CARDIOVASCULAR JOURNAL OF AFRICA • Vol 22, No 1, January/February 2011
40
AFRICA
need for revascularisation procedures,
30
or CAD by myocardial
perfusion scan.
14
However, there is a higher frequency of angina
compared to case-matched controls and it is more common in
patients with LVH.
30
Angina is present in 13 to 20% of patients,
and equal among men and women.
9,31,33
This is not to say that patient’s with Fabry’s are not more
susceptible to thrombotic or embolic phenomena. In one series
of Japanese patients there was a high incidence of throm-
botic events (9/65 with strokes).
34
Despite a lack of significant
coronary artery disease on angiograms, in one series, diffuse
hypo-echogenic plaques were more common in Fabry’s patients
compared to age-matched controls by intravascular ultrasound.
35
It has also been shown that coronary flow is reduced in patients
with Fabry’s despite normal peripheral endothelial function.
36
However, patients with hypertrophic cardiomyopathy also have
slow coronary flow compared to controls.
37
Therefore the exact
aetiology of angina in patients with Fabry’s is still open to debate.
Arrhythmias
Patients with Fabry’s are more prone to both atrial and ventricu-
lar arrhythmias due to glycosphingolipid deposition and fibrosis
as well as atrial dilatation and relative ischaemia secondary to
LVH. However, in one series,
9
both systolic and diastolic blood
pressures were normal, implying that atrial arrhythmias, in
particular atrial fibrillation, were not secondary to long-standing
hypertension or renal disease but to Fabry’s. Interestingly, in this
series arrhythmias were 1.5 times more likely in males (although
other cardiac events were similar),
9
possibly due to more
advanced cardiomyopathy and renal dysfunction in hemizygotes
versus heterozytgotes.
The most common ECG finding is voltage criteria for LVH,
13
although various degrees of block as well as PR-interval short-
ening have also been described.
38
Atrial arrhythmias, e.g. atrial
fibrillation are more common than ventricular arrhythmias. In
a case series of 78 patients over 10 years, 13% had paroxysmal
atrial fibrillation (which was four times that of the general popu-
lation for age) and 8% had non-sustained ventricular tachycardia
(VT); all patients with VT had LVH. Predictors for atrial fibrilla-
tion were age, left atrial size, LV wall thickness, LV mass index
and angina. Permanent pacemakers were implanted in 10.6% of
patients for complete heart block or symptomatic bradycardia.
39
An international series of 714 patients also found that the
incidence of arrhythmias was increased in patients with LVH.
31
Corresponding to observational reports that women die from
cardiac aetiology, ventricular arrhythmias were more common
in women, 20 versus 14%, in a large cohort of 1 448 patients.
32
Valvular disease
Mitral and aortic valve abnormalities were present in 57 and
47% of one series (where LVH was present in 61%), although no
severe regurgitation was noted.
23
Another larger series documents
mitral valve regurgitation in 32% of patients.
32
In a series of 111
patients, there were no cases of severe valvular disease, includ-
ing nine patients with end-stage disease. The most common
valvular abnormality was mild mitral regurgitation (
n
=
57).
40
The incidence of Fabry’s patients undergoing valve replacement
surgery is low.
31
Therefore valves are not the most significantly
affected tissue of the heart in patients with Fabry’s and do not
cause much burden.
Evaluating for cardiac involvement
In a retrospective cohort analysis of patients with amyloidosis,
Fabry’s, hypertrophic obstructive cardiomyopathy and hyper-
tensive heart disease patients, no single clinical characteristic,
ECG finding or echocardiographic feature could differentiate
between the various causes of LV hypertrophy, including echo-
genicity, valvular abnormalities, renal dysfunction and diastolic
dysfunction. However, painful neuropathy, anhydrosis, lack of
hypertension and presence of Sokolow-Lyon criteria for LVH on
ECG were significant for Fabry’s disease by univariate analysis.
Furthermore, if none of hypertension, orthostasis, pericardial
effusion or a papillarymuscle anomaly was present, the sensitivity
and specificity for Fabry’s disease was 92 and 87%, respectively.
41
Echocardiography
As previously described, the echocardiogram shows varying
degrees of hypertrophy and usually a preserved systolic func-
tion, with LVH more often seen in older individuals.
23
Diastolic
dysfunction, sometimes displaying a restrictive pattern, can be
present,
23
but peak E velocity, peak A velocity and decelera-
tion time of the mitral valve are most often normal,
28
and dias-
tolic dysfunction is usually not present in the absence of LVH.
42
Furthermore, diastolic dysfunction does not distinguish Fabry’s
cardiomyopathy from hypertrophic obstructive cardiomyopathy.
43
As previously discussed, valvular abnormalities, most often
mitral and aortic, are not associated with severe regurgitation.
23
A case-control series of 40 consecutive patients showed that
82.5% of Fabry’s patients – 94% of Fabry’s patients with LVH
– had a ‘binary appearance’ of the endocardial border, which
was not present in any matched hypertrophic cardiomyopthy
patient, hypertensive or otherwise, representing a sensitivity and
specificity of 94 and 100%, respectively, for detecting Fabry’s
cardiomyopathy. Furthermore, pathological examination of the
‘binary’ areas revealed endocardium and myocardium laden
with glycosphingolipids, separated by a subendocardial ‘empty
space’.
43
However, relying on the ‘binary sign’ has been challenged
recently by Kounas
et al
., where they found the sensitivity and
specificity of the binary appearance on echocardiogram to differ-
entiate Fabry’s from hypertrophic obstructive cardiomyopathy to
be 35 and 79%, respectively. Furthermore, only 3.5% of patients
with LV wall thickness less than 15 mm had a binary sign, albeit
the number of subjects examined was small.
44
A recent small,
blinded study reported that a binary appearance of the endocar-
dium on echocardiography has a sensitivity of only 15.4%, but a
specificity of 73.3%.
45
Tissue Doppler imaging (TDI)
Although two-dimensional echocardiogram cannot be used to
screen patients with Fabry’s disease for signs of cardiac involve-
ment before the development of LVH, the addition of TDI may
be reliable in detecting sub-clinical involvement. In a case-
control series of 20 patients with Fabry’s (half with LVH) and 10
control patients, those with Fabry’s showed reduced contraction
and relaxation even before the development of LVH. Lateral or
septal systolic velocities (Sa)
<
10 cm/s or early diastolic veloci-
ties (Ea)
<
10 cm/s each showed a sensitivity and specificity of
100% in mutation-positive patients without LVH.
46
