Role of Echocardiography in Amyloid Cardiomyopathy

Introduction: Cardiac Amyloidosis is a rapidly progressive infiltrative cardiomyopathy associated with significant mortality and morbidity. It is characterised by the deposition of amyloid fibrils, misfolded protein deposits, in the myocardium and there are two types described. One, light chain amyloidosis (AL), is due to myocardial deposition of monoclonal immunoglobulin light chain proteins produced by bone marrow plasma cells and is rare. The second, transthyretin amyloidosis (ATTR), is caused by a transport protein that is synthesized primarily by the liver. ATTR has two sub-types, hereditary or variant (vATTR) and wild-type (wtATTR), with wildtype being the significantly more prevalent of the two.

AL and ATTR have different clinical courses, prognosis, and potential treatments. Cardiac involvement is common in both, occurring in up to 75% of AL cases and it is the dominant clinical feature in ATTR patients. ATTR has a slower and more varied clinical course with a median survival of around 3-5 years if untreated, compared to the more rapid progression of AL, with an untreated median survival of 6 months.

Previously, treatment for ATTR cardiomyopathy (ATTR-CM) was solid organ transplantation but the new effective treatments and emerging therapies have increased the importance of recognising it at an early stage.

While the gold standard for diagnosis of the disease is myocardial biopsy, echo plays a major role in the diagnosis. However, echo by itself is insufficient to make the diagnosis with other investigations required to confirm the presence of cardiac amyloid and differentiate between AL and ATTR. Consensus guidelines recommend it should be combined with electrocardiographic, clinical, biomarker, and other imaging findings to maximize diagnostic accuracy.1 Some proposed scoring systems combine ECG, echo, and biomarker measurements. One group has suggested a combination of troponin T >30ng/L, left ventricular wall thickness >13.6mm and QRS width >120ms had strong positive and negative predictive value.2 They also highlight the importance of other imaging modalities with potential sequential use of echo, cardiac MRI, and SPECT scanning to aid in the diagnosis and the differentiation between AL and ATTR cardiomyopathies.

Whilst echocardiography is widely available, it does lack the tissue characterisation ability of MRI, so echo relies on the presences of highly suggestive features which include.

• Increased LV and RV wall thickness

• Biatrial enlargement and thickening of the interatrial septum

• Evidence of stage II diastolic dysfunction or worse

• Reduced mitral annular velocities on tissue doppler imaging

• Reduced global longitudinal strain

• Low flow-low gradient aortic stenosis

• Diffuse valve thickening

• Pericardial effusion

One of the challenges with the often-quoted list2 is that these findings are quite non-specific and the typical features, such as a markedly increased wall thickness and a restrictive filling pattern may not be present until advanced stages of the disease.

Increased wall thickness, or left ventricular hypertrophy (LVH), occurs in other conditions, such as hypertension, hypertrophic cardiomyopathy (HCM) or indeed other infiltrative conditions such as Fabry’s disease. A wall thickness >12mm, without another plausible cause, is utilised and this represents only a mild thickening which may be present in undiagnosed hypertension. LVH is also a key feature of significant aortic stenosis (AS), a very plausible cause, but it has been shown that ATTR cardiomyopathy can be present in 16% of elderly TAVI patients,3 so we need to maintain an appropriate level of suspicion, even in cases of mild LVH or in cases of AS.

Whilst challenging to differentiate LVH due to amyloid from other causes, amyloid cardiomyopathy is associated with small chamber sizes and low stroke volumes, compared to hypertensive heart disease. A relative wall thickness (calculated as twice the posterior wall thickness divided by LV diastolic diameter) of >0.42 can be suggestive of amyloid cardiomyopathy. In true LVH an increase in size of the muscle cells leads to an increase in voltage, i.e., larger amplitude QRS complexes on ECG. However, in amyloid, as the wall thickness is due to infiltration, there is an ECG-LVH mismatch, with the QRS amplitudes not reflecting the increased wall thickness seen. This mismatch can also raise a suspicion that the increased thickness is due to cardiac amyloid. A key feature of HCM is asymmetrical hypertrophy and we would also expect large QRS amplitudes on ECG. It is proposed that a septal to posterior wall thickness <1.6 combined with a summed QRS amplitude of <30mm could differentiate between hypertrophic and amyloid cardiomyopathies.4 The combination of right and left ventricular hypertrophy is a classic finding seen in cardiac amyloid but not in other cases of left ventricular hypertrophy. The pattern of LVH seen on echo can help differentiate AL from ATTR cardiomyopathy with the former demonstrating concentric pattern of LVH compared to the latter where septal hypertrophy is more likely.

The classic phenotype of amyloid cardiomyopathy is heart failure with preserved ejection fraction (HFpEF) due to the infiltrative process leading to significant diastolic dysfunction. However, this is not an essential element as this may only manifest in more advanced stages of the disease. Patients with cardiac amyloid often have stage II or stage III (a restrictive filling pattern) diastolic dysfunction on echocardiography. This is seen as reduced E wave deceleration times and raised E:A ratios on transmitral doppler (Fig 1), raised E/e’ measurements, and left atrial (LA) enlargement. Amyloid cardiomyopathy is also associated with reduced measurements on tissue doppler imaging with S’, e’ and a’ all measuring <5cm/s (Fig 2). Echo grading of diastolic dysfunction can be complex and the above measurements, in combination with the velocity of tricuspid regurgitation should be assessed and classified according to current guidelines.5 LA enlargement in diastolic dysfunction is due to atrial remodelling in the setting of raised LA pressures, secondary to raised LV diastolic pressures. We also see atrial dilatation in the setting of atrial fibrillation, which is also commonly associated cardiac amyloid.

The increasingly used tool in echocardiography of global longitudinal strain (GLS) is useful in the diagnosis of amyloid cardiomyopathy. GLS is a unitless measure of longitudinal LV function and can demonstrate a reduction before a frank reduction in ejection fraction can be measured. GLS tracks the echo “speckles” within the myocardium. In an actively contracting segment of myocardium, speckles will move closer to each other. Reflecting this shortening, the rate of shortening is expressed as a negative value, with more negative values indicating more deformation or shortening. Thus, a segment showing a GLS of -20 is contracting more than a segment with a value of -10. There are software differences amongst vendors resulting in variances in quoted normal values. In general values more negative than -18 to -20 are considered normal, but it is recommended that serial studies on a patient are carried out on the same manufacturer machine as the original study. In a GLS study each of the seventeen echo segments from basal to apical is given a score and an average GLS is also calculated.

In amyloid cardiomyopathy, the GLS “bullseye” plot, can show a classical apical sparing, where basal and mid GLS decreases with normal values in the apex (Fig 3). This pattern is seen in both AL and ATTR cardiomyopathy and is sometimes referred to as having a “cherry on top” appearance, as brighter red values signify greater shortening. The exact mechanism for this is not fully understood but an accepted hypothesis is that the amyloid fibrils are preferentially deposited in the basal and mid segments of the ventricles. Several ratios using GLS, such as average apical to average basal and mid values; septal apical to basal values; and EF to average GLS have been studied.

GLS can be prone to errors in its performance and the apical sparing pattern is not exclusive to amyloid cardiomyopathy, therefore, alone it is insufficient to make the diagnosis.

Where there is echocardiographic evidence or a clinical suspicion of amyloid cardiomyopathy the echo report should explicitly comment on the likelihood of amyloidosis based on the imaging as not suggestive, strongly suggestive, or equivocal for cardiac amyloid.

From an echocardiography point of view AL and ATTR cardiomyopathies demonstrate similar echo findings, with perhaps the only differentiator being the asymmetrical hypertrophy seen more commonly in ATTR. For that reason, those in the field of echocardiography should be aware of some key clinical features that would raise a suspicion of ATTR cardiomyopathy. The most striking is that nearly 50% have bilateral carpal tunnel syndrome.6 They can also present with peripheral and autonomic neuropathies. As the amyloid deposits also affect the cardiac conduction tissue, they can develop arrhythmias and high grade atrio-ventricular block.

Amyloid cardiomyopathy is associated with significant mortality and morbidity and with new and emerging treatments early recognition is important. Echocardiography plays a central role in this and whilst no one feature is diagnostic, key red-flags should raise a suspicion when you see unexplained LVH, particularly in older patients. We should also remember the association between aortic stenosis and ATTR cardiomyopathy, keeping this dual diagnosis in mind in the presence of other features.

Classic echo findings include left and right ventricular hypertrophy, biatrial enlargement, significant diastolic dysfunction with reduced annular tissue doppler velocities and the apical sparing pattern seen in GLS. One should remember that no single feature is diagnostic for the disease, but rather it is a constellation of findings. Those working in echo should have an index of suspicion for amyloid cardiomyopathy in the setting of HFpEF and/or the echo features described in this article.

About the author

Paul Nolan is a Clinical Lecturer on the Clinical Measurement Physiology programmes in Atlantic Technological University and is a Cardiac Physiologist of over 20 years’ experience. He is a holds certification from the European Association of Cardiovascular Imaging in Adult Echocardiography.