A 35-year-old man is admitted for evaluation of non-ST-elevation myocardial infarction (STEMI) and heart failure. He has a history of leukemia treated with chemotherapy and radiation as a child, but has not had a cardiac evaluation in 10 years. He had been well until he developed dyspnea with exertion that progressed to dyspnea at rest over two days. At presentation, his blood pressure is 80/60mmHg, his heart rate is 110bpm, and his oxygen saturation is 91% in room air. Examination demonstrates bibasilar crackles, elevated jugular venous pressure, and a summation gallop without murmur. His extremities are notable for bilateral lower-extremity edema and he is cool to the touch from foot to knee. Chest X-ray demonstrates bilateral effusions, pulmonary edema, and cardiomegaly. Electrocardiogram (ECG) shows relatively low-voltage and non-specific ST changes (see Figure 1). Laboratory data demonstrate a troponin of 1.0mg/dl (upper limit of normal 0.4mg/dl). He is taken urgently to cardiac catheterization out of concern that his symptoms could be related to an acute coronary syndrome. His coronary arteries are normal; however, his left ventricular end-diastolic pressure (LVEDP) is 30mmHg. An echocardiogram is ordered to evaluate global LV systolic function and diastolic function and to assess in more detail the reason(s) for the high filling pressures.
A transthoracic echo study demonstrates moderately decreased LV systolic function with mild biventricular enlargement and moderately increased LV and right ventricular (RV) wall thickness (see Figure 2). It is noted that the myocardium appears bright and granular. The mitral and tricuspid valves appear thickened, with only mild regurgitation. Both atria are significantly dilated (see Figure 3). There is a small pericardial effusion and bilateral pleural effusions. On spectral Doppler recordings, the mitral E velocity is elevated, early deceleration time is shortened, and mitral A velocity is decreased (see Figure 4). The patient is admitted and given intravenous (IV) diuretics and standard heart failure therapies, with rapid clinical improvement. Based on his history and echocardiographic findings, the possibility of amyloidosis is raised. Urine and serum protein electrophoreses are performed, demonstrating a monoclonal immunoglobulin G (IgG) lambda spike concerning for amyloid. Colonic biopsy shows amyloid deposition, and bone marrow biopsy demonstrates plasma cell dyscrasia. Further biopsies are planned to help distinguish between AA and AL amyloid, as the therapeutic and prognostic implications are different.
Amyloidosis is a disorder marked by deposition of amyloid material in tissues. Amyloid fibrils are precursors of a variety of serum proteins, and the type of protein is related to the underlying etiology of the disease. There are many types of amyloidosis, some acquired and some hereditary, with different underlying pathophysiologies and prognoses. AA and AL amyloidosis are the most common types seen in adults and in tertiary care centers. AL (primary) amyloidosis is due to deposition of monoclonal protein fragments, and is most commonly associated with plasma cell dyscrasias such as multiple myeloma. AA (secondary) is due to deposition of acute-phase protein fragments, and is most commonly associated with chronic inflammatory conditions such as rheumatic diseases or chronic infections that produce acute-phase responses and result in markedly increased serum amyloid A protein. Specific organ involvement depends on the type of amyloid. Although many types will affect the skin, gastrointestinal (GI) tract, or kidneys, cardiac involvement is typical of AL amyloid but not of AA amyloid.
The prognosis of AL amyloid is different from that of AA amyloid. Patients with AL amyloid have a poor prognosis,1 particularly if there is cardiac involvement. The underlying plasma cell dyscrasia is difficult to treat, and the protein deposition is generally not reversible. Therapies include melphalan and prednisone, and in patients without cardiac involvement stem cell transplantation may be an option. Conversely, in patients with AA amyloid treatment of the underlying inflammatory process will decrease circulating serum amyloid A proteins, and partial resolution of the protein deposition may be seen.2
In the case described, the clinical suspicion of amyloidosis was raised on the basis of the echocardiographic findings. Although there is no pathognomic echocardiographic feature that makes the diagnosis of amyloid likely, a suite of features (discussed below) should raise suspicion. Definitive diagnosis is made by serum and urine protein electrophoresis, bone marrow biopsy, and biopsy of affected organs. Even in those patients with ‘classic’ echocardiographic findings suggesting amyloidosis, the specific type or underlying pathology is not discernable by cardiac imaging.
