Cardiac amyloidosis is a form of restrictive infiltrative cardiomyopathy that confers significant mortality. Due to the relative rarity of cardiac amyloidosis, clinical and diagnostic expertise in the recognition and evaluation of individuals with suspected amyloidosis is mostly limited to a few expert centers. Electrocardiography, echocardiography, and radionuclide imaging have been used for the evaluation of cardiac amyloidosis for over 40 years.^1-3 Although cardiovascular magnetic resonance (CMR) has also been in clinical practice for several decades, it was not applied to cardiac amyloidosis until the late 1990s. Despite an abundance of diagnostic imaging options, cardiac amyloidosis remains largely underrecognized or delayed in diagnosis.⁴ While advanced imaging options for noninvasive evaluation have substantially expanded, the evidence is predominately confined to single-center small studies or limited multicenter larger experiences, and there continues to be no clear consensus on standardized imaging pathways in cardiac amyloidosis. This lack of guidance is particularly problematic given that there are numerous emerging therapeutic options for this morbid disease, increasing the importance of accurate recognition at earlier stages. Imaging provides non-invasive tools for follow-up of disease remission/progression complementing clinical evaluation. Additional areas not defined include appropriate clinical indications for imaging, optimal imaging utilization by clinical presentation, accepted imaging methods, accurate image interpretation, and comprehensive and clear reporting. Prospective randomized clinical trial data for the diagnosis of amyloidosis and for imaging-based strategies for treatment are not available. A consensus of expert opinion is greatly needed to guide the appropriate clinical utilization of imaging in cardiac amyloidosis.

The American Society of Nuclear Cardiology (ASNC) has assembled a writing group with expertise in cardiovascular imaging and amyloidosis, with representatives from the American College of Cardiology (ACC), the American Heart Association (AHA), the American Society of Echocardiography (ASE), the European Association of Nuclear Medicine (EANM), the Heart Failure Society of America (HFSA), the International Society of Amyloidosis (ISA), the Society for Cardiovascular Magnetic Resonance (SCMR), and the Society of Nuclear Medicine and Molecular Imaging (SNMMI). This writing group has developed a joint expert consensus document on imaging cardiac amyloidosis, divided into two parts. Part 1 has the following aims:

Perform and document a comprehensive review of existing evidence on the utility of echocardiography, CMR, and radionuclide imaging in screening, diagnosis, and management of cardiac amyloidosis.

Define standardized technical protocols for the acquisition, interpretation, and reporting of these noninvasive imaging techniques in the evaluation of cardiac amyloidosis.

Part 2 of this expert consensus statement addresses the development of consensus diagnostic criteria for cardiac amyloidosis, identifies consensus clinical indications, and provides ratings on appropriate utilization in these clinical scenarios.

The overall goal of this multi-societal expert consensus document on noninvasive cardiovascular imaging in cardiac amyloidosis is to standardize the selection and performance of echocardiography, CMR, and radionuclide imaging in the evaluation of this highly morbid condition, and thereby improve healthcare quality and outcomes of individuals with known or suspected cardiac amyloidosis. We hope that research generated to validate the recommendations of this consensus document will form the basis for evidence-based guidelines on cardiac amyloidosis imaging within the next few years.

Cardiac amyloidosis is a cardiomyopathy that results in restrictive physiology from the myocardial accumulation of misfolded protein deposits, termed amyloid fibrils, causing a clinically diverse spectrum of systemic diseases. Most cases of cardiac amyloidosis result from two protein precursors: amyloid immunoglobulin light chain (AL), in which the misfolded protein is a monoclonal immunoglobulin light chain typically produced by bone marrow plasma cells, and amyloid transthyretin (ATTR) amyloidosis, in which the misfolded protein is transthyretin (TTR), a serum transport protein for thyroid hormone and retinol that is synthesized primarily by the liver.³ ATTR amyloidosis is further subtyped by the sequence of the TTR protein into wild-type (ATTRwt) or hereditary (ATTRv), the latter resulting from genetic variants in the TTR gene.^5,6 Cardiac involvement in systemic AL amyloidosis is common (up to 75%, depending on diagnostic criteria),⁷ and in the case of ATTRwt amyloidosis, is the dominant clinical feature seen in all cases.

The different types of cardiac amyloidosis display significant heterogeneity in clinical course, prognosis, and treatment approach.⁸ AL amyloidosis is characterized by a rapidly progressive clinical course, and if untreated, the median survival is less than 6 months. ATTRv amyloidosis follows a varied clinical course depending upon the specific mutation inherited with either cardiomyopathy and/or sensory/autonomic polyneuropathy.⁹ Furthermore, ATTR amyloidosis (both wild-type and hereditary) is characterized by an age-dependent penetrance, with the clinical phenotype developing as age advances.

The diagnosis of cardiac amyloidosis remains challenging owing to a number of factors, which include the relative rarity of the disease, clinical overlap with more common diseases that result in thickening of the myocardium (ie, hypertension, chronic renal failure, hypertrophic cardiomyopathy, aortic stenosis), unfamiliarity with the proper diagnostic algorithm, and a perceived lack of definitive treatment. While systemic AL amyloidosis is indeed a rare disease affecting approximately 8 to 12^10,11 per million person years, and as high as 40.5 per million person years in 2015,¹² ATTRwt cardiac amyloidosis appears quite common, with recent reports using contemporary diagnostic strategies that place the prevalence in as many as 10% to 16% of older patients with heart failure or with aortic stenosis.^13-15 In addition, the most common mutation associated with ATTRv amyloidosis has been reproducibly demonstrated in 3.4% of African Americans.¹⁶ While the penetrance remains disputed, this suggests there are approximately 2 million people in the United States who are carriers of an amyloidogenic mutation and are at risk for cardiac amyloidosis. It is clear both ATTRv and ATTRwt cardiac amyloidosis are underrecognized, yet important causes of diastolic heart failure.¹⁷

Treatment options are rapidly expanding. Anti-plasma cell therapeutics have extended median survival in AL amyloidosis beyond 5 years, 7 with increasing survival beyond 10 years. We are potentially nearing a similar sea change in the management of ATTR amyloidosis. ATTR amyloidosis was previously only treated by solid-organ transplantation, as conventional highly effective heart-failure therapy is poorly tolerated and contraindicated in advanced cardiac amyloidosis. Although early clinical trials of amyloid specific antibodies have been unsuccessful to date,^18-20 one remains under study in a Phase I clinical trial.²¹ Novel therapeutics that suppress TTR expression have been studied in Phase 3 clinical trials and received FDA approval^18,19 for ATTRv with polyneuropathy. Additionally, a randomized clinical trial of TTR stabilizer therapy demonstrated a reduction in all-cause mortality in ATTR cardiomyopathy²²; this agent has recently received FDA approval for ATTR cardiomyopathy. As these exciting prospects move into the clinical realm, it is evident early diagnosis will be essential to afford the most effective treatment options for both AL and ATTR cardiac amyloidosis.

Despite these advances in treatment, the challenge persists to increase recognition and achieve effective, timely diagnosis. In the past, a diagnosis of cardiac amyloidosis required an endomyocardial biopsy, which remains the gold standard, as it is virtually 100% accurate, assuming appropriate sampling, for the detection of amyloid deposits.²³ Specific identification of the precursor protein can be accomplished from the tissue specimen through immunohistochemistry, albeit with limitations,²⁴ or laser-capture tandem mass spectrometry (LC/MS/MS). This latter technique is considered the definitive test for precursor protein identification.²⁵ While ATTR cardiac amyloidosis can now be diagnosed accurately without the need of cardiac biopsy,³ AL amyloidosis requires demonstration of light-chain amyloid fibrils in tissue (although not necessarily the heart) prior to administration of chemotherapy. Even for ATTR cardiac amyloidosis, a cardiac biopsy remains necessary in the context of equivocal imaging or the co-existence of a monoclonal gammopathy.

Clinical suspicion of cardiac amyloidosis can be raised by the constellation of clinical signs and symptoms, specific demographics (ie, age, race, country of family origin), electrocardiography, and suggestive non-invasive imaging findings. Endomyocardial biopsy, although highly sensitive (100%),²³ is impractical as a screening test for cardiac amyloidosis, given its inherent risk and requirement of pathologic expertise, which is limited to a few academic centers. Other limitations of endomyocardial biopsy include: inability to quantify whole-heart amyloid burden, inability to evaluate systemic disease burden, and, for these same reasons, limited assessment of response to therapy. Thus, contemporary imaging techniques, including CMR, radionuclide imaging with bone-avid radiotracers, and echocardiography with longitudinal strain quantification, have evolved as the principal means for diagnosis and management of cardiac amyloidosis.

The current diagnostic approach for cardiac amyloidosis involves the use of one or more of these imaging modalities in conjunction with assessment of a plasma-cell disorder (Figure 1).³ Serum plasma electrophoresis is an insensitive test for AL amyloidosis and thus is unreliable for diagnosing AL amyloidosis. Serum and urine immunofixation and the measurement of serum free light chains (FLC) are necessary for the diagnosis of AL amyloidosis. In cases of confirmed ATTR amyloidosis, TTR gene sequencing is performed to establish ATTRwt vs ATTRv. In AL amyloidosis, the concentration of the affected FLC, in conjunction with serum N-terminal-pro brain natriuretic peptide (NT-proBNP) and cardiac troponin T or I, can be utilized to assign a disease stage that confers highly reproducible prognostic information.²⁶ Furthermore, a cardiac staging system based on NT-proBNP and cardiac troponins (along with differential FLC levels) allows the stratification of patients into stages widely used in clinical practice for modulating the therapy intensity in AL amyloidosis.²⁶ A European study identified a stage 3b subgroup with very advanced cardiac involvement; these patients had high concentrations of NT-proBNP (>8500 ng/L) and a very poor prognosis, which warrants further study.²⁷ Furthermore, a reduction in FLC following anti-plasma cell treatment, termed a hematologic response, is typically followed within 6 to 12 months by a reduction in NT-pro-BNP and troponin, termed an organ-specific response, which is associated with improved symptoms of heart failure and extended survival.²⁸ The FLC-based and NT-proBNP-based hematology and cardiac responses have been extensively validated in AL amyloidosis.²⁹ In ATTR cardiac amyloidosis, NT-pro-BNP, cardiac troponin, and estimated glomerular filtration rate have also been validated as diagnostic markers in different risk-prediction models,^30-32 with changes in NT pro-BNP useful to follow disease progression.^18,22 Biomarker evaluation is an integral part of the management of patients with AL and ATTR cardiac amyloidosis.

Figure 1. Systematic evaluation of cardiac amyloidosis. A comprehensive evaluation of cardiac amyloidosis includes consideration of clinical symptoms, evaluation of cardiac involvement (biomarkers and cardiac imaging), evaluation of systemic amyloidosis (serum, urine testing, and biopsy), followed by typing of amyloid deposits into AL or ATTR, and documentation of mutations in patients with ATTR amyloidosis. *Clinical symptoms: heart failure, peripheral/autonomic neuropathy, macroglossia, carpal tunnel syndrome, periorbital bruising, stroke, atrial fibrillation, postural hypotension, fatigue, weight loss, pedal edema, renal dysfunction, diarrhea, constipation. †Evaluation for cardiac amyloidosis: ECG, ECHO, CMR, EMB, ^99mTc-PYP/DPD/HMDP/¹²³I-mIBG/PET, NT-proBNP, troponin T. ‡Evaluation for systemic amyloidosis: AL: detect plasma cell clone: serum and urine immunofixation, serum FLC assay and immunoglobulin analysis; AL: detect systemic organ involvement: 24-hour urine protein, Alkaline phosphatase, eGFR, cardiac biomarkers (NT-proBNP, troponins); Tissue biopsy: EMB/Fatpad/ Bone marrow/Other with Congo red staining. §Confirm Amyloidosis Type: ATTR: IHC and MS of Biopsy or ^99mTc- PYP/DPD/HMDP Grade 2 or 3 if a clonal process is excluded; AL: MS or IHC of Biopsy. ¥Confirm TTR Mutation in Patients with ATTR amyloidosis: genetic testing for TTR mutations. AL, amyloid light chain; ATTR, amyloid transthyretin; CMR, cardiac magnetic resonance imaging; DPD, -3,3-diphosphono-1,2-propanodicarboxylic acid; ECG, electrocardiogram; EMB, endomyocardial biopsy; ECHO, echocardiogram; eGFR, estimated glomerular filtration rate; HMDP, hydroxymethylenediphosphonate; IHC, immunohistochemistry; mIBG, meta-iodobenzylguanidine; MS, mass spectroscopy; v, hereditary; PYP, pyrophosphate; Tc, technetium; wt, wild-type.

Despite the widespread utilization of serum biomarkers for risk assessment of cardiac amyloidosis, biomarkers themselves are non-specific for the diagnosis of amyloidosis. This lack of specificity is primarily due to confounding by renal function and overlap with other cardiomyopathies that also result in abnormalities of NT-pro BNP and troponin. For this reason, imaging remains a requisite component of the diagnostic algorithm for cardiac amyloidosis. In addition, imaging alone captures the cardiac functional impairment caused by amyloid infiltration and affords insight into hemodynamics. Finally, imaging has the potential to directly visualize cardiac remodeling that may result from both FLC reduction, TTR stabilization/suppression, and/or the anti-amyloid specific therapies in development. This consensus document serves as means to summarize the interpretation and application of multimodal imaging in cardiac amyloidosis.

The first descriptions of echocardiographic findings in cardiac amyloidosis were reported more than 40 years ago.^1,33 Since that time, echocardiography has become a standard part of the diagnostic assessment in patients with suspected or confirmed cardiac amyloidosis.^34-37 The initial studies of echocardiography in cardiac amyloidosis occurred when only M-mode echocardiography was routinely available and predated the advent of clinical 2D and Doppler echocardiography. Nevertheless, these early studies recognized many of the findings of cardiac amyloidosis still used today in clinical practice,^34-41 along with more recent advances as discussed in subsequent sections.^1,33 Echocardiography has the advantage of portability, bedside availability, conspicuous presence, and superior diastolic function assessment. Thus, while echocardiography is not sufficient by itself, to make the diagnosis of cardiac amyloidosis, it is an essential part of the diagnostic evaluation and ongoing management of patients with this disorder.

Cardiovascular magnetic resonance in cardiac amyloidosis provides structural and functional information that complements echocardiography.⁴² Cardiovascular magnetic resonance may have advantages when acoustic windows are poor, for characterization of the right ventricle, tissue characterization based on the contrast-enhanced patterns of myocardial infiltration, and precise quantification of cardiac chamber volumes and ventricular mass. However, CMR with late gadolinium enhancement (LGE) may be relatively contraindicated in patients with suspected cardiac amyloidosis and concomitant renal failure—a frequent occurrence. Moreover, in centers where CMR scanning in patients with pacemakers is not yet routine, echocardiography may be the only option for imaging cardiac structure and function. Although both the echocardiographic and CMR assessment of structure and function alone may be non-specific, some features provide more specificity, including biventricular long axis function impairment, apical sparing, reduced stroke volume index, pericardial effusion, marked biatrial enlargement, atrial appendage thrombus in sinus rhythm, sparkling texture of the myocardium, and/or disproportionate increase in left ventricular (LV) mass for electrocardiogram (ECG) voltages. Given the limitations of assessment of structure and function alone (by echo or CMR), tissue characterization by CMR adds high value, as discussed in subsequent sections.

Radionuclide imaging provides critical information on amyloid type that complements cardiac structural and functional characterization by echocardiography and CMR. It has long been appreciated that there is a unique myocardial uptake pattern in amyloid by scintigraphy with ^99mTechnetium (Tc)-bisphosphonate derivatives (^99mTc-pyrophosphate [PYP], ^99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (^99mTc-DPD), ^99mTc hydroxymethylene- diphosphonate [^99mTc-HMDP]). Many studies dating from the 1970s and 80s suggested ^99mTc-PYP could assist in diagnosing amyloidosis.^2,43-48 However, there was variable diagnostic accuracy, which limited early use of the technique, owing to the study of mixed patients populations with undifferentiated ATTR and AL subtypes. Subsequent studies comparing ^99mTc-bisphosphonate scintigraphy to gold standard endomyocardial biopsy discovered that ATTR cardiac amyloidosis has avidity for bone radiotracers, whereas AL cardiac amyloidosis has minimal or no avidity for these tracers. Therefore, bone-avid radiotracers can definitively diagnose amyloid type when a plasma cell dyscrasia is excluded. Recognition of preferential ATTR binding to bone-avid ^99mTc-bisphosphonate-based radiotracers resulted in renewed interest and greater clinical application of cardiac scintigraphy with ^99mTc-PYP, ^99mTc-DPD, and ^99mTc-HMDP. Although there is no direct comparison between these tracers, the information available suggests they can be used interchangeably. This is fortunate, given that there is limited access to ^99mTc-DPD and ^99mTc-HDMP in the United States and ^99mTc-PYP in Europe.

Cardiac amyloidosis is substantially underdiagnosed due to varied clinical manifestations, especially in the early stages of disease. An ideal non-invasive diagnostic method would identify cardiac involvement in amyloidosis and would also confirm the etiologic subtype. No existing diagnostic tools can provide this information individually, necessitating a multimodality cardiac imaging approach.

Echocardiography plays a major role in the non-invasive diagnosis of cardiac amyloidosis due to its assessment of structure and function and its pervasive use in patients with concerning cardiac symptoms. The evaluation of cardiac amyloidosis using echocardiography focuses on morphological findings related to amyloid infiltration, in particular, thickened LV walls >1.2 cm in the absence of any other plausible causes of LV hypertrophy (Figure 2).²⁸ Although increased LV mass in the setting of low voltage ECG is suggestive of cardiac amyloidosis, a definitive distinction by echocardiography of amyloidosis from hypertrophic cardiomyopathy or other causes of LV hypertrophy is challenging.⁴⁹ Other echocardiographic findings that suggest infiltrative disease include normal to small LV cavity size; biatrial enlargement and dysfunction⁴¹; left atrial and left atrial appendage stasis and thrombi; thickened valves; right ventricular and interatrial septal thickening; pericardial effusion; and a restrictive transmitral Doppler filling pattern.^50-54 Several of these features, including an overt restrictive mitral inflow pattern are uncommon until late in the disease process.^34,51 However, reduced LV systolic thickening, filling pressures, cardiac output,³⁸ early diastolic dysfunction,^39,40 and signs of raised filling pressures are commonly seen.^34,52 A granular sparkling appearance of the myocardial walls may be appreciated, but it is not considered a highly specific finding and can be seen in other conditions, such as end-stage renal disease. The echocardiographic shift from fundamental to harmonic imaging has confounded this phenotype.

Figure 2. Characteristic appearance of cardiac amyloidosis on echocardiography. (A)-(D) 2D echocardiography. (A) (parasternal long axis) and (B) (parasternal short axis) demonstrate increased LV wall thickness with a sparkling texture of the myocardium (yellow arrows) in a patient with primary (AL) cardiac amyloidosis. Also, note the small pericardial effusion (white arrows), which is often seen in patients with cardiac amyloidosis. (C) (apical 4-chamber view) demonstrates increased biventricular wall thickness, biatrial enlargement, and increased thickening of the interatrial septum (yellow arrow) and mitral valve leaflets (white arrow) in a patient with wild-type transthyretin cardiac amyloidosis. (D) Tissue Doppler imaging (TDI) tracing taken at the septal mitral annulus in a patient with ATTR cardiac amyloidosis. The TDI tracings shows the ‘‘5-5-5’’ sign (s’ [systolic], e’ [early diastolic], and a’ [late (atrial) diastolic] tissue velocities are all <5 cm/s), which is seen in patients with more advanced cardiac amyloidosis. The dotted lines denote the 5 cm/s cut-off for systolic and diastolic tissue velocities. In addition to the decreased tissue velocities, isovolumic contraction and relaxation times (IVCT and IVRT, respectively) are increased and ejection time (ET) is decreased, findings also seen in patients with cardiac amyloidosis especially as the disease becomes more advanced.

Tissue Doppler imaging (TDI) and speckle-tracking echocardiography (STE) refine the non-invasive recognition of cardiac amyloidosis by quantitating longitudinal systolic function.^51,55,56 A pattern of reduced longitudinal shortening with preserved LV ejection fraction and radial shortening is characteristic of cardiac amyloidosis and can differentiate it from other causes of increased LV wall thickness. Longitudinal systolic function is commonly impaired, even in the earlier phases of the disease, when radial thickening and circumferential shortening are still preserved.^34,51,57-62 Both AL and ATTR cardiac amyloidosis patients demonstrate a typical pattern of distribution of STE-derived longitudinal strain in which basal LV segments are severely impaired while apical segments are relatively spared (Figure 3).^51,63 Conversely, patients with other causes of LV hypertrophy (ie, aortic stenosis, hypertrophic cardiomyopathy) typically show reduced LV longitudinal strain in the regions of maximal hypertrophy.^63,64

Figure 3. Left ventricular longitudinal strain abnormalities. (A) (apical 4-chamber view), (B) (apical 2-chamber view), (C) (apical 3-chamber view) all show abnormal longitudinal strain in the basal and mid segments with relative preservation in the apical segments (purple and green curves, white arrows) in a patient with ATTRv cardiac amyloidosis. (D) shows the corresponding bullseye map of the longitudinal strain pattern throughout the left ventricle with the ‘‘cherry-on-the-top’’ sign (red denotes normal longitudinal strain at the apex and pink/blue denotes abnormal longitudinal strain at the mid/basal left ventricle).

Another abnormal quantitative measure of LV contractility in cardiac amyloidosis is the myocardial contraction fraction (MCF), the ratio of stroke volume to myocardial volume. The MCF is an index of the volumetric shortening of the myocardium that is independent of chamber size and geometry and highly correlated with LV longitudinal strain.^65-67 Abnormalities beyond the left ventricle can also suggest cardiac amyloidosis. Recently, it has been reported that the stroke volume index has a prognostic performance similar to LV strain in predicting survival in AL cardiac amyloidosis, independently of biomarker staging. Because the stroke volume index is routinely calculated and widely available, it could serve as the preferred echocardiographic measure to predict outcomes in AL cardiac amyloidosis patients. Left atrial reservoir and pump functions measured by strain are frequently impaired, irrespective of left atrial size, suggesting that both raised LV filling pressures and direct atrial amyloid infiltration (as documented by CMR studies) contribute to left atrial dysfunction.^41,68 This dysfunction may result in the formation of atrial and atrial appendage thrombi, even in the setting of normal sinus rhythm, exposing patients to higher relative risk for embolic strokes. Although data are not available, clinical experience from major amyloidosis centers suggest the highly thrombogenic milieu of the left atrium increases cardioembolic risk in these patients.⁶⁹ The right ventricle is often affected due to a combination of increased afterload from pulmonary hypertension and intrinsic right ventricular amyloid infiltration, resulting in reduced tricuspid annular plane systolic excursion, tissue Doppler systolic velocity, and longitudinal strain.⁷⁰

As echocardiographic findings lack the tissue characterization provided by CMR, echocardiographic diagnosis of cardiac amyloidosis relies on the presence of highly suggestive findings that can confirm diagnostic suspicion.^34,71Table 1 lists the echocardiographic parameters for acquisition, interpretation, and reporting in cardiac amyloidosis. Moreover, abnormal parameters are provided that suggest cardiac amyloidosis and warrant further evaluation. The combination of these echocardiographic “red flags” with other parameters, such as abnormal cardiac biomarkers and electrocardiographic findings, maximizes diagnostic accuracy.⁷² For instance, the combination of pericardial effusion and symmetric LV wall thickening in the presence of low or normal QRS voltages should prompt a strong suspicion of cardiac amyloidosis.^50,72,73 In particular, the ratio of QRS voltage to echocardiographic LV wall thickness is useful in diagnosing cardiac amyloidosis.⁴⁹

Table 1. Standardized Acquisition, Interpretation, and Reporting of Echocardiography for Cardiac Amyloidosis

2D, 2 dimensional; A, late (atrial) mitral inflow velocity; AL, amyloid light chain; ATTR, amyloid transthyretin; E, early mitral inflow velocity; E/A, ratio of early to late (atrial) mitral inflow velocities; ECG, electrocardiogram; ESRD, end-stage renal disease; IVCT, isovolumic contraction time; IVRT, isovolumic relaxation time; LA, left atrial; LV, left ventricular; PA, pulmonary artery; RA, right atrium; STE, speckle-tracking echocardiography; TDI, tissue Doppler imaging.

Comprehensive 2D echocardiography, including quantitative tissue Doppler and speckle-tracking strain analysis (when available) should be performed in all patients with unexplained LV wall thickening and a clinical suspicion of cardiac amyloidosis.

To increase identification of this underdiagnosed disease, any echocardiographic abnormalities suggestive of cardiac amyloidosis should prompt further evaluation.

• Echocardiographic parameters should be combined with electrocardiographic, clinical, biomarker, and other imaging findings to maximize diagnostic accuracy.

Cardiac magnetic resonance has a central role in the non-invasive diagnosis of cardiac amyloidosis due to its ability to provide tissue characterization in addition to high-resolution morphologic and functional assessment. Cardiac magnetic resonance offers value in two clinical scenarios: the differentiation of cardiac amyloidosis from other cardiomyopathic processes with increased wall thickening and potentially in detection of early cardiac involvement in patients with evidence of systemic amyloidosis. A comprehensive CMR evaluation for cardiac amyloidosis includes morphologic and functional assessment of the left and right ventricles and atria using cine imaging, evaluation of native T1 signal (assessed on non-contrast T1 mapping), assessment of LGE, and extracellular volume (ECV) measurement. Overall, current published reports from single-center studies demonstrated heterogeneity in study design, sample size, and types of amyloidosis included. See the Appendix for a summary of the published literature on diagnosis of cardiac amyloidosis using CMR.

Maceira et al described a typical LGE pattern in cardiac amyloidosis of global subendocardial enhancement.⁷⁴ Initial observations were that nulling—rendering remote myocardium dark—was difficult in cardiac amyloidosis. The blood pool and myocardium null together due to expansion of the extra cellular myocardial volume (from amyloid infiltration) which approaches plasma volume. An inversion time scout (TI-scout) technique (obtaining a series of images with various inversion time values) could be useful to select the optimal inversion time for the LGE sequence.⁷⁵ Traditional LGE imaging techniques, however, can be difficult to acquire and interpret in cardiac amyloidosis. Late gadolinium enhancement using the widely available relatively new phase-sensitive inversion recovery sequence (PSIR) eliminates the need to optimize null-point settings, making LGE in cardiac amyloidosis more robust and operator independent. Using the PSIR technique, LGE is significantly more specific and sensitive than echo or CMR functional assessment. Although multiple LGE distributions have been described in cardiac amyloidosis, subendocardial and transmural LGE patterns predominate. Both patterns are present in AL and ATTR cardiac amyloidosis, but to different extents, with subendocardial LGE being more prevalent in AL and transmural LGE more prevalent in ATTR cardiac amyloidosis.⁷⁶ Late gadolinium enhancement shows an initial basal predilection but with biventricular transmurality in advanced disease.^77-80

At 4 minutes post-gadolinium administration, a subendocardial-blood T1 difference of 191 ms detected cardiac amyloidosis at 90% and 87% sensitivity and specificity, respectively.⁷⁴ In several studies where the results of an endomyocardial biopsy has been used as a reference standard, a typical LGE pattern has consistently been shown to have a diagnostic sensitivity of 85% to 90%.^{74,77,78,80-82} However, the true specificity of LGE in diagnosing cardiac amyloidosis with reference to histologic evidence cannot be accurately determined, given verification bias (typically only positive CMR cases are referred for endomyocardial biopsy). A recent meta-analysis based on seven published studies, estimated a sensitivity and specificity of 85% and 92%, respectively, for CMR-based LGE in diagnosing cardiac amyloidosis.⁸³

Other CMR methods include native (non-contrast)^82,84 and post-contrast T1 mapping,⁸⁵ left atrial LGE,⁶⁸ and qualitative visual T1 comparison between the myocardium and cardiac blood pool.⁷⁹ The method of a nulling comparison between the myocardium and the blood pool allows a rapid confirmation of cardiac amyloidosis diagnosis as an adjunct to LGE findings, at an excellent sensitivity but a moderate specificity.⁷⁹ Late gadolinium enhancement in non-ischemic cardiomyopathies, especially cardiac amyloidosis, is not easy to quantify; therefore, using LGE to track changes over time can be difficult. T1 mapping is a new technique where a direct quantitative signal from the myocardium is measured, either pre-contrast (native T1) or post-contrast (ECV).⁸⁶ T1 mapping before and after contrast administration allows a quantitative measure of the contrast exchange between the blood pool and the expanded extracellular compartment, thus permitting an incremental characterization and detection of the degree of infiltration.

Native T1 may find particular utility when administration of contrast is contraindicated. Of note, a recent report demonstrated that native myocardial T1 measured by the shortened modified look-locker inversion recovery (ShMOLLI) method achieved a diagnostic sensitivity and specificity of 92% and 91%, respectively.⁸² Native T1, however, is a composite signal from the extra- and intracellular space, and administration of contrast with ECV measurement enables us to isolate the signal from the extracellular space.⁸⁶ Amyloidosis is an exemplar of interstitial disease, and this is reflected by substantial elevation of ECV in patients with AL and ATTR cardiac amyloidosis.^85,87 Extracellular volume is also elevated even when conventional testing and LGE suggest no cardiac involvement, highlighting a potential role of ECV as an early disease marker.⁸⁸ Both native T1 and ECV track a variety of markers of disease activity, and there is early evidence they could be used to track changes in amyloid burden over time.

Advanced techniques, such as T2 mapping and perfusion are being used to assess additional aspects of the cardiac amyloidosis phenotype, including myocardial edema⁸⁹ and coronary microvascular dysfunction. Using a combination of CMR features, a measure of the likelihood of cardiac amyloid type (ATTR vs AL), and likelihood of ATTR vs AL can be gleaned^90,91; but, this is typically not sufficient for excluding AL cardiac amyloidosis. Free light chains combined with cardiac scintigraphy with bone tracers have advantages over echo and CMR for differentiation of the type of cardiac amyloidosis.³

Comprehensive CMR-based evaluation of cardiac structure, function, and myocardial tissue characterization is helpful for diagnosis of cardiac amyloidosis, particularly when echocardiographic findings are suggestive or indeterminate.

In patients with biopsy-proven systemic amyloidosis, typical CMR findings, including diffuse LGE, nulling of myocardium before or at the same inversion time as the blood pool, and extensive ECV expansion are combined with structural findings of increased wall thickness and myocardial mass to diagnose cardiac involvement. In the absence of documented systemic amyloidosis, typical CMR features should prompt further evaluation for cardiac amyloidosis.

Cardiovascular magnetic resonance, however, is typically unable to definitively distinguish AL from ATTR cardiac amyloidosis.

Cardiovascular magnetic resonance parameters should be combined with electrocardiographic, clinical, biomarker, and other imaging findings to maximize diagnostic accuracy.

Radionuclide imaging plays a unique role in the non-invasive diagnosis of cardiac amyloidosis. A variety of ^99mTc-labeled diphosphonate and PYP (bone-avid) compounds diagnose ATTR cardiac amyloidosis with high sensitivity and specificity.³ Targeted amyloid binding ¹⁸F-positron emission tomography (PET) tracers are highly specific to image amyloid deposits and appear to bind to both AL and ATTR.^92-96123I-meta-iodobenzylguanidine (mIBG), an established tracer for imaging myocardial denervation, has been utilized to image myocardial denervation in familial ATTR cardiac amyloidosis.^97,98 A substantial additional benefit of radionuclide evaluation of cardiac amyloidosis is that whole-body imaging can be performed concurrently, allowing evaluation of multi-organ systemic involvement.

The explanation for this differential uptake in ATTR vs AL cardiac amyloidosis is unknown, but it has been suggested that the preferential uptake by ATTR may be a result of higher calcium content.^99,100 Furthermore, the type of mutation and the result of the proteolysis of myocardial fibers (full-length only vs full length plus C-terminal ATTR fragments) also modulate uptake of bone radiotracers by amyloid fibrils.¹⁰⁰

Systematic evaluation of diphosphonate radiotracers suggests that cardiac uptake of ^99mTc-PYP, ^99mTc-DPD, and ^99mTc-HMDP are remarkably sensitive (but not completely specific) for ATTR cardiac amyloidosis.^3,31,100-106 Notably in the absence of cardiac amyloidosis (or previous myocardial infarction), there is no myocardial uptake of bone tracers; therefore, cardiac scintigraphy with bone-avid radiotracers may reliably distinguish cardiac amyloidosis from other entities that mimic cardiac amyloidosis, such as hypertrophic cardiomyopathy.^31,102 Cardiac scintigraphy with bone-avid radiotracers is particularly sensitive in the early identification of ATTR cardiac amyloidosis, including carriers without apparent cardiac involvement by other diagnostic techniques.^105,107,108 Furthermore, ^99mTc-DPD/HMDP allow the possibility of detecting extra-cardiac (skeletal muscle and lung) amyloid infiltration.^109,110 See the Appendix for a full summary of the published literature on diagnosis of ATTR cardiac amyloidosis using ^99mTc-PYP/DPD/HMDP.

A multicenter experience in 1498 patients showed a positive predictive value for ATTR cardiac amyloidosis of 100% (95% confidence interval, 98.0-100) in patients with an echocardiogram or CMR consistent with or suggestive of cardiac amyloidosis, and absence of monoclonal protein using urine and serum, with serum FLC assay and immunofixation electrophoresis.³ A recent bivariate meta-analysis confirmed the accuracy of bone scintigraphy in the assessment of ATTR cardiac amyloidosis.¹¹¹ Again, these high sensitivities and specificities were reported from major centers of expertise and in patients with advanced stages of the disease, and often with New York Heart Association (NYHA) heart failure greater than Class II. The yield of ^99mTc-PYP/DPD/HMDP cardiac scintigraphy in patients with earlier stages of disease or with pre-clinical disease is yet to be confirmed.

Several diagnostic parameters have been evaluated on cardiac scintigraphy with bone-avid tracers. The ratio of heart-to-contralateral (H/CL) lung uptake (semi-quantitative scoring), heart-to-whole-body (H/WB) retention, and a heart-to-bone ratio (visual grade) have been assessed at both 1 and 3 hours (see Standardized Imaging Techniques). Early work by Perugini and colleagues found that a visual grade ≥2 on ^99mTc-DPD (ie, moderate or strong myocardial uptake) was 100% sensitive to identify ATTR cardiac amyloidosis and 100% specific to distinguish from AL and control subjects.¹¹² Subsequent studies have confirmed the high sensitivity to detect ATTR cardiac amyloidosis and showed that mild uptake of ^99mTc-DPD (Grade 1) may be noted in patients with other subtypes of cardiac amyloidosis (ie, AL, Amyloid A amyloidosis, and Apolipoprotein A1).^43,106 Rapezzi et al¹⁰⁵ evaluated the ratio of heart-to-whole-body retention of ^99mTc-DPD, on the late (3-hour) images, in patients with TTR mutation, and demonstrated that individuals with increased LV myocardial wall thickness >1.2 cm had much higher heart-to-whole-body retention ratio compared to individuals with normal LV wall thickness. In a single-center experience, Bokhari et al¹¹³ identified a very high diagnostic accuracy (area under the curve of 0.992, P<0.0001) for visual Grade ≥2 and a H/CL ratio ≥1.5 on 1-hour images to distinguish ATTR from AL cardiac amyloidosis.^3,111 A H/CL ratio ≥1.3 has been proposed to distinguish ATTR accurately from AL cardiac amyloidosis on the late (3-hour) ^99mTc-PYP images.¹¹⁴

The recently-developed consensus algorithm for non-invasive diagnosis of cardiac amyloidosis attributes a central role to ^99mTc-PYP/DPD/HMDP cardiac scintigraphy (Figure 4).³ If cardiac amyloidosis is suspected clinically or based on echocardiography/CMR, blood and urine should be analyzed for evidence of a monoclonal protein and ^99mTc-PYP/DPD/HMDP cardiac scintigraphy should be considered if ATTR cardiac amyloidosis is suspected. If both tests are negative, then current evidence suggests that cardiac amyloidosis is very unlikely. It is still possible, however, for patients with ATTRv to have negative findings on DPD scintigraphy¹⁰⁵ in case of some rare non-V30M mutations and in some V30M mutations with early onset and only full-length TTR fibrils.¹¹⁵ In the presence of a Grade 2 or 3 positive ^99mTc-PYP/DPD/HMDP cardiac scan (see section on standardized imaging techniques) without evidence for monoclonal proteins in blood and urine, a diagnosis of ATTR cardiac amyloidosis can be made without a biopsy (specificity and positive predictive value >98%).³ For those patients with evidence of a plasma cell dyscrasia, a histological diagnosis is still required because the presence of low-grade uptake on a ^99mTc-PYP/DPD/HMDP scan is not 100% specific for ATTR cardiac amyloidosis, and substantial uptake (Grade 2 or 3) has been reported in more than 20% of patients with AL cardiac amyloidosis.³ The writing group would like to emphasize the importance of excluding monoclonal process with serum/urine immunofixation and a serum FLC assay in all patients with suspected amyloidosis.

Figure 4. Consensus algorithm for noninvasive diagnosis of cardiac amyloidosis. This algorithm provides an approach to the evaluation of patients with cardiac amyloidosis. Among patients with suspected cardiac amyloidosis, Grade 2 or 3 uptake of ^99mTc-PYP/DPD/HMDP uptake in the absence of a clonal abnormality is highly specific to diagnose ATTR cardiac amyloidosis avoiding the need for endomyocardial biopsy. Patients with any abnormal serum/ urine immunofixation or a positive serum free light chain assay should be referred for further evaluation to a hematologist ideally with amyloidosis experience. (Figure reproduced with permission from Gillmore JD, et al. Circulation. 2016;133:2404-2412.)

^99mTc-PYP/DPD/HMDP scintigraphy has been recently used to detect ATTR cardiac amyloidosis in previously unexplored clinical settings, including heart failure with preserved ejection fraction (prevalence 15%)^13,15 and severe degenerative aortic stenosis,^14,116 including the “paradoxical low-flow low-gradient” subtype (18%).¹¹⁷

Based on the utility of cardiac scans with SPECT bone-avid radiotracers, there has been interest in ¹⁸F-NaF, a PET bone radiotracer, for imaging cardiac amyloidosis.^118,119 Early reports, however, suggest limited utility for imaging ATTR cardiac amyloidosis, and further studies are warranted to examine its utility.

Several amyloid binding SPECT and PET radiotracers are available for amyloidosis imaging. ^99mTc-aprotinin^120-122 and ¹²³I-serum amyloid P-component (¹²³I-SAP)¹²³ were originally developed to image systemic amyloidosis but have limited availability. They have not been useful to image cardiac amyloidosis due to poor signal-to-noise ratio¹²³ and concerns for risk of bovine encephalopathy.¹²¹ In contrast, several PET amyloid-binding radiotracers, structurally similar to thioflavin-T and likely binding to the amyloid fibril structure, approved for imaging beta amyloid in Alzheimer’s disease,¹²⁴ have been successfully used to image cardiac amyloidosis. ¹¹C-Pittsburgh compound B (PIB) was one of the first PET radiotracers developed for beta-amyloid imaging but is limited in availability to sites with a cyclotron. ¹⁸F-florbetapir, ¹⁸F-florbetaben, and ¹⁸F-flutemetamol developed subsequently and are currently FDA approved for beta-amyloid imaging and are widely commercially available. Several additional tracers are still under development.¹²⁵

¹¹C-PIB,^92,9518F-florbetapir,^93,96 and ¹⁸F-florbetaben⁹⁴ have been evaluated in patients with AL and ATTR cardiac amyloidosis. In these pilot studies, high cardiac radiotracer uptake was consistently reported in patients with cardiac amyloidosis compared to controls, including hypertensive controls. A target to background (LV myocardium to blood pool) ratio >1.5 and a retention index of >0.025 min^-1 with ¹⁸F-florbetapir⁹³ and ¹⁸F-florbetaben⁹⁴ have been shown to separate patients with cardiac amyloidosis from controls without amyloidosis. Myocardial retention of ¹¹C-PIB,⁹²¹⁸F-florbetapir,⁹³ and ¹⁸F-florbetaben⁹⁴ was significantly greater in cardiac amyloidosis patients compared to controls. In one study, although ¹⁸F-florbetapir myocardial retention index was lower in ATTR compared to AL cases, definitive subtype differentiation was not feasible⁹³; similar findings were confirmed with ¹⁸F-florbetaben.⁹⁵ Although not studied serially, retention of ¹¹C-PIB was lower in AL cardiac amyloidosis patients treated with chemotherapy as compared to those who did not undergo treatment,⁹⁵ suggesting it is possible this radiotracer will be useful for disease monitoring. Finally, unlike echocardiography or CMR, amyloid-binding PET tracers can image systemic amyloid deposits in various other organs^126,127 and offer the potential to quantify the load of amyloidosis in the whole body.

As literature on PET amyloid-binding radiotracers is limited, sections on risk assessment and standardized protocols are not provided for these radiotracers.

Patients with amyloidosis are prone to autonomic dysfunction from amyloid infiltration of myocardial and nerve conduction tissue, resulting in rhythm disorders.¹²⁸ Autonomic dysfunction is most common in ATTR cardiac amyloidosis, particularly ATTRv, where it has been studied extensively.^129,130 Notably, cardiac dysautonomia may occur independent of the presence of a typical restrictive cardiomyopathy.¹³¹ In patients with ATTRwt cardiac amyloidosis, polyneuropathy and dysautonomia are less common, seen in approximately 9%.¹³² While AL cardiac amyloidosis patients less commonly manifest autonomic dysfunction,¹³³ it may develop as a complication of AL amyloidosis treatment.¹³⁴ Therefore, autonomic denervation is a non-specific finding. ¹²³I-mIBG scintigraphy is not able to discriminate between cardiac amyloidosis subtypes nor differentiate cardiac amyloidosis from other forms of cardiomyopathy.¹³⁵ However, cardiac denervation evidenced by mIBG occurs earlier than amyloid deposit detection by diphosphonate scintigraphy in TTR mutation carriers.¹³⁶ Although secondary (Amyloid A, AA) amyloidosis rarely shows cardiac manifestations, myocardial denervation has been reported in one study.¹³⁵

While amyloid infiltration of the cardiac autonomic system cannot be directly imaged, multiple tracers assess autonomic myocardial denervation, including ¹²³I-mIBG, ¹²⁴I-mIBG, N-[3-Bromo-4-3-[¹⁸F-]fluoro-propoxy)-benzyl]-guanidine LM1195, and ¹¹C-hydroxy-ephedrine. ¹²³I-mIBG, a chemically modified analogue of norepinephrine, is stored in vesicles in presynaptic sympathetic nerve terminals and is not further catabolized. ¹²³I-mIBG has been specifically studied in cardiac amyloidosis, and semi-quantitative analysis of ¹²³I-mIBG cardiac uptake compared to background (heart-to-mediastinal ratio [HMR]), provides indirect information of amyloid infiltration in the sympathetic nerve system.^{97,98,130,135,137-141} Decreased HMR at 4 hours after tracer administration (late HMR) reflects the degree of sympathetic dystonia, and is an independent prognostic factor in the development of ventricular dysrhythmia. PET imaging of sympathetic innervation in cardiac amyloidosis has not yet been studied. See the Appendix for a summary of the published literature on assessment of autonomic myocardial innervation imaging in amyloidosis using ¹²³I-mIBG.

Angina, in the absence of coronary artery disease, is common in patients with cardiac amyloidosis. Endothelial¹⁴² and microvascular dysfunction¹⁴³ have been described and may precede the clinical diagnosis of cardiac amyloidosis.^143,144 In one study, focal and global subendocardial hypoperfusion at rest and post-vasodilator stress were ubiquitous in patients with AL and ATTR cardiac amyloidosis.¹⁴⁵ Absolute myocardial blood flow¹⁴⁵ and coronary flow reserve^144,145 are substantially reduced in patients with cardiac amyloidosis, despite absence of epicardial coronary artery disease. Whether coronary microvascular dysfunction improve after successful anti-amyloid therapy is not known.

Myocardial imaging with ^99mTc-PYP/DPD/HMDP, in the appropriate clinical context, is highly sensitive and specific to diagnose ATTR cardiac amyloidosis and may aid in its early detection.

In the absence of a light-chain clone, myocardial uptake of ^99mTc-PYP/DPD/HMDP of Grade ≥2 is diagnostic of ATTR cardiac amyloidosis, obviating the need for endomyocardial biopsy.

To facilitate early diagnosis of ATTR cardiac amyloidosis, cardiac ^99mTc-PYP/DPD/HMDP scintigraphy should be more broadly considered in patients with unexplained increased LV wall thickness, heart failure with preserved ejection fraction, familial amyloid polyneuropathy (FAP), family history of amyloidosis, degenerative aortic stenosis with low-flow low gradient in the elderly, and a history of bilateral carpal tunnel syndrome.

¹²³I-mIBG can detect cardiac denervation in patients with hereditary ATTR amyloidosis.

Cardiac involvement is common in systemic AL and ATTR amyloidosis and markedly impacts quality of life and outcome.¹⁴⁶ Thus, cardiac assessment in patients with systemic amyloidosis is crucial for risk stratification and treatment decisions. Imaging plays a key role in risk stratification of patients with AL and ATTR cardiac amyloidosis and may add to the existing clinical and biomarker-based risk stratification as discussed previously (see section “Biomarkers and Biopsy in Cardiac Amyloidosis”).

Abnormalities in several echocardiographic imaging parameters (eg, LV longitudinal strain, early mitral inflow [E-wave], deceleration time, myocardial performance index, pericardial effusion) are associated with worse outcomes and should alert the clinician to the potential of advanced disease.^{9,50,59,65,70,147-165} At the present time, however, there is no formal staging system for ATTRv, ATTRwt, or AL cardiac amyloidosis that uses echocardiographic parameters. Therefore, echocardiography findings in isolation should not be used to determine risk in the individual patient with cardiac amyloidosis. Additional studies to assess the optimal risk-stratification algorithm that incorporates multiple echocardiographic parameters are needed. Moreover, further study is needed to demonstrate the incremental value of echocardiographic parameters over simple clinical markers (eg, New York Heart Association functional class, B-type natriuretic peptide, troponin, glomerular filtration rate) and radionuclide and CMR imaging findings. See the Appendix for a summary of the published literature on the prognostic value of echocardiography in cardiac amyloidosis.

Multiple CMR measures have prognostic significance in cardiac amyloidosis, including LGE presence and pattern, native T1, post-contrast T1, and multiple morphologic parameters.¹⁶⁶ Despite the excellent discriminative capacity of LGE, conflicting results were initially reported describing its prognostic impact in cardiac amyloidosis.^78,80,81,167 At that time, LGE patterns of cardiac amyloidosis were heterogeneous due to non-standardized acquisition and analysis. The transition to more robust LGE approaches, such as PSIR,¹⁶⁸ has markedly improved image quality. This tool has provided insight into progression of both AL and ATTR cardiac amyloidosis through visualization of a continuum of amyloid accumulation as determined by progression of the LGE pattern from normal to subendocardial to transmural.^76,169 As a result, several studies now show that the LGE pattern can serve as an independent predictor of prognosis after adjustments for echocardiographic characteristics and blood biomarkers (NT-proBNP and troponin) have been performed.¹⁷⁰ Importantly, the LGE pattern confers prognosis in both AL and ATTR cardiac amyloidosis. Despite its prognostic usefulness, LGE does not lend itself readily toward quantification of myocardial infiltration, owing to different patterns and signal intensities. Thus, the capacity of LGE to track changes accurately over time and monitor response to treatment is unknown. Parametric T1 mapping has the potential to overcome these limitations.⁸⁶ Recent studies have shown that higher native myocardial T1 can accurately stratify worse prognosis in AL cardiac amyloidosis¹⁷¹ but not in ATTR cardiac amyloidosis.⁸⁷ Alternatively, T1-derived ECV has been associated with prognosis in AL and ATTR cardiac amyloidosis after adjustment for known independent predictors.^171,172 T2 mapping, a measure of myocardial edema, adds a third dimension to the tissue characterization; in patients with AL cardiac amyloidosis, it is an independent predictor of prognosis.⁸⁹ See the Appendix for a summary of the published literature on the prognostic value of CMR in cardiac amyloidosis.

The prognostic role of ^99mTc-PYP/DPD/HMDP scintigraphy and ¹²³I-mIBG have been explored in several studies. ^99mTc-PYP/DPD/HMDP cardiac uptake moderately correlates positively with LV wall thickness and mass, troponin T, NT-proBNP, and ECV; it correlates negatively with LV ejection fraction.^{31,102,105,173-175} The degree of cardiac uptake correlates with overall mortality and survival free from major adverse cardiac events. Multiple semi-quantitative markers of cardiac uptake have been studied, including heart and heart-to-whole-body retention,^105,174 heart/skull ratio, 102 H/CL ratio,¹⁷³ and visual scoring.¹⁷⁵ In a multicenter study using ^99mTc-PYP, an H/CL ratio of >1.5 was associated with worse survival among patients with ATTR cardiac amyloidosis.¹⁷³ Similar data was found in a single-center study in patients with suspected ATTR cardiac amyloidosis,¹⁷⁵ and these same authors found that regional variability of ^99mTc-PYP uptake may also predict mortality.¹⁷⁶ In all these studies, combining the degree of cardiac uptake with an anatomical (interventricular septal thickness) or functional (Class NYHA, NT-proBNP) variable improved prognostic risk stratification. Of note, visual grading of ^99mTc-PYP/DPD/HMDP has not been shown to be an independent predictor of outcomes.^31,114

Cardiac sympathetic denervation is associated with decreased survival in ATTRv cardiac amyloidosis.^129,131 A late decreased HMR <1.6 portends a poor prognosis and can be used to identify ATTRv cardiac amyloidosis patients who would benefit from liver transplantation.¹³¹ After liver transplantation, cardiac sympathetic denervation does not appear to progress¹³⁸ and has questionable independent prognostic significance.¹³¹ The prognostic relevance of late-HMR reduction is less clear in AL and ATTRwt cardiac amyloidosis.^135,138-140 See the Appendix for a summary of the published literature on the prognostic value of radionuclide imaging in ATTR cardiac amyloidosis.

Multiple imaging parameters predict a worse prognosis, including increased LV mass, lower global longitudinal strain, increased right ventricular wall thickness, higher native T1 and ECV, higher H/CL ratio, and ¹²³I-mIBG increased HMR and delayed washout rate.

Although not formally incorporated into current risk-assessment algorithms, radionuclide results should be combined with electrocardiographic, clinical, biomarker, and other imaging findings for optimal prognostication.

The ideal method for evaluating the time course of the disease and the response to treatment, particularly disease-modifying treatments, should provide a precise quantitative measure of systemic and cardiac amyloid burden. In AL cardiac amyloidosis, cardiac response is assessed using the serum NT-proBNP concentration, a substantial reduction of which consistently predicts clinical improvement and extended survival.¹⁷⁷ However, the advent of anti-amyloid therapies demands the development of imaging techniques that can estimate the amyloid burden in the heart.

Echocardiography remains the cornerstone of serial assessment of LV dysfunction in patients with heart failure. However, there is relatively little information in the echocardiographic cardiac amyloidosis literature regarding assessment of disease progression and response to therapy. A few studies have shown potential benefit for the use of echocardiography in the following areas: (1) to demonstrate changes in cardiac disease in response to treatment in patients with AL cardiac amyloidosis^29,178,179; (2) to determine whether patients with cardiac amyloidosis need to be anticoagulated for stroke prophylaxis; (3) to diagnose progressive cardiac involvement after liver transplantation in patients with ATTRv amyloidosis^180-182; and (4) to assess LV ejection fraction in patients with AL amyloidosis being considered for stem-cell transplantation.¹⁸³ Due to a higher incidence of cardiac thrombi in patients with cardiac amyloidosis, some centers consider a transesophageal echocardiogram prior to cardioversion of atrial arrhythmias, even in patients on therapeutic anticoagulation. Emerging data suggests that echocardiographic LV global longitudinal strain may be a marker of disease progression and response to therapy.¹⁸⁴ In contrast, T1 mapping with ECV measurement by CMR can track multiple parameters of structural change (amyloid burden and cardiomyocyte response). In a small retrospective study, the prevalence of a decrease in LV mass and ECV on CMR was higher in patients with AL cardiac amyloidosis and a complete response or very good partial response to chemotherapy.⁸⁸ The quantitative nature of CMR makes it a promising tool to monitor disease progression and response to therapy. Although ^99mTc-PYP/DPD/HMDP scintigraphy correlates well with anatomic and functional variables, this technique has not been definitively proven to quantify changes in response to current therapies, and thus repeat studies are not typically clinically useful.¹⁸⁵ Positron emission tomography is inherently more sensitive and quantitative, and holds the possibility of monitoring response to therapy with PET amyloid-binding tracers once adequately studied. Serial myocardial denervation studies have been studied in ATTRv amyloidosis to guide timing of liver transplantation.¹⁸⁶ Experience with implantable cardioverter defibrillators (ICDs) in cardiac amyloidosis is limited,^187,188 and the indication for ICD implantation in these patients is unclear even in the setting of myocardial denervation. Prospective studies are needed in this area. The role of imaging to guide referral to cardiac transplantation and monitor for recurrence post-transplant is not well elucidated and needs further study.

Notably, none of the imaging techniques have been validated for assessing response to therapy, and no study has correlated changes in imaging findings after therapy with survival.

Transthoracic echocardiography is reasonable to monitor disease progression and/or response to therapy in cardiac amyloidosis because echocardiography is often done clinically for other reasons (ie, heart failure management).

Transthoracic echocardiography (for the evaluation of left atrial size and function) and transesophageal echocardiography (for the evaluation of the left atrial appendage) are useful to guide initiation and management of anticoagulation in patients with cardiac amyloidosis.

Cardiovascular magnetic resonance assessment of LV wall thickness, LV mass, and particularly ECV is emerging as a tool to assess disease progression and response to therapy.

Serial SPECT ^99mTc-PYP/DPD/HMDP scintigraphy is currently not recommended to assess disease progression or response to therapy.

Extensive research has been performed in cardiac amyloidosis using varied protocols without a clear consensus. This section will provide recommendations for standardized image acquisition, interpretation, and reporting in the assessment of cardiac amyloidosis using echocardiography, CMR, and radionuclide imaging. Standardization would facilitate comparability and reproducibility within and across institutions and enable pooling of data for research purposes.

2D and Doppler echocardiographic acquisition in patients with suspected or known cardiac amyloidosis should follow the ASE/European Association of Cardiovascular Imaging (EACVI) guidelines,^40,189 and all standard transthoracic echocardiography views should be obtained. Required and optional reporting recommendations are provided in Table 1. When reporting results of the echocardiogram in this population, it is important to distinguish other forms of LV hypertrophy from increased LV wall thickness due to amyloid infiltration. Therefore, the report should include not only wall-thickness measurements but also qualitative assessment of the “texture” of the myocardium. Other morphologic features that can be helpful for the diagnosis of cardiac amyloidosis (eg, atrial enlargement; increased relative wall thickness defined as two times posterior wall thickness/LV end-diastolic dimension; thickening of the interatrial septum and/or valves; and the presence of a pericardial effusion) should also be reported. The visual assessment of the loss of longitudinal motion of the heart on 2D imaging (ie, minimal descent of the base in the apical views) can be helpful to include in the report as it increases the likelihood of cardiac amyloidosis.

In patients with cardiac amyloidosis, right ventricular involvement confers a worse prognosis; thus, right ventricular wall thickness (measured in the subcostal view at end-diastole) and assessment of right ventricular systolic function should be included in the report.¹⁸⁹ On Doppler assessment, evaluation of diastolic function (mitral inflow velocities, early mitral inflow [E] deceleration time, and early diastolic relaxation velocity on tissue Doppler imaging (TDI), [see section below]) should be reported.^39,40 In addition, estimation of hemodynamics (including right atrial pressure, pulmonary artery systolic pressure, LV filling pressure [based on E/e' ratio], and cardiac output [based on LV outflow tract diameter, velocity-time integral (on pulse wave Doppler)]) is helpful for the management of heart failure.³⁸

Accurate tissue Doppler images should be obtained per ASE and EACVI recommendations.⁴⁰ As shown in Figure 2, in the setting of cardiac amyloidosis, s', e', and a' velocities are all often reduced, and should be reported. The right ventricular free wall TDI should be measured, and the s' velocity reported as a measure of right ventricular longitudinal systolic function (<10 cm/s is abnormal).¹⁹⁰ In addition, isovolumic relaxation and contraction times are increased, and ejection time is decreased. Although not widely used in clinical practice, these three indices can be combined to calculate the myocardial performance index (ejection time/[isovolumic relaxation time+isovolumic contraction time]), which is also reduced in the majority of patients with overt cardiac amyloidosis.¹⁶⁵

High-quality longitudinal strain STE LV curves should be obtained in the apical 2-, 3-, and 4-chamber views at frame rates of 50-80 fps with good endocardial border definition (Figure 3). Right ventricular free wall strain is calculated as the average of the basal, mid, and apical longitudinal segmental strains. The curves for the left atrium should be generated using P-P gating, if the patient is in normal sinus rhythm. In patients with atrial fibrillation or other rhythm with a lack of P waves, there will be no booster component to the left atrial strain curve, and the left atrial conduit and reservoir strains will be equal to each other.^191,192 Emerging literature (scientific abstract not yet published) suggests that transesophageal echocardiography should be considered in patients with suspected cardiac amyloidosis and distal embolization to rule out left atrial and left atrial appendage thrombi even in the setting of normal sinus rhythm.

Ideally, in all patients with suspected or known cardiac amyloidosis, the global LV longitudinal strain value (which is calculated using the peak negative instantaneous average of the 18 longitudinal segmental strains [6 in each of the apical views]) should be reported.^193,194 In addition, a description and assessment of the pattern displayed on the global longitudinal strain bullseye map (as shown in Figure 3) should be included in the report.⁶³ Right ventricular free wall strain can also be reported. If left atrial strain is performed, the values of the reservoir, conduit, and booster strains can be reported.

Echocardiograms in patients with suspected or known cardiac amyloidosis should be obtained using ASE/EACVI guidelines on comprehensive echocardiography.

Reporting should include assessment of wall thickness and myocardial “texture”; thickening of other cardiac structures; pericardial effusion; tissue Doppler velocities (s', e', and a'); diastolic function; and hemodynamics.

Speckle-tracking echocardiography should be performed routinely in patients with suspected or known cardiac amyloidosis when available, and efforts should be made to optimize the apical 2D imaging views for speckle-tracking analysis. The global longitudinal strain and pattern of segmental strains (ie, ‘bullseye’ map) should be reported. RV and LA strain can be reported when performed.

An overall reporting on likelihood of amyloidosis based on imaging findings is recommended (not suggestive, strongly suggestive, or equivocal for cardiac amyloidosis).

Cardiovascular magnetic resonance assessment of structure and function in patients with suspected or known cardiac amyloidosis follows well-standardized protocols (Table 2).¹⁹⁵ Image interpretation and reporting should highlight effusions, atrial thrombi, long axis function, and stroke volumes in addition to LV and right ventricular ejection fraction. These and other parameters are specified in Table 3.

Table 2. Recommendations for Standardized Acquisition of CMR in Cardiac Amyloidosis

The overall imaging protocol as described above will take approximately 45-60 minutes. This table provides a general guide to the steps of a CMR imaging protocol. Some variation between sites may exist. Each of these sequences assesses a unique myocardial characteristic as discussed in the text and Table 3.

ECV, extracellular volume; SCMR, Society for Cardiovascular Magnetic Resonance.

Table 3. Recommendations for Standardized Interpretation and Reporting of CMR for Cardiac Amyloidosis

T2 mapping is currently not part of the standard clinical amyloidosis imaging protocol.

AL, amyloid light chain; ATTR, amyloid transthyretin; CMR, cardiac magnetic resonance imaging; ECV, extracellular volume; EF, ejection fraction; FLC, free light chain; LGE, late gadolinium enhancement; LV, left ventricular; MOLLI, modified Look-Locker inversion recovery; SSFP, steady state free precession; ShMOLLI, Shortened MOdified Look-Locker Inversion Recovery; ULN, upper limit of normal and per Ref. 205 at mid-cavity level ULN for women/men were 7 mm/9 mm (long axis) and 7 mm/8 mm (short axis), respectively.

Protocols for LGE assessment in cardiac amyloidosis are likewise well-defined.¹⁹⁵ Late gadolinium enhancement visualizes the extracellular space expansion that occurs in cardiac amyloidosis. Late gadolinium enhancement imaging depends on “nulling” of normal myocardium in order to detect LGE from slowed gadolinium washout (thus signal enhancement) in abnormal tissue. Initial LGE evaluation of cardiac amyloidosis was challenging due to similar nulling of both the myocardium and blood pool. The more recent PSIR technique, which ensures appropriate nulling, overcomes this limitation.^76,168 There are two phenomena that are unique to the LGE assessment of cardiac amyloidosis. First, there is rapid movement of gadolinium into the ECV due to the high burden of amyloid protein. This results in myocardial nulling prior to or concurrent with the blood pool, which can be identified visually on the TI scout.⁷⁹ Second, there is a global delayed washout of gadolinium from the ECV, resulting in diffuse LGE at time points at which LGE are typically assessed in scar imaging.⁷⁴

A limitation of LGE assessment in cardiac amyloidosis is the requirement for gadolinium administration in the setting of a high coincidence of renal failure in ATTR and AL amyloidosis due to age and multiple myeloma and renal involvement, respectively. Cyclic gadolinium agents need to be administered to decrease risk of nephrogenic systemic fibrosis and other complications. Partially protein-bound contrast agents (gadolinium-BOPTA MultiHance®) should not be used, as neither the ECV technique nor the characteristic amyloid LGE pattern are reliable.¹⁶⁹

In contrast to LGE, T1 and T2 mapping techniques are quantitative tools. Their acquisition has been standardized in a recent consensus statement.^86,196 Per this guideline, T1 map acquisition is recommended in two short-axis slices and a 4-chamber view before and after contrast; T2 map acquisition is recommended in one mid-short-axis slice. Use of local reference ranges and quality control phantoms has been emphasized. A potential concern is the time required for these multiple acquisitions.

T1 mapping can measure the longitudinal magnetization of the myocardium before contrast (native T1). In addition, by measuring T1 before and after contrast and correcting for the blood volume of distribution (1-hematocrit), ECV can be derived (Figure 5). In combination with pre-contrast T1, an approach using one post-contrast T1 has been validated in cardiac amyloidosis¹⁹⁷ and is used by many centers. Other centers perform serial post-contrast measurements, as the fidelity of mapping the myocardial vs blood exchange of contrast may be improved.¹⁹⁸ T1 mapping has advanced from a cumbersome multi breath-hold technique with contrast infusion; current techniques require a single breath-hold and generate an ECV map automatically, in some cases without the need for hematocrit sampling or off-line processing.^199,200

Figure 5. Characteristic appearance of cardiac amyloidosis on CMR. Two patients [upper and lower row, (A) and (B)] with cardiac amyloidosis: similar mass (cine), but significantly different amyloid burden, with the patient at the bottom (B) showing a significant higher amyloid burden (higher native T1, higher ECV, transmural LGE) and lower myocardial resting perfusion (also, after adjusting for ECV expansion). (C) Inversion scout images in two patients, upper row amyloid, lower row non-amyloid control. These images show a distinct pattern of myocardial and blood pool nulling. In the non-amyloid subject, the blood pool nulls prior to myocardium; in contrast, in the subject with cardiac amyloidosis, the myocardium nulls prior to the blood pool.

More recently, CMR with multiparametric mapping has been driving a change in disease understanding: cardiac amyloidosis is not a disease of solely infiltration. T2 mapping, a marker of myocardial edema, has been highlighting other processes in the myocardium—a possible new aspect of the evolution of the myocardial phenotype in cardiac amyloidosis.⁸⁹

Other techniques may also add value: perfusion is profoundly abnormal in cardiac amyloidosis with vasodilator stress revealing marked endo to epicardial gradients (Figure 5).²⁰¹

Cardiovascular magnetic resonance should be performed using standard parameters, as listed in this section.

Cardiac structure, function, and PSIR LGE should be assessed and reported per SCMR guidelines.

Cardiac amyloidosis-specific CMR markers, such as native T1 mapping and ECV, should be assessed and reported when available, as discussed in this document.

An overall reporting on likelihood of cardiac amyloidosis based on imaging findings is recommended (eg, not suggestive, strongly suggestive, or equivocal for cardiac amyloidosis).

Recommendations for standardized radionuclide image acquisition for cardiac amyloidosis using ^99mTc-PYP/DPD/HMDP are provided in Table 4. Images should be acquired early (1 hour) or late (2-3 hours). There is a stepwise approach to interpretation as shown in Table 5. The first step of interpretation is to visually confirm diffuse myocardial radiotracer uptake and differentiate this uptake from residual blood pool activity or overlapping bone using SPECT and planar images.

Table 4. Recommendations for Standardized Acquisition of ^99mTc-PYP/DPD/HMDP for Cardiac Amyloidosis

Adapted from Ref. 207.

ECG, electrocardiogram; PYP, pyrophosphate.

* Parameters for NaI SPECT scanners.

† Anterior and lateral views are obtained at the same time; lateral planar views or SPECT imaging may help separate sternal from myocardial uptake.

Table 5. Recommendations for Interpretation of ^99mTc-PYP/DPD/HMDP for Cardiac Amyloidosis

Adapted from Ref. 207.

AL, amyloid light chain; ATTR, amyloid transthyretin; H/CL, heart/contralateral lung; ROI, region of interest.

* Fig. 6 and 7 refer to figures in the original document.

If myocardial uptake is confirmed visually, there are two approaches to differentiate AL from ATTR cardiac amyloidosis, depending on the tracer used and time between injection and scan acquisition. The 1-hour approach has been validated for ^99mTc-PYP and involves generation of an elliptical/circular region of interest (ROI) over the heart on the anterior planar images with care to avoid sternal overlap and with size adjusted to maximize coverage of the heart without inclusion of adjacent lung. This ROI should be mirrored over the contralateral chest to adjust for background and rib uptake (Figure 6). A semi-quantitative H/CL ratio is calculated as a ratio-of-heart ROI mean counts to contralateral chest ROI mean counts; a ratio of ≥1.5 at 1 hour can accurately differentiate ATTR cardiac amyloidosis from AL cardiac amyloidosis.¹¹³

Figure 6. Characteristic appearance of cardiac amyloidosis on ^99mTc-PYP/DPD/HMDP imaging. Semi-quantitative H/CL Ratio on ^99mTc-PYP Planar Imaging. Anterior planar chest views one hour after injection of ^99mTc-PYP a patient with Grade 3 (A), and Grade 0 (B) ^99mTc-PYP uptake. On the right are the corresponding H/CL (heart/contralateral lung) lung-ratio methodology with measurement of mean counts per pixel for target (heart) and background (contralateral chest). As shown in this figure, the ROIs (region of interest) should be positioned to minimize overlap with sternal or focal rib uptake and maximize coverage of the heart without including adjacent lung.

Alternatively, a 2- or 3-hour approach can be used (as typically performed for ^99mTc-DPD/HMDP) in which a visual grading scale is used (Table 5). Grade 2 or Grade 3 myocardial uptake of ^99mTc-PYP/DPD/HMDP, in the absence of a clonal disorder, is diagnostic of ATTR cardiac amyloidosis (Figure 7). Both planar and SPECT imaging should be reviewed and interpreted using visual and quantitative approaches irrespective of the timing of acquisition.

Figure 7. ^99mTc-PYP/DPD/HMDP. Anterior planar chest images (Top row), SPECT cardiac imaging (Middle row) and planar whole-body imaging (Bottom row). Cardiac uptake is visually compared with surrounding ribs for a visual grading score as described in Table 5. Images with Grade 0, Grade 1, Grade 2, and Grade 3 myocardial uptake of ^99mTc-PYP are shown. (Top panel provided by ASNC Cardiac Amyloidosis Practice Points.²⁰⁹)

SPECT imaging is necessary for studies that show planar myocardial uptake because they can help differentiate myocardial uptake from blood pool or overlying bone uptake. Interpretation should also include comment on focal vs diffuse radiotracer uptake; diffuse uptake is typically consistent with cardiac amyloidosis, while focal uptake may represent early cardiac amyloidosis but has also been described in acute or subacute myocardial infarction. Guidelines for standardized reporting are provided in Table 6.

Table 6. Recommendations for Standardized Reporting of ^99mTc-PYP/DPD/HMDP Imaging for Cardiac Amyloidosis

Adapted from Ref. 207.

AL, amyloid light chain; ATTR, amyloid transthyretin; CMR, cardiovascular magnetic resonance; echo, echocardiography; FLC, free light chain; H/CL, heart-to-contralateral lung ratio.

An H/CL ratio may be falsely low in patients who had suffered a prior large remote myocardial infarction; myocardial uptake of the tracer will be limited to non-infarcted zone. Careful evaluation of these imaging using SPECT and non-planar image display are recommended to visualize regional uptake.

An overview of the imaging acquisition parameters for ¹²³I-mIBG is available in the Appendix. Sources of variability in late HMR include non-homogeneity in ¹²³I-mIBG imaging acquisition; differing gamma camera systems; and low- vs medium-energy collimators.^{131,186,202,203} Recommendations for the reporting of ¹²³I-mIBG are provided in the Appendix and are predominately based on the HMR and washout-rate quantification. As with ^99mTc-PYP/DPD/HMDP, SPECT imaging is of value in addition to planar imaging to evaluate regional cardiac sympathetic innervation abnormalities. The majority of patients (in both AL and ATTR cardiac amyloidosis) with low HMR show reduced tracer accumulation in the inferolateral segments.^{97,98,130,135,137,139} This, however, is not a finding specific to cardiac amyloidosis; reduced radiotracer uptake in the inferolateral myocardial wall is also reported in healthy control subjects due to physiological over projection of ¹²³I-mIBG accumulation of the liver into this region.²⁰⁴ Also, this technique should be avoided in patients with suspected cardiac amyloidosis and prior myocardial infarction.

^99mTc-PYP/DPD/HMDP and ¹²³I-mIBG imaging should be performed using standard protocols as discussed in this section.

SPECT imaging is useful particularly in positive or equivocal cases to differentiate myocardial from blood pool signal and to describe regional heterogeneity.

Visual and semi-quantitative interpretation of ^99mTc-PYP/DPD/HMDP planar and SPECT images should be employed to evaluate heart-to-bone ratio and/or H/CL lung ratio. The HMR is used to interpret ¹²³I-mIBG images.

An overall reporting on likelihood of amyloidosis based on imaging findings is recommended (eg, not suggestive, strongly suggestive, or equivocal for cardiac amyloidosis and for extra-cardiac findings).

The field of imaging in cardiac amyloidosis is expanding rapidly and more research is needed in several key areas.

Early detection with imaging remains an unmet need in cardiac amyloidosis, and techniques that identify disease at an earlier stage are needed. ^99mTc PYP/DPD/HMDP have the potential for early detection of ATTR cardiac amyloidosis prior to echocardiography and CMR. This needs to be further validated.

Molecular imaging techniques, including amyloid binding PET radionuclide tracers and ECV by CMR are particularly well suited to detect early disease. Further studies are needed.

Early detection of cardiac amyloidosis could allow targeted therapy prior to symptom onset and improve clinical outcomes. This needs to be studied further.

Methods for quantitative assessment of systemic and cardiac burden of amyloidosis are needed. ECV assessment by CMR, and ¹⁸F-labelled PET tracers have the potential to provide accurate quantification but require additional evaluation and more widespread dissemination of technology and broader clinical use to reach their full potential.

Precise detection of changes in the burden of cardiac amyloidosis using imaging can allow evaluation of the efficacy of emerging novel therapies aimed at stabilization and even resorption of amyloid fibrils.

Advanced echocardiography, including 3D echocardiographic strain, dynamic echocardiography, left atrial mechanics, and automated, machine learning-based methods over standard approaches are being investigated.

Prospective studies evaluating the incremental diagnostic and prognostic value of non-invasive imaging techniques, including advanced echocardiographic methods, ^99mTc PYP/DPD/HMDP, ¹²³I-mIBG, and CMR should be undertaken. The incremental value of imaging markers over clinical and laboratory markers needs to be studied further.

The majority of existing literature arises from small, single-center studies of highly selected patients. Multicenter studies, including larger patient cohorts and standardized imaging methods, are needed to advance the evaluation and management of cardiac amyloidosis. In particular, large prospective studies are needed to validate the clinical utility of cardiac imaging in assessing the response to therapy and predicting clinical outcome.

The purpose of Part 1 of this consensus statement has been to establish the available diagnostic and prognostic literature for imaging in cardiac amyloidosis and provide comprehensive expert recommendations based on this evidence and expert opinion regarding the role of imaging in cardiac amyloidosis, including standardized image acquisition, interpretation, and reporting. We hope that use of these consensus recommendations on standardized imaging techniques will improve patient care and outcomes. We also hope we have identified gaps in the literature that can spur relevant research to broaden our understanding of this complex disease and support guideline development.

A summary of literature that supports the recommendations provided in this consensus statement on the prognostic value of echocardiography (Table 7); diagnostic and prognostic value of CMR (Tables 8 and 9), and diagnostic and prognostic value of radionuclide imaging with ^99mTc- PYP/DPD/HMDP (Tables 10, 11, 12, 13, and 14) in the evaluation of cardiac amyloidosis are provided in the Appendix. The diagnostic value, prognostic value, standardized image acquisition and reporting of ¹²³I-mIBG in cardiac amyloidosis are provided in Tables 15, 16, 17, and 18.

Table 7. Key Literature Summarizing the Utility of Echocardiography for Risk Assessment in Cardiac Amyloidosis

AL, light-chain amyloidosis; ATTR, transthyretin amyloidosis; ATTRv, amyloidosis due to transthyretin gene mutations, ATTRwt, amyloidosis due to deposition of wild-type transthyretin (senile amyloidosis); AAPSE, anterior aortic plane systolic excursion; CV-1B, cyclic variation of integrated back scatter; CHF, congestive heart failure; GLS, global longitudinal strain; LV, left ventricular; LSsys, longitudinal strain; GLSGE, global longitudinal strain; MCF, myocardial contraction fraction; MAPSE, mitral annular plane systolic excursion; n.a., not available; N number; RV, right ventricular; RRSR, relative regional strain ratio; TAPSE, tricuspid annular plane systolic excursion.

Table 8. Key Literature Summarizing the Diagnostic Value of CMR for Cardiac Amyloidosis

AL, light chain amyloidosis; ATTR, transthyretin amyloidosis; CMR, cardiac magnetic resonance imaging; DCM, dilated cardiomyopathy; ECG, electrocardiogram; HHD, hypertensive heart disease; LGE, late gadolinium enhancement; LV, left ventricular; NA, not applicable; SD, standard deviation; ShMOLLI, Shortened MOdified