Outline
Epidemiology, 301
Natural History, 302
Clinical Features, 303
Misdiagnosis and Obstacles to Early Diagnosis, 303
Raising Clinical Suspicion for Cardiac Amyloidosis, 303
Diagnosis, 303
Electrocardiography, 303
Laboratory Testing, 303
Cardiac Imaging, 303
Tissue Biopsy and Confirmation of the Causative Precursor Protein, 305
Genetic Analysis, 306
Clinical Management, 306
Biomarker Staging Systems, 306
Prognostication From Advanced Cardiac Imaging, 306
Supportive Nondisease-Modifying Therapies, 307
Diuretics, 307
Drug Intolerances in Advanced CA, 307
Calcium Channel Blockers and Digoxin, 307
Atrial Fibrillation, Thromboembolic Complications, and Anticoagulation, 307
Permanent Pacemaker Implantation, 307
Implantable Cardioverter Defibrillators, 307
Organ Transplantation, 308
Disease-Targeted Therapies for AL-CA, 308
Chemotherapy and Stem Cell Transplantation, 308
Disease-Modifying Therapeutic Opportunities for ATTR-CA, 308
Silencers of TTR Gene Expression, 308
TTR Stabilization, 309
Fibril Degradation and Reabsorption, 310
Future Perspectives, 310
Amyloidosis is a deposition disease in which proteins with unstable structures misfold and aggregate into amyloid fibrils, which deposit in the heart, kidneys, liver, peripheral nerves, gastrointestinal tract, lungs, and soft tissues. Amyloid fibrils are insoluble nonbranching structures 7 nm to 10 nm wide and of variable length that are resistant to proteolysis. Under normal light microscopy when stained with Congo red, amyloid deposits display a hyaline pink appearance and under polarized light, change to a characteristic apple-green birefringence. When viewed by electron microscopy, amyloid fibrils appear as needle-like filamentous bundles with a unique cross β-pleated sheet configuration.
Cardiac amyloidosis (CA) is an increasingly recognized cause of heart failure (HF). Among the thirty proteins known to form amyloid fibrils that deposit in the extracellular space leading to disruption of tissue architecture and organ dysfunction, cardiologists predominantly encounter amyloidosis caused by three proteins that infiltrate the heart: (1) immunoglobulin light chain (AL) and transthyretin (ATTR) amyloidosis due to (2) a mutation (hATTR, also known as hereditary or familial amyloid cardiomyopathy) or (3) wild-type (wtATTR, also known as senile or age-related CA). In developing nations, secondary amyloid A (AA) is more prevalent due to chronic infections and inflammatory conditions. Uncommon variants also known to affect the heart or aorta include atrial natriuretic peptide (ANP), apolipoprotein A1 (AApoA1), fibrinogen (Afib), gelsolin (Agel), and lactadherin amyloid ( Table 22.1 ).
| Amyloid Type | Amyloidogenic Protein | Extent of Cardiac Involvement |
|---|---|---|
| AL | Immunoglobulin light chain of κ or λ subtype | Frequent and severe cardiac involvement |
| ATTR | Transthyretin | |
| hATTR ( familial ) | Mutant transthyretin | Severe with particular mutations (Val122Ile, Thr60Ala, Ile68Leu, Leu11Met) |
| wtATTR ( senile ) | Wild-type transthyretin | Frequent and severe cardiac involvement is typical |
| AA | Serum amyloid A protein | Usually involves kidney first causing nephrotic syndrome; severe cardiac involvement can occur |
| Afib | Mutant fibrinogen | Cardiac autonomic dysfunction can occur |
| AGel | Gelsolin | Cardiac involvement possible |
| Apo A1 | Mutant apolipoprotein A1 | Severe cardiac involvement can occur |
| ANP | Atrial natriuretic peptide | Atrial involvement may occur |
Epidemiology
Light-chain cardiac amyloidosis (AL-CA) is a rare condition with an estimated incidence of 10 cases per million individuals and thus ∼3000 new diagnoses per year in the United States. Among new cases, 30% to 50% present with symptomatic cardiac involvement. AL amyloid is caused by plasma cell dyscrasia, which spans the spectrum from monoclonal gammopathy of undetermined significance (MGUS) to AL and multiple myeloma. Approximately 10% to 15% of patients with AL amyloidosis meet criteria for multiple myeloma.
Emerging data suggest that transthyretin cardiac amyloidosis (ATTR-CA) is not uncommon, especially due to wtATTR, which will become the most commonly diagnosed form of CA. Evidence suggests it is underdiagnosed as a cause of common cardiovascular conditions in older adults, including heart failure with a preserved ejection fraction (HFpEF) and aortic stenosis (AS), and is commonly seen in subjects with degenerative orthopedic conditions such as lumbar spinal stenosis, biceps tendon rupture, and bilateral carpal tunnel syndrome. Among patients with HFpEF over 75 years of age at autopsy, 32% have amyloid deposits, compared with 8% in those less than 75 years of age. In hospitalized patients with HFpEF who underwent nuclear scintigraphy, 13% had wtATTR-CA. Similarly, nuclear scintigraphy in patients undergoing transcatheter aortic valve replacement (TAVR) for severe calcific AS revealed a prevalence of 16% overall and 22% among men. In patients undergoing surgery for degenerative lumbar spinal stenosis, almost all had amyloid in surgical specimens of the ligamentum flavum, and transthyretin (TTR) was identified immunohistochemically as the causative protein in one-third of resected tissues. Among patients with wtATTR-CA, biceps tendon rupture has been observed in 33% of patients, occurring in the dominant arm in 95% and bilaterally in 24% of patients.
Mutations in the TTR gene, which are inherited in an autosomal dominant fashion, can also lead to ATTR-CA (see also Chapter 24 ). Over 120 mutations have been described, of which some have varying phenotypes and are endemic to specific geographic regions. For example, the Val122Ile or Leu111Met mutations lead to a predominant cardiomyopathy, while Val30Met causes a predominant neuropathy, also known as familial amyloidotic polyneuropathy (FAP), and Glu89Gln and Thr60Ala lead to a mixed cardiomyopathy/neuropathy ( Table 22.2 ). Among mutations in the USA, Val122Ile is the most common. Val122Ile affects almost exclusively individuals of African or Afro-Caribbean descent with a population prevalence of 3.4% and is under-recognized as a cause of HF. Another common mutation, Thr60Ala, is the most common variant in the UK with an estimated population prevalence of 1% in northwest Ireland. The prevalence and penetrance of Val30Met, which is endemic to Portugal, Sweden, Japan, and Brazil, varies widely with region and age of onset. The disease prevalence is 1 in 100,000 individuals in nonendemic areas such as the United States, compared with 1 in 538 individuals affected in endemic areas of Portugal. Penetrance can also vary geographically with 22% at age 60 years in Sweden, compared with 80% at 50 years in Portugal. Patients with early-onset disease manifest with peripheral neuropathy, while those with late-onset disease typically present with a concomitant cardiomyopathy.
| Clinical Characteristic | AL | wtATTR | ATTR Val122Ie | ATTR Thr60Ala |
|---|---|---|---|---|
| Median age, years | 55 | 75 | 70 | 60 |
| Sex | Equal | >90% male reported to date | Male predominant | Equal |
| Ethnic background | None | White | African-American or Afro-Caribbean | Irish |
| Peripheral neuropathy | +++ | + | + | +++ |
| Frequency of neuropathy | ++ | + | + | ++ |
Natural History
In AL-CA, the presence and extent of cardiac involvement is the major determinant of survival. Without successful treatment, patients who present with AL-CA and symptomatic HF have a median survival of 6 months. The efficacy of chemotherapy targeting the underlying plasma cell clone is the next most important determinant of survival. Overall survival has improved with advances in therapy for plasma cell dyscrasias but still remains low at ∼42% at 4 years. Despite a hematologic response, survival is dependent on underlying cardiac function, and patients may die of progressive HF or sudden cardiac death (SCD). Pulseless electrical activity and electromechanical dissociation are often the cause of SCD, although ventricular tachyarrhythmias, bradycardia, and thromboembolism are not uncommon sequelae.
Original reports suggested that the median survival of patients with wtATTR-CA was greater than 5 years, but contemporary studies demonstrate worse outcomes with a median survival of 3.5 years from initial evaluation. This discrepancy stems largely from the difficulty in determining the precise time of disease onset. Initial clinical manifestations of TTR amyloid deposition may include bilateral carpal tunnel syndrome, which is present in ∼70% of individuals an average 5 to 7 years before the start of cardiac manifestations. Other early signs of amyloid deposition include atrial arrhythmias and progressive effort intolerance, which are difficult to definitively attribute to amyloidosis given these occur commonly in older adults. In an early longitudinal multicenter investigation, the Transthyretin Amyloid Cardiac Study (TRACS), which studied advanced patients, mortality with ATTR-CA was high and was worse in patients with hATTR Val22Ile compared to wtATTR. Common causes of death included HF, SCD, and sepsis. These data have been corroborated by the Transthyretin Amyloidosis Outcome Survey (THAOS), which reported significantly worse age-adjusted survival from time of enrollment over 3 years in patients with hATTR-CA Val122Ile compared to wtATTR-CA (HR 1.947, P = .013). A single-center study demonstrated significantly worse overall and age-adjusted survival from time of diagnosis in patients with hATTR-CA Val122Ile compared with wtATTR-CA (median survival 47 months vs. 59 months, P = .01), with survival diverging well after 24 months from diagnosis.
Physiologic derangements in ATTR-CA include a reduction in chamber capacitance, decline in contractility, and increase in arterial elastance, which lead to progressive left ventricular (LV) dysfunction. Blood pressure falls due to a reduction in cardiac output and heart rate increases to compensate for a reduced forward stroke volume. Among patients with wtATTR-CA, a positive troponin T, presence of a pacemaker, and New York Heart Association (NYHA) class IV symptoms were associated with a worse outcome. Clinical assessments at 6-month increments in the TRACS study in patients with advanced disease demonstrated marked disease progression: mean 6-minute hall walk declined 26 m, N-terminal pro-B-type natriuretic peptide (NT-proBNP) increased 1816 pg/mL, and left ventricular ejection fraction (LVEF) fell 3%. In advanced ATTR-CA, cardiac cachexia ensues, which may be mediated by right HF, liver congestion, and altered bowel flora.
Clinical Features
Misdiagnosis and Obstacles to Early Diagnosis
CA is often misdiagnosed for other causes of left ventricular hypertrophy (LVH) such as hypertrophic cardiomyopathy (HCM), hypertensive cardiomyopathy, and HFpEF. Neuropathy in patients with FAP is often mistaken for chronic inflammatory demyelinating polyneuropathy (CIDP). Consequent delays in diagnosis have detrimental consequences for patients. A survey in 533 participants with AL amyloidosis, 37% of whom had cardiac involvement, revealed that the average time from initial symptoms to diagnosis was 2 years. Nearly one-third of patients reported visiting more than five physicians before receiving a diagnosis of amyloidosis, while only 8% received the diagnosis after visiting one physician. Despite seeing cardiologists more frequently than hematologists and nephrologists, cardiologists only made the diagnosis in 19% of cases. Similarly, approximately 50% of patients with ATTR-CA receive a diagnosis within 6 months, typically by a cardiologist.
CA is a “great mimicker,” and the diagnosis is delayed due to both physician and disease-related factors. A multidisciplinary amyloid team of cardiologists, neurologists, nephrologists, and hematologists is optimal for diagnosis and management, but such teams are rare and isolated to a few academic medical centers. The late presentation of AL-CA and the absence of disease-modifying treatments for ATTR-CA have also led physicians to express nihilism about prognosis and management options. An important disease-specific cause of diagnostic uncertainty is the heterogeneous phenotype of CA, which ranges from an isolated cardiomyopathy to systemic involvement with minimal cardiac involvement. Phenotypic variability within inherited ATTR is also influenced by genetic heterogeneity, geography, endemic region of origin, age, sex of the proband and transmitting parent, as well as amyloid fibril composition. The traditional requirement of histopathologic evidence of amyloid infiltration in involved tissues also delays diagnosis, as expertise in endomyocardial biopsy and amyloid-specific histopathologic techniques is typically restricted to specialty centers.
Raising Clinical Suspicion for Cardiac Amyloidosis
A heightened clinical suspicion for CA is the key step in avoiding misdiagnosis and delayed diagnosis. The presence of noncardiac signs such macroglossia and periorbital purpura, while specific, are present only in a minority of cases of AL-CA and absent in ATTR-CA. A patient with unexplained LVH or restrictive cardiomyopathy should immediately prompt consideration of CA. Additional red flags may include a history of carpal tunnel syndrome in ATTR (especially bilateral with a history of repeated carpal tunnel surgeries), atraumatic rupture of the biceps tendon, a history of total knee or hip arthroplasty, surgeries for degenerative lumbar spinal stenosis, unexplained neuropathic pain, and orthostatic hypotension. In addition, a diagnosis of “hypertensive cardiomyopathy” in a patient with normal or low blood pressure or “HCM” in a patient aged 60 years or older should raise clinical suspicion for CA.
Diagnosis
When CA is suspected, the diagnosis must be systematic and directed toward defining the specific amyloid subtype and assessing the burden of amyloid infiltration in the myocardium and other organs.
Electrocardiography
Low QRS voltage on electrocardiography (ECG) is touted as a classic and pathognomonic sign specific to CA but occurs as a late phase phenomenon associated with poor prognosis. In contemporary series, the prevalence of low voltage is relatively low and varies with CA subtype, ranging from 20% in ATTR to 60% in AL. The absence of low QRS voltage therefore should not preclude the diagnosis of CA, particularly in patients with ATTR-CA in whom 30% present with LVH or left bundle branch block (LBBB). The hallmark ECG feature of CA is a disproportionately low QRS voltage to LV mass ratio. A pseudo-infarct pattern, defined as pathologic Q waves in at least two contiguous leads without obstructive coronary artery disease, is present in up to 70% of cases and is more sensitive for CA than low voltage. Patients with CA may also develop progressive conduction disease such that AV block in an older patient with increased LV wall thickness should raise suspicion for CA.
Laboratory Testing
To date, no blood test exists to detect TTR oligomers to diagnose ATTR-CA. In AL-CA, however, quantification of serum free light chains (FLCs) and identification of an abnormal monoclonal band on immunofixation of serum and/or urine have a combined sensitivity of 99% for identifying AL-CA. However, up to 5% of the population over 65 years of age has MGUS, and therefore an abnormal ratio of kappa (κ) to lambda (λ) FLCs alone is not specific for AL amyloidosis. An abnormal κ/λ ratio in elderly patients who in fact have ATTR-CA with concurrent MGUS can lead to misdiagnosis of AL-CA in up to 10% of cases (even at referral centers). Interpretation of FLC concentrations must also take into account renal function, as FLCs are filtered by the glomeruli. Renal dysfunction can lead to increased serum FLC concentrations and affects κ and λ light chains differently. Therefore making a diagnosis of AL-CA is more challenging in patients with chronic kidney disease, and a wider reference range for a normal κ/λ ratio has been proposed in this population.
Natriuretic peptides tend to be markedly elevated in CA, typically out of proportion to LV systolic function. In AL-CA, circulating light chains are directly toxic to cardiomyocytes by modulating p38 mitogen-activated protein kinase and subsequent downstream upregulation of NT-proBNP expression. Direct injection of light chains from subjects with AL amyloidosis into zebrafish results in impaired cardiac function, pericardial edema, and increased cell death with eventual 100% mortality relative to light chain control samples isolated from patients with multiple myeloma. Therefore for the same degree of hemodynamic derangement, plasma NT-proBNP levels are higher in patients with AL compared with ATTR-CA. Amyloid infiltration can also trigger cardiomyocyte apoptotic pathways with subsequent elevation in serum troponin levels, which can lead to false diagnoses of an acute coronary syndrome. Consequently, in patients with unexplained new-onset HF, measurement of serum FLCs, NT-proBNP, and troponin may raise the suspicion of AL-CA. These tests are especially important in patients with a preexisting clonal plasma cell dyscrasia such as MGUS or smoldering myeloma, or in patients with unexplained increased LV wall thickness.
Cardiac Imaging (see also Chapter 32 )
In clinical practice, transthoracic echocardiography, speckle tracking strain echocardiography, and cardiac magnetic resonance (CMR) often raise suspicion for CA ( Fig. 22.1 ), while bone radiotracer scintigraphy can be diagnostic of ATTR-CA. As these modalities are complementary to one another, the focus should be to define the specific diagnostic goal and selecting the corresponding imaging test.
Transthoracic Echocardiography
Echocardiographic features of CA are nonspecific, but in clinical context are highly suggestive of CA. In advanced disease, predominant abnormalities include symmetric thickening of the LV (though asymmetric hypertrophy is common), thickening of the right ventricular (RV) free wall, and a small pericardial effusion. Classically, thickening of the atrioventricular valves and interatrial septum and a “speckled” myocardium also occur, although “speckling” is less reliable with harmonic imaging. Amyloidosis is the archetype of LV diastolic dysfunction, with stages of diastolic impairment paralleling progressive disease. Impaired LV relaxation occurs in early CA, which progresses to restrictive pathophysiology. Analogous progression in RV diastolic dysfunction is reflected in Doppler signals of the RV inflow, superior vena cava, and hepatic vein flow velocities. CA also leads to impaired systolic performance, which precedes the onset of HF. In parametric polar maps of the LV, relative apical sparing of global longitudinal strain (LS) represents an important diagnostic clue for CA relative to other causes of LVH. While differentiating AL from ATTR-CA by echocardiography is not possible, on average, patients with AL-CA tend to have more restrictive diastolic dysfunction and patients with ATTR-CA tend to have greater LV wall thickness.
Cardiac MRI
Structural information derived from CMR is similar to that acquired from echocardiography, but the opportunity to interrogate tissue composition with gadolinium-based contrast agents in CMR has led to an increase in the diagnosis of CA. Gadolinium is a purely extracellular agent, which does not enter intact cardiomyocytes. Global subendocardial late gadolinium enhancement (LGE) in a noncoronary artery distribution is pathognomonic for CA, but LGE can also be diffuse and transmural or focal and patchy. T1 mapping, in which a quantitative signal from the myocardium is measured before or after contrast administration, has shown increased precontrast T1 signal in patients with CA compared with HCM or healthy controls. Pre- and postcontrast T1 data can also be used to calculate extracellular volume (ECV), a measurement of interstitial expansion, which is significantly elevated in patients with CA due to interstitial amyloid deposition. Native T1 mapping does not require administration of contrast and is therefore safe in patients with renal impairment. Contemporary CMR enables assessment of amyloid burden via direct measurement of ECV, edema via measurement of native T1, and myocyte response via measurement of intracellular volume. Limitations to T1 mapping include nonstandardized reference ranges with different software platforms and limited availability. While CMR is safe, painless, and requires no radiation, its use is limited in patients with CA who have implanted ferromagnetic hardware or cannot tolerate lying immobile for the duration of the exam due to comorbidities such as spinal stenosis or HF. Protocols requiring the use of contrast may also be contraindicated in patients with renal impairment. Therefore in clinical practice, CMR may raise clinical suspicion of CA but is not diagnostic. Advanced CMR techniques provide insights into the pathophysiological processes underlying CA and monitoring of disease progression and response to therapy.
Myocardial Radiotracer Scintigraphy
Nuclear cardiac scintigraphy using technetium-labeled bone radiotracer has demonstrated high diagnostic accuracy in multiple studies. Radiotracers include technetium-3,3-diphosphono-1,2-propanodicarboxylic ( 99m Tc-DPD), technetium-pyrophosphate ( 99m Tc-PYP), and technetium-hydroxymethylene diphosphonate ( 99m Tc-HMDP). A revival in nuclear cardiac scintigraphy for CA occurred when 99m Tc-DPD planar scintigraphy demonstrated high sensitivity and specificity for differentiating ATTR-CA from other causes of LVH using a semiquantitative visual score (range 0–3 with score 0 = absent cardiac uptake, normal bone uptake; 1 = mild cardiac uptake inferior to bone uptake; 2 = moderate cardiac uptake with attenuated bone uptake; 3 = strong cardiac uptake with mild/absent bone uptake) and a quantitative heart-to-whole body (H/WB) ratio. Like 99m Tc-DPD, 99m Tc-PYP in North America, using a quantitative heart-to-contralateral (H/CL) ratio, and 99m Tc-HMDP in Europe using a quantitative heart-to-skull (H/S) ratio, exhibited high diagnostic accuracy for differentiating ATTR-CA from AL-CA and nonamyloid HFpEF.
An international multicenter collaboration in 857 patients with CA who underwent either 99m Tc-DPD, 99m Tc-PYP, or 99m Tc-HMDP planar radiotracer scintigraphy showed that grade 2 or 3 myocardial uptake was 100% specific for the diagnosis of ATTR-CA in patients with no evidence of an abnormal monoclonal protein . The presence of a monoclonal protein was defined as an abnormal FLC ratio (<0.26 or >1.65) on serum Freelite assay or presence of a monoclonal protein on immunofluorescence of serum or urine. These data led to an international consensus that in select patients without evidence of an abnormal monoclonal protein, a positive radiotracer nuclear scan can diagnose ATTR-CA without the need for a biopsy. Planar nuclear scintigraphy is therefore an accurate and feasible imaging modality that may spare select elderly patients the need for invasive biopsy. These scans are faster than CMR, though they do require a supine position and minimal radiation exposure. A standardized protocol using 99m Tc-PYP demonstrated the feasibility of a faster incubation period, less radiation exposure compared to other protocols, and shorter imaging time. The molecular mechanism by which 99m Tc-based bone radiotracers selectively bind to ATTR amyloid fibrils in the myocardium is unknown, but it has been suggested that the preferential binding may be a result of higher calcium content relative to other fibril types. Bone tracers can also detect early ATTR cardiac infiltration in asymptomatic allele carriers before echocardiographic or biomarker changes occur.
Cardiac Positron Emission Tomography
Positron emission tomography (PET) radiotracers have also demonstrated high affinity for amyloid fibrils. Several radiotracers studied in small patient cohorts include 11 C-Pittsburgh compound B (PIB), and the 18 fluorine-labeled ( 18 F) compounds, 18 F-florbetapir, 18 F-florbetaben, and 18 F-sodium fluoride (NaF). Intense myocardial uptake of 11 C-PIB, 18 F-florbetapir, and 18 F-florbetaben in patients with CA (irrespective of AL and ATTR subtype) occurs with no uptake in amyloid negative controls. A limitation of 11 C-PIB is its short radioactive half-life of 20 minutes, which restricts its use to centers with a nearby cyclotron and radiopharmacy infrastructure, while 18 F-florbetapir and 18 F-florbetaben have longer half-lives of 110 minutes allowing for widespread distribution to PET centers without a cyclotron. While these tracers may have potential for evaluating for CA, they do not differentiate between CA subtypes. Increased myocardial uptake of 18 F-NaF, which also has a half-life of 110 minutes, occurred in small cohorts of patients with ATTR-CA but not in AL or nonamyloid controls, suggesting specificity for ATTR.
Tissue Biopsy and Confirmation of the Causative Precursor Protein
The diagnostic accuracy of a noncardiac biopsy is dependent on amyloid type and the source of examined tissue ( Table 22.3 ). The yield of a noncardiac biopsy from the abdominal fat pad is higher in patients with AL compared with ATTR. In AL-CA, the yield of an abdominal fat pad biopsy can be as high as 84% while it is lower in hATTR-CA at 45% and in wtATTR-CA at 15%. Therefore while the abdominal fat pad is a preferred initial site for harvesting tissue in subjects with AL-CA, a negative result is not sufficient to exclude the diagnosis, and in the context of high clinical suspicion, an endomyocardial biopsy should be pursued. Biopsy specimens also delineate the amyloidogenic precursor protein. Immunohistochemistry is the most widespread method of tissue typing and is commonly performed on formalin-fixed, paraffin-embedded sections for κ, λ, TTR, and serum amyloid A protein. A common pitfall is false-positive immunostaining for TTR in CA, which is confounded by the frequent lack of staining for immunoglobulin light chains. Mass spectrometry–based proteomic analysis is considered the gold standard for fibril typing and can frequently identify a mutant TTR genotype, although genetic sequencing is the gold standard for identifying TTR mutations.
| Misconception | Clinical Reality |
|---|---|
| Cardiac amyloid is rare |
|
| A negative fat pad biopsy rules out the diagnosis of amyloid |
|
| Low-voltage on an electrocardiography is an early diagnosis marker of CA |
|
| Lack of disease-modifying treatment renders accurate diagnosis merely an academic exercise |
|
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
