Closure of ventricular septal defects: a study of factors influencing spontaneous and surgical closure

Introduction: Aspects of the management of ventricular septal defects which remain uncertain include the role of screening, and the need for closure of defects that remain patent. Aim: To ascertain the natural history of clinically significant ventricular septal defects, and to examine uncertainties In strategies of management. Methods: We studied prospectively a regional cohort of children born with isolated ventricular septal defects, noting age at presentation; sex; morphology and size of the defect; age at closure, if it occurred, and the means of closure. Results: We identified 290 children with isolated ventricular septal defects, of whom 147 (51%) were male. The mean follow up period was 65 months, with five cases being lost to follow up. Surgical closure was required in 41 (14%) cases, and 155 (54%) defects closed spontaneously. Spontaneous closure occurred in 123 (68%) out of 180 cases with completely muscular borders, and in 31 (29%) of the 107 patients with perimembranous defects. There were two cases with associated aortic regurgitation and no cases of endocarditis. Size and morphology of the defect were significant predictors of spontaneous and surgical closure. In addition, young age at diagnosis was a significant predictor of spontaneous closure. Conclusions: The size and morphology of a ventricular septal defect are important determinants of spontaneous closure and to the need for surgical intervention. Early age at presentation, in contrast, is not predictive of the need for surgical intervention. In early childhood, there appears to be very little risk of endocarditis or aortic valvar prolapse. Neither screening of populations to identify ventricular septal defects, nor surgical closure of asymptomatic defects in childhood, are supported by our findings.