Amyloid deposition in the heart results in a restrictive cardiomyopathy. Restrictive cardiomyopathies are often due to systemic disorders that result in tissue infiltration. In amyloidosis, protein infiltrates the myocardium, while in hemachromatosis it is iron deposition that causes myocardial infiltration and leads to restriction. Although cardiac amyloid may be a restrictive process with normal to small cavity size and preserved systolic function early in the disease, ventricular dilation and systolic dysfunction can develop in the later stages.
Both 2D and Doppler findings are useful in the diagnosis of cardiac amyloid. On 2D imaging, LV wall thickness will be increased, as may RV wall thickness. Wall thickening is due to abnormal protein deposition rather than hypertrophy of the myocytes; therefore, ‘hypertrophy’ is not an accurate description. This concept helps to explain the typical ECG as well, which demonstrates low voltage (see Figure 1) in spite of a marked increase in LV wall thickness on echocardiography. In a multivariate analysis of patients with biopsy-proven amyloid, the combination of increased septal thickness >1.98cm and low voltage on ECG had a positive predictive value of 79% and a negative predictive value of 88%.3
On echocardiography, the myocardium may appear bright and is sometimes termed ‘speckled’ or ‘granular’ or as appearing like a ‘starry, starry night’ (see Figure 2). Each of these features in isolation is not suggestive of amyloid. Increased wall thickness will most often be due to hypertension, but even in the absence of hypertension there is still a long differential diagnosis, including hypertrophic cardiomyopathy and other infiltrative disorders. The appearance of a speckled myocardium is also not specific, particularly with current harmonic tissue imaging, which can make normal myocardium appear speckled or granular.
In addition to thickened ventricular myocardium, the valves and interatrial septum may also appear thick due to amyloid deposition (see Figure 2). In a study of the echocardiographic features of cardiac amyloidosis, the interatrial septum was thickened in 60% of patients with documented amyloid, and the combination of atrial thickening and increased myocardial echogenicity was 60% sensitive and 100% specific for amyloid compared with patients with LVH of other etiologies.4
Atrioventricular (AV) valve regurgitation may be present, whether from primary valve dysfunction or due to functional regurgitation related to ventricular dilation. Biatrial enlargement is typical (see Figure 3), and marked biatrial enlargement in the absence of ventricular dilation or significant valvular dysfunction may be a clue to underlying restrictive physiology.
Infiltrative cardiomyopathies such as amyloidosis have a significant impact on cardiac hemodynamics. Amyloid deposition impairs diastolic function and alters ventricular compliance. Early in the disease, diastolic dysfunction in the form of impaired relaxation may be present, with normal filling pressures. As the disease progresses, ventricular compliance is impaired. The poorly compliant ventricle requires increased preload; accordingly, both ventricular and atrial diastolic pressures increase.
Poor compliance leads to rapid equalization of atrial and ventricular pressures during diastole, causing abbreviation of the early diastolic filling period. In late diastole, atrial contraction normally provides additional ventricular filling. However, in restrictive processes, equalization of atrial and ventricular pressures in early diastole, combined with atrial dysfunction, results in relatively little forward flow with atrial contraction. Elevated filling pressures in the setting of impaired compliance also shorten the isovolumic relaxation period.
These manifestations of restrictive filling are readily seen with Doppler echocardiography. The early equalization of pressures and diminished effect of atrial contraction is seen in the diastolic time–velocity curve sampled at the mitral leaflet tips (see Figure 4). The E-wave velocity will be increased and the A-wave velocity decreased, resulting in an abnormal E/A ratio of ≥1.5. Rapid equalization of atrial and ventricular pressures in early diastole results in a short deceleration time (<150msec). Additionally, as left atrial pressure is increased at the time of atrioventricular pressure cross-over, isovolumic relaxation time will be shortened. The findings in our patient fit the restrictive pattern seen in amyloid heart disease. The presence of restrictive filling often occurs late in the disease5 and may portend a poor prognosis in patients with known amyloidosis.6
The assessment of ventricular filling may be straightforward; however, multiple confounders exist. These parameters are load- and heart-rate-dependent, and thus may vary over time. Although restrictive filling is seen in amyloid heart disease and restrictive cardiomyopathies in general, it may also be seen in other situations, such as systolic dysfunction with high filling pressures and constrictive pericarditis.
In particular, the distinction between restriction and constriction can be difficult. Both constriction and restriction result in elevated filling pressures with rapid equalization of pressures in early diastole and impaired filling with atrial contraction. However, salient differences in the underlying pathology make distinction between the two possible. In constrictive pericarditis, there is typically normal systolic function and normal ventricular septal mobility, but the thickened pericardium prevents normal transmission of intrathoracic pressure during respiration. In constrictive pericarditis, despite restrictive filling seen as an abnormal E/A ratio at the mitral tips in diastole, systolic myocardial velocities on tissue Doppler will typically be normal, reflecting normal myocardial function.
Although intrathoracic pressure is not transmitted to the cardiac chambers, inspiration does change pulmonary capillary wedge pressure (PCWP) in constriction. During inspiration, PCWP decreases while ventricular pressure is unchanged, causing a decrease in the pressure gradient between left atrium and left ventricle. Normal septal mobility allows the RV cavity size to increase with inspiration, when LV cavity size decreases. This combination of effects results in respiratory variation in mitral inflow with reciprocal changes in the right ventricle. Also seen is a septal ‘bounce’ with respiration, reflecting the respiratory changes in RV and LV volumes.
In patients with restriction, intrathoracic pressures are transmitted equally to the atria and ventricles, and septal mobility is typically impaired. Additionally, some degree of systolic dysfunction is often present, which is manifest by a decreased myocardial systolic velocity on tissue Doppler.7 Respiratory variation in mitral inflow is not seen, since both the PCWP and LV pressures decrease with inspiration. Impaired septal mobility does not allow the septum to bow with changes in ventricular volume, therefore the respiratory septal bounce is not seen. In patients with restrictive physiology, mitral inflow velocity variation of ≥10% change with respiration has a reported sensitivity and specificity for constrictive pericarditis of 84 and 91%, respectively.8 Unfortunately, with markedly elevated pressures the respiratory changes typical of constriction are not seen. Consequently, the Doppler filling patterns must be taken in the context of other available findings.
In our patient, a constellation of findings that include normal cavity size, increased ventricular wall thickness, biatrial enlargement, ‘sparkling’ myocardium, and restrictive filling makes the diagnosis of amyloid likely.9 Low voltage on the ECG in the setting of increased myocardial mass also increases the likelihood of amyloid.10
The patient presented highlights two issues. First is the role of echocardiography in the diagnosis of amyloid heart disease. Despite this patient’s young age, amyloidosis with cardiac involvement may be a consequence of his childhood cancer treatments, which place survivors at risk for secondary malignancies later in life. Second is the issue of cardiac follow-up for childhood cancer survivors. There are long-term consequences of anthracycline treatment in childhood, particularly if there was concomitant chest radiation. Guidelines for follow-up include recommendations for assessment of ventricular function. Although amyloid is a very uncommon consequence, late systolic dysfunction can be seen and can be amenable to standard heart failure therapy.
- Kyle RA, Gertz MA, Greipp PR, et al., A trial of three regimens for primary amyloidosis: colchicine alone, melphalan and prednisone, and melphalan, prednisone, and colchicine, N Engl J Med, 1997;336(17):1202–7.
- Lachmann HJ, Goodman HJ, Gilbertson JA, et al., Natural history and outcome in systemic AA amyloidosis, N Engl J Med, 2007;356(23):2361–71.
- Rahman JE, Helou EF, Gelzer-Bell R, et al., Noninvasive diagnosis of biopsy-proven cardiac amyloidosis, J Am Coll Cardiol, 2004;43(3): 410–15.
- Falk RH, Plehn JF, Deering T, et al., Sensitivity and specificity of the echocardiographic features of cardiac amyloidosis, Am J Cardiol, 1987;59(5):418–22.
- Klein AL, Tajik AJ, Doppler assessment of diastolic function in cardiac amyloidosis, Echocardiography, 1991;8(2):233–51.
- Klein AL, Hatle LK, Taliercio CP, et al., Prognostic significance of Doppler measures of diastolic function in cardiac amyloidosis. A Doppler echocardiography study, Circulation, 1991;83(3):808–16.
- Ha JW, Ommen SR, Tajik AJ, et al., Differentiation of constrictive pericarditis from restrictive cardiomyopathy using mitral annular velocity by tissue Doppler echocardiography, Am J Cardiol, 2004;94(3):316–19.
- Rajagopalan N, Garcia MJ, Rodriguez L, et al., Comparison of new Doppler echocardiographic methods to differentiate constrictive pericardial heart disease and restrictive cardiomyopathy, Am J Cardiol, 2001;87(1):86–94.
- Bhandari AK, Nanda NC, Myocardial texture characterization by two-dimensional echocardiography, Am J Cardiol, 1983;51(5): 817–25.
- Carroll JD, Gaasch WH, McAdam KP, Amyloid cardiomyopathy: characterization by a distinctive voltage/mass relation, Am J Cardiol, 1982;49(1):9–13.