Robert H. Anderson,1 Diane E.
Spicer,2 James A. Quintessenza,2Hani K. Najm,3 and Justin T.
Tretter4
1 – Cardiovascular Research Centre, Biosciences
Institute, Newcastle University, Newcastle upon Tyne, UK
2 – Heart Institute, Johns Hopkins All Children’s
Hospital, St. Petersburg, FL, USA
3 – Department of Pediatric Cardiac Surgery,
Cleveland Clinic Children’s, and The Heart, Vascular, and Thoracic
Institute, Cleveland Clinic, Cleveland, OH, USA
4 – Department of Pediatric Cardiology, Cleveland
Clinic Children’s, and The Heart, Vascular, and Thoracic Institute,
Cleveland Clinic, Cleveland, OH, USA
Key words: Aortic root; Arterial valves; Sinuses; Leaflets;
Nomenclature
Corresponding author: Justin T. Tretter, M.D., Pediatric
and Adult Congenital Heart Center, Cleveland Clinic, 9500 Euclid Avenue,
M-41, Cleveland, OH, 44195. Fax: 216-445-3692. Phone: 216-219-6374.
Email: trettej3@ccf.org
In a brief report published in the current issue of the Journal, the
group working at Brussels in Belgium describe their experience with
“bicuspidisation” of a congenitally unicuspid aortic
valve.1 We applaud the Belgian group in terms of its
meticulous reparative approach, which is based on the understanding of
the three-dimensional anatomy of the aortic root. We support the general
concepts as described in their current account as the potential way
forward in treating those with functionally unileaflet aortic valves
amenable to surgical repair. In this regard, those attempting to repair
the congenitally malformed aortic root should remember that, when
repositioning the commissures, account should also be taken of the
asymmetry of the leaflets and their sinuses.2
The general approach described by the Belgian group, which we have
adopted ourselves,3 is founded on unambiguous
communication between the imager and the surgeon. In this light, during
the review process culminating in the publication of their significant
contribution, several exchanges took place between the referee, who was
one of ourselves (RHA), and the authors. The content of the exchanges is
also pertinent to recent exchanges in the pages of JCTVS Open, in which
one of the Belgian authors was a contributor.4,5 The
exchanges related to a “consensus” document regarding the description
of the normal and abnormal aortic root.6 They reveal
that the schisms identified in the questionnaire recognising the “Tower
of Babel”, which itself was said to exist with description of the
root,7 have yet to be resolved. The reasons for the
ongoing disagreements are multiple.
In the first instance, the view obtained of the aortic root in the
operating room, or in the echocardiographic laboratory, is quite
different from that seen in the dissecting or autopsy room. This
difference can now be overcome by the sophistication provided by
clinical techniques such as computed tomography. These techniques are
able to reveal the three-dimensional features of the arterial roots as
seen in the living heart.3 A second problem relates to
the frequency with which the different abnormal aortic phenotypes are
encountered by those working in adult as opposed to pediatric cardiac
surgical practises. Also of significance is the use of words by those
who seek to establish the superiority of their own preferred
nomenclature. The authors who responded to our own letter on “the
aortic valve with two leaflets” cited the oft-used aphorism originating
from the works of Lewis Carroll.5 As they emphasised,
the question in terms of the words used is which feature is to be the
master. Should it be the general meaning of the word, or the way it is
defined by those who choose to use the given word for a specific
purpose?
In seeking to justify their need to achieve compromise rather than
consensus, the authors of the response claim to have illustrated the
deficiencies of our own chosen preference.4 There are,
however, fundamental flaws in their described
justifications.5 This now becomes of significance with
regard to the description by the Belgian authors of their approach to
“bicuspidisation”.1 The evidence provided by the
three-dimensional reconstructions,3 to which we will
return, is obviously of equal, if not greater, significance. But first
consider the linguistic problems. It was the disagreements in the use of
words that came to the fore in the “Tower of Babel” identified by the
questionnaire circulated by the German cardiac
surgeons.7 Such disagreements are unlikely properly to
be demolished until, as we suggested in our letter,4distinction is made between the moving components of the arterial roots
and the arterial valvar sinuses that support them. Our preference for
description of the moving components is to account for them as
“leaflets”. We use this word also to account for the moving components
of the atrioventricular valves. In describing their technique for
“bicuspidisation”,1 the Belgian authors have also,
on this occasion, described the moving components as “leaflets”. The
response to our letter in JCTVS Open,4 of which de
Kerchove is a co-author,5 along with the account of
bicuspidisation,1 suggests that he, and his
colleagues, would much prefer to have used the word “cusp” when
accounting for the moving components in the arterial roots.
We had pointed out the deficiency of such usage in our
letter.4 We emphasised that the authors of the
consensus document that prompted our letter6 had
themselves created confusion when suggesting that the coronary arteries
arose from the “cusps” of the aortic root. In their
response,5 the authors indicated that this was no more
than a mistake, or an “oversight”. Cardiac surgeons are well aware
that mistakes, or oversights, made in the operating room can have
catastrophic consequences. Attention to the current cardiological
literature, furthermore, particularly when used by electrophysiologists,
will show that “cusp” is used in equal fashion not only to describe
the arterial valvar leaflets, but also their supporting sinuses. The
latest issue of the European Heart Journal – Case Reports, for example,
contains a report of premature ventricular complexes ablated from the
non-coronary cusp.8 “Cusp” is similarly used to
account for the valvar sinuses in a recent publication in “Heart
Rhythm”.9 Or do the authors of the
response5 imagine that the electrophysiologists are
ablating the valvar leaflets? Perhaps this again would be no more than
an “oversight”? If so, ablation of the leaflet is unlikely to be
advantageous to post-intervention valvar function.
There are also, as we have stated, additional major flaws in the reasons
provided by the responding authors, including de Kerchove, to
substantiate their preference for “cusp” over
“leaflet”.5 In the first place, they suggest that
“the term “cusp” is specific to the semilunar valves”. There is an
additional problem here, in that the valves to which they refer are
arterial, whereas it is the hinges of their moving parts that are
semilunar. But that is a mere bagatelle when set against their assertion
that “cusp” is specific to the arterial valves. Could it be that they
have forgotten that the counterpart of the mitral valve is usually
described as the tricuspid valve? Or perhaps they are suggesting that
the valve guarding the right atrioventricular junction of the normal
heart should now be re-named as the “trileaflet” valve? We find it
unlikely that such a suggestion would find general favour. It is
incorrect, therefore, to suggest that “cusp” is limited to the moving
components of the arterial valves.
The authors do proceed accurately to account for the derivation of
“cusp”.5 As they describe, the word “indicates a
pointed end formed by the intersection of 2 arcs or curved lines that
meet (as in the tip of a spear)”.5 We agree with this
definition. But they are in error when suggesting that such arcs and
points are integral to the arterial valves, but not the atrioventricular
valves. The edges of the leaflets of the atrioventricular valves, at the
level of the atrioventricular junction, come together to produce such
meetings of curved lines (Figure 1A). These meetings are directly
comparable to those of the arterial valves at the sinutubular junction
(Figure 1B). The area that is identified by the joining of the curved
lines at the sinutubular junction, and hence correctly identified as the
“cusp”, demonstrates the peripheral end of the zone of apposition
between the adjacent leaflets. This feature is usually described as the
“commissure”. It does not account for the entirety of the moving
components of the arterial valves (Figure 1B).
Description of the “cusps” of the arterial valves in linguistically
appropriate fashion would create still further problems. This is
because, as emphasised above, the feature of the meeting of the two arcs
is usually described as the valvar commissure. These “commissures” are
the peripheral attachments, at the sinutubular junction, of the zones of
apposition between the moving components of the roots. Suggestions that
“cusp” be restricted to the moving components,5therefore, does no more than add to the potential confusion. Such
commissures are also readily identified between the moving components of
the valve itself usually described as being “tricuspid” (Figure 1A).
The reason that the leaflets of the arterial valves were initially
considered to be “cusps” is more likely to be because of their
resemblance, when viewed in closed position from their ventricular
aspect, to the surfaces of the molar and premolar teeth. The arrangement
of the leaflets of the tricuspid valve, furthermore, would present a
similar appearance when viewed in closed position from the atrial
aspect. The argument that the derivation of “cusp” justifies use of
the word to describe only the moving components of the arterial valves,
therefore, is both illogical and unjustified.
Of far greater importance, as was acknowledged by the authors of the
response to our letter,5 is the ability to distinguish
between the moving components of the arterial valves and the sinuses
that support them. The very fact that “cusp” is currently used
interchangeably to describe both features, in our opinion, disqualifies
it from use exclusively to describe the moving components. Far better to
use “leaflet” to describe these components in both sets of valves, not
least because it is the competence or incompetence of these components
in both settings that determines valvar function.
There is then another problem in the use of words that has emerged from
the recommendations of the consensus document,6 the
response to our letter,5 and the approach now taken by
those recommending “bicuspidation” for some variants of the unicuspid
and unicommissural aortic valve.1 This is in the use
of “annulus”. As was shown by the questionnaire prompting the
recognition of the “Tower of Babel”,7 not all
surgeons equated this ring with the diameter measured by
echocardiographers, better described as the virtual basal ring. This
virtual plane is created by joining together the nadirs of the moving
components. In the response to our letter,5 it was
indeed the virtual basal ring that was illustrated as representing the
“annulus”. In the described approach to “bicuspidisation”, however,
the Belgian authors now argue in favour of a “functional annulus”,
which they argue extends from the virtual basal ring to the sinutubular
junction. It would seem that this corresponds with the hinges of the
moving components. And the response to the German questionnaire showed
that some surgeons considered these attachments to represent a
“semilunar ring”.7 Could it be, therefore, that the
group are now recognising both potential candidates as the valvar
“annulus”? Such an approach must provide the potential for still
further confusion, and additional “oversights”, the more so since some
surgeons in the response to the questionnaire equated the virtual basal
ring with the ventriculo-arterial junction.7
As we indicated in our opening paragraph, the ability now to use
computed tomographic interrogation so as to demonstrate the
three-dimensional arrangement in the living heart emphasises the need to
use words that are specific for the different components that come
together to form competent arterial roots. These images show that, as
also demonstrated in the drawing provided by the authors responding to
our letter,5 the arterial roots extend between the
virtual basal ring and the sinutubular junction (Figure 2, Video 1). It
is the extent of the semilunar hinges of the moving components within
the roots that delimits these boundaries. If we understand their
definition correctly, it is these semilunar hinges that the Belgian
group now define as the “functional annulus”.1Rather than producing confusion with those who continue to recognise the
virtual basal ring as the echocardiographic annulus,7it would surely be preferable simply to describe the attachments of the
moving components as the semilunar hinges. And, as shown in Figure 2, it
is now an easy matter to reconstruct these hinges. When viewed in
three-dimensions, they produce a crown-like configuration. Equally
important, the reconstruction shows that, interposing between the
adjacent hinges, and separating the valvar sinuses, are to be found the
fibrous interleaflet triangles.10 It is the extent of
formation of these interleaflet triangles that serves to identify the
different phenotypes encountered when the arterial roots are
congenitally malformed.11 As the Belgian authors
recognise in their account, the phenotype that they treat on the basis
of “bicuspidisation” is built on the basis of a trisinuate
scaffold.3 It is because of hypoplasia and incomplete
formation of two of the three interleaflet triangles that the skirt of
leaflet tissue takes on a unicuspid and unicommissural appearance. And
it is because of the presence of a solitary zone of apposition within
the skirt of leaflet tissue that the phenotype lends itself to creation
of a new “bicuspid” valve, as described in their brief
report.1 It also follows that the commonest variants
of “bicuspid” valves are similarly built on a trisinuate scaffold, but
with incomplete formation of only one of the interleaflet
triangles.3 The variation in such bileaflet valves
housed within trisinuate roots then depends on which interleaflet
triangle is incompletely formed. In the rarest variant of the
“bicuspid” valve, there are but two sinuses formed.3This phenotype then varies according to whether the missing sinus, and
its supporting leaflet, is the non-coronary primordium or one of the
primordiums that gives rise to a coronary artery. It is a surprising
paradox that, when the interleaflet triangles are incompletely formed,
as in the so-called unicuspid variant, then the hinges of the skirt of
leaflet tissue are much more annular, rather than showing the crown-like
configuration of the normal root.
There is then an even bigger paradox emerging from the brief report of
the Belgian group,1 and the other recent publications
we have highlighted in our commentary.4-6 We all have
the same goal – namely to demolish the Tower of Babel. And we all share
the desire to use words in their most appropriate fashion. We submit,
however, that the reasons offered thus far for using “cusp”
exclusively to account for the moving components of the arterial roots
are without any logical foundation. It is spurious to suggest that the
use of “cusp” to account for the valvar sinuses rather than the moving
components is no more than an oversight. The tower will only properly be
demolished when separate words are used to account for the moving
components of the roots and the walls of the root that support them.
This, we submit, is best achieved by describing the moving parts as the
leaflets, and recognising that such moving components are also to be
found within the atrioventricular valves. In both sets of valves, it is
snug apposition of these moving parts that ensures valvar competence.
And the achievement of valvar competence is the major goal of cardiac
surgeons when operating on malformed valves, be they arterial or
atrioventricular.