Arrhythmogenic prolapse of the mitral valve is a result of the increased mechanical stress placed on the subvalvar apparatus and surrounding myocardium by prolapsing leaflets and elongated tendinous cords, leading to localised myocardial fibrosis readily detected using cardiac magnetic resonance.1,2 Individuals who experience sudden cardiac death with this problem are predominantly women with prolapse of both leaflets, ventricular arrhythmias originating within the left ventricle, with those originating from the outflow tract alternating with others having papillary muscle or fascicular origin and frequent repolarisation abnormalities recognisable in the inferior electrocardiographic leads.3–5 Mitral regurgitation, and the Pickelhaube sign – defined as an increased peak systolic lateral velocity greater than 16 m/s measured across the atrioventricular junction – are also predictors of ventricular arrhythmias.5,6 Ventricular arrhythmias occur more frequently in patients with replacement fibrosis, but are not associated with the grade of mitral regurgitation.1,7
However, it appears that the presence of ventricular arrhythmias in patients with prolapsing valves in the absence of regurgitation, in addition to the underlying myocardial fibrosis, is the expression of more pronounced morphologic and mechanical abnormalities of the valvar apparatus.8 In this regard, prominent mitral annular disjunction (Figure 1 ) has also been associated with sudden cardiac death.2,9 Exaggerated disjunction is increasingly recognised in association with prolapse. This has led to the suggestion of a genetic link between the two, which frequently run together in families. The genes FLNA, DCHS1 and DZIP1 have all been implicated, along with links to TGF-β pathway genes, such as SMAD3 and TGFBR2. These latter genes are known to underscore heritable aortic diseases and connective tissue disorders, but explain only a small proportion of instances of prolapse.10–12
An association between arrhythmic events and disjunction has been suggested to be independent of the presence of prolapse. This has led some to propose that the exaggerated disjunction itself may be an arrhythmogenic entity,6 although this has not been shown to translate into excess mortality.13 In the Montalcino Aortic Consortium Registry, furthermore, mitral valvar regurgitation and prolapse – but not disjunction – were associated with adverse aortic and cardiac events.11 Those seeking to implicate disjunction as a causative factor, however, ignore the fact that the feature is a ubiquitous finding in the heart when the metrics of the hinge of the mural leaflet are assessed within the overall atrioventricular junction, albeit not always in the same place.14,15
Preliminary data suggest that catheter ablation and valvar repair, especially in young patients, might reduce the risk of malignant arrhythmia.5,16 In a recent study, however, the risk of arrhythmic events in individuals with both disjunction and valvar prolapse did not decrease after successful reparative surgery.17 It seems, therefore, that sudden cardiac death and arrhythmias do occur in association with exaggerated disjunction and valvar prolapse, but at a frequency insufficient to establish causality. At present, more data are needed to better prove and define the arrhythmogenic potential of a largely normal cardiac finding.
The term ‘left ventricular non-compaction’ is currently used frequently to account for an arrangement of the ventricular mural anatomy characterised by a prominent trabecular myocardial component relative to the width of the compact layer, with deep intertrabecular recesses extending from the ventricular cavity but separated from the epicardial coronary arteries.18,19 The classical clinical presentation is heart failure, arrhythmias and embolic events.20 The prevalence of such alleged “non-compaction” is currently estimated to be from 0.01 to 0.05% in adults.19,21 Uncertainty with regard to the assessment of trabeculations that can occur as a response to increased loading conditions, however, makes it difficult to define a true prevalence.22
The non-compaction phenotype has been observed in the setting of cardiomyopathy, leading to suggestions of an association with sarcomere gene mutations, particularly MYH7, MYBPC3 and TTN.23–26 It has also been observed in individuals having aortic valves with two leaflets, again prompting proposals of a genetic link between the two conditions, with MIB1–NOTCH argued to play a crucial role.26 The identification of a pathogenic mutation on genetic testing or a family history, for those with bicuspid aortic valves has been proposed to support a definitive diagnosis of non-compaction.25 If the condition were truly linked genetically, however, the question would need to be asked as to why more individuals with bicuspid aortic valves did not also have non-compaction.
The phenotype currently recognised as non-compaction, however, is to be found in healthy subjects having left ventricles of normal size and function, and it can be acquired and reversible. Of note with regard to the notion of non-compaction, no evidence has been provided thus far to show that the compact components of the ventricular walls are produced by compaction of the pre-existing trabecular layer. Indeed, there is now significant evidence to indicate that the feature is the consequence of the ongoing proliferation of the trabecular layer that is prominent during embryonic development (Figure 2 ).27 Further evidence indicates that the feature, properly described as excessive trabeculation, is a morphologic trait, and one which can be found in the healthy population. As such, it is hardly surprising that it can be found in individuals suffering different cardiomyopathies, rather than representing a distinct cardiomyopathy in its own right.18,28–30 Such excessive trabeculations fulfilling the diagnostic criteria for what is termed non-compaction cardiomyopathy, when seen in the general population, therefore, can represent a benign phenomenon. They can also be seen in athletes, patients with sickle cell anaemia, and pregnant women, in whom they resolve within the next 2 years.18,28,31,32 The feature is becoming increasingly recognised with the advent of newer imaging technologies that identify the excessive trabeculations more readily, such as resonance imaging or computed tomography. The clinical significance of these findings remains unclear, but there is significant potential for overdiagnosis.21 The extent of trabeculations measured in end-diastole in asymptomatic individuals recruited in the MESA study, furthermore, was not associated with any deterioration in left ventricular volumes or function during a 10-year period.32 In patients with excessive trabeculation, the extent of the trabecular layer when detected using resonance imaging has less prognostic significance for adverse cardiac events than left ventricular dilation, systolic dysfunction, or fibrosis.21,33 The thickness of the trabecular, as opposed to the compact, layers of the ventricular walls in the adult is determined by the differential – or allometric – growth of each myocardial layer. Because the trabeculated components of the developing walls do not coalesce to form the compact layers, therefore, the term “left ventricular noncompaction” has no scientific basis.
In our efforts to identify substrates for cardiac disease, and especially arrhythmias, it is possible we tend to over-emphasise the extremes of the range of normal characteristics. In so doing, we may create the impression of imaginary conditions. Disjunction and what is termed non-compaction could well represent two notable examples. On the other hand, it might be argued that increased identification of the relevant populations with malignant mitral valvar prolapse, or non-compaction, might be expanded by the increased use of biobanks, aided by the ever-increasing resolution of clinical phenotyping, and through the integration of genetic and genomic analysis.11,34 Such an approach might allow for clearer delineation between variants, particularly those without clinical consequences, differentiating them from the variants associated with added risk. If such a policy proves to be correct, it would certainly permit development of more accurate descriptors, replacing the current broad, potentially inaccurate categorisations and, hence, provide a tighter nosology. Until we are able to identify such conditions, we must appreciate that the features of so-called mitral annular disjunction and excessive trabeculation are most probably epiphenomena, and need to be recognised as such.