSD) from the fetal to postnatal period at home and abroad. Objective：To provide data support for prenatal counseling and prognosis assessment of isolated VSD. Design：Retrospective cohort study. Methods：This study included children with isolated VSD, who were diagnosed at the outpatient clinic of the Cardiac Center of Guangdong Women and Children Hospital from January 2017 to December 2019 and had regular postnatal follow-up and documented records. Those with maternal termination of pregnancy, chromosomal or genetic abnormalities (identified through amniocentesis or umbilical blood sampling), or those who were lost to follow-up for more than 1 year were excluded. A routine follow-up form was established to collect the maternal age, gestational age, VSD diameter and location, and aortic diameter during the first echocardiographic examination. Follow-up echocardiograms were conducted within 48 hours after birth, at 3 months, at 6 months, and annually thereafter, until March 31, 2023. Patients were categorized into the spontaneous closure group, surgical closure group, and persistent defect group. Main outcome measures：Closure rate and closure time of isolated VSD for spontaneous or surgical closure. Results：A total of 264 cases were included in this study, including 154 cases (58.3%) in the spontaneous closure group, 78 cases (29.5%) in the surgical closure group, and 32 cases (12.1%) in the persistent defect group. The males were 109（41.3%）. At the time of VSD detection, the gestational age of the fetus was 28.3 (20-38) weeks, and the defect diameter was 2.6 (1.0-7.5) mm. There were 75 cases of muscular VSD (28.4%), 186 cases of perimembranous VSD (70.5%) and 3 cases of subarterial VSD. In total, 60 cases were large (22.7%), 179 cases were medium (67.8%), and 25 cases were small (9.5%).The 264 cases were followed up for 7 (0.1-69) months. The closure time was 8 (0-48) months in the spontaneous closure group and 4 (1-53) months in the surgical closure group.The time of followup was 45 (38-69) months in the persistent defect group. The proportions of musclar VSD and small VSD in the spontaneous closure group were higher than those in the surgical closure group and the persistent defect group. The proportions of perimembranous VSD and large VSD, R-value and defect diameter in the spontaneous closure group were lower than those in the surgical closure group and the persistent defect group. The follow-up time of the spontaneous closure group was longer than that of the surgical closure group and shorter than that of the persistent defect group, and the differences were statistically significant. The spontaneous closure rates of muscular VSD and perimembranous VSD were 92.0% and 45.7%, respectively, with a statistically significant difference. The spontaneous closure rates of large VSD, medium VSD and small VSD were 26.7%, 63.1% and 100%, respectively, and the differences of spontaneous closure rates between small and medium VSD,and between small and large VSD were statistically significant. In the surgical close group of 78 patients, 23 cases of medium VSD (12.8% of all medium VSD cases) progressed to large VSD by the time of operation. Conclusion：Perimembranous VSD was the most common type of VSD in fetuses, followed by muscular VSD. Spontaneous closure was observed in 58.3% of VSD cases, with higher rates in muscular VSD and small VSD. An increase in size was observed in 12.8% of medium VSD cases after birth.

Key words: Echocardiography, Fetus, Ventricular septal defect., Muscular, Perimembranous