The Natural History of Ventricular Septal Defects

Abstract & Commentary

Synopsis: Ventricular septal defects (VSDs) are relatively common and are perimembranous, muscular, or rarely doubly committed in location. Muscular defects account for about 60% of VSDs. These defects are often small, have few hemodynamic abnormalities, and often close spontaneously. Perimembranous VSDs are often associated with hemodynamic abnormalities. They are less likely to spontaneously close and so often require surgery.

Source: Turner SW, et al. The natural history of ventricular septal defects. Arch Dis Child 1999;81:413-416.

Turner and associates set out to investigate the incidence and natural history of clinically detectable ventricular septal defects (VSDs) and designed a longitudinal, population-based cohort study with a mean follow-up of 76 months. Entry into the study was based upon clinical suspicion or recognition of VSD rather than routine echocardiography without signs or symptoms.

VSDs can be classified based on their size and their location. Turner et al used a standard classification system whereby VSDs were described as being perimembranous, muscular, or doubly committed. The latter refers to a group of VSDs seen primarily in Asian populations, where the defect is located directly beneath the aortic and pulmonary valves. This type of VSD accounts for less than 5% of the VSDs seen in North American children. Echocardiography was used to verify the presence of the defect, to locate its position in the septum and to determine its size. Small defects were defined as those VSDs that were seen only during part of the cardiac cycle in two-dimensional imaging or those that were only seen with the use of color Doppler mapping.

The study was carried out in the northern United Kingdom, where there is a stable population of about 3.1 million people. Virtually all pediatric cardiology patients in this part of the United Kingdom are referred to a centralized pediatric cardiology service. Local GPs were asked to immediately refer all infants and children in whom a VSD was known or suspected.

Closure of the VSD was determined to have occurred under three situations: 1) surgical intervention; 2) on follow-up, neither two-dimensional echocardiography or color Doppler could identify a persistent defect; 3) there was no murmur of VSD at follow-up.

Muscular defects comprised nearly 60% of all VSDs and perimembranous defects accounted for about 40%. No doubly committed defects were seen among the total of 68 VSDs. A perimembranous defect is more likely to be hemodynamically significant than a muscular defect and so more often requires surgery (39% vs 3%). Moreover, perimembranous defects are less likely to undergo spontaneous resolution when compared to muscular defects (29% vs 69%). It is important to note that almost 11% of the perimembranous VSDs that were not large ultimately required surgical closure because of associated aortic valve prolapse or right ventricular outflow obstruction, two known complicating features of perimembranous VSDs.

Comment by Alan Friedman, MD, FAAP

It has long been known that VSDs are the most common of the congenital heart defects when the bicuspid aortic valve is excluded. In fact, with the introduction of highly sensitive color Doppler mapping imaging techniques, the incidence of VSD in the newborn population has been reported to be between 2-4.2%. A good number of these defects are small and undergo spontaneous closure over time, and, therefore, require no intervention or therapy. The results of this study by and large corroborate our clinical experience as pediatricians: muscular defects tend to be small and usually get smaller, remaining hemodynamically insignificant or spontaneously close; perimembranous defects are less likely to close and are more likely to require surgical intervention. This latter finding may result from the defect’s size and the concomitant volume load with or without a pressure load on the pulmonary vasculature, or closure may be indicated because of the development of associated problems such as aortic valve prolapse.

This study also confirms the fact that VSDs are common. In this study, isolated VSDs were found to occur in 1.76/1000 live births. This incidence is lower than in studies that use echocardiography with color Doppler mapping instead of auscultation as an entry criterion because color Doppler will undoubtedly recognize "tiny," clinically inaudible muscular VSDs, which have a high rate of closure in the first year of life.

We as pediatricians should recognize that VSDs are common in our patients and when they are present in the muscular part of the septum, they have a high rate of spontaneous closure, usually before the age of 6 years, and a low incidence of hemodynamic alteration. In contrast, perimembranous VSDs tend to be larger and are more likely to lead to clinically relevant hemodynamic abnormalities. Perimembranous VSDs require more diligent long-term follow-up, as there may be the development of associated aortic valve disease.

Ventricular septal defects:

a. have similar prognoses regardless of the location of the defect.

b. have a high rate of spontaneous closure when located in the muscular septum.

c. rarely are hemodynamically significant when the membranous septum is involved.

d. continue to close spontaneously at about the same rate throughout childhood and adolescence.