Development relative to the variable patterns of multiple
ventricular septal defects
It is the changes that take place during normal development that now
underscore our understanding of the phenotypic variations to be found
amongst isolated ventricular septal defects. The inferences made from
normal development, furthermore, have now been validated by the finding
of such defects in developing mice in which the Furin enzyme had been
genetically perturbed. We are unsure precisely why perturbation of the
Furin enzyme should interfere with ventricular septation, but the
interrogation of three-dimensional datasets from developing fetuses
sacrificed at the stage when, in normal development, the ventricular
septum is intact, show the phenotypic features in different fetuses of
the different types of ventricular septal defect. Analysis of normal
development has also shown that the muscular ventricular septum is
formed by the coalescence of the components of the initial trabecular
meshwork. It is often presumed that these trabeculations come together
to form the compact layer of the ventricular walls. This is not the
case.14 The trabeculations do coalesce, nonetheless,
to form the papillary muscles of the atrioventricular valves, and also
the muscular part of the ventricular septum. In the mice in which the
Furin enzyme was perturbed, the process of septal coalescence was
disturbed, producing the arrangement also known as the Swiss-cheese
septum (Figure 3A). The inference can be made that failure of such
coalescence might be less severe, leaving discrete but multiple defects
with muscular borders at any site within the septum. In a proportion of
the mice having undergone perturbation of the Furin enzyme, there was
persistence of the tertiary embryonic interventricular communication. In
these mice, there was fibrous continuity in the postero-inferior
quadrants of the defects between the leaflets of the tricuspid and
mitral valves. This, of course, is the phenotypic feature of the
perimembranous defect.16 In a small number of the mice
with perturbation of the Furin enzyme, furthermore, there was evidence
that the proximal outflow cushions had fused so as to separate the
aortic and pulmonary roots, but had failed to muscularise. This had left
an interventricular communication between the outflow tracts that was
bordered cranially by the fused proximal outflow cushions. It is this
feature that is diagnostic for the ventricular septal defects that are
juxta-arterial.15 The inference can also be made, on
the basis of the development observed in the normal mice, and those
suffering perturbation of the Furin enzyme, that either the
perimembranous or juxta-arterial defects could co-exist with defects in
the apical muscular part of the septum. It also follows that multiple
individual defects might be anticipated to exist within the muscular
septum, or that failure of coalescence could hav been sufficiently
severe to produce the Swiss-cheese arrangement. All of these
possibilities are borne out by examination of hearts with ventricular
septal defects as found in archival collections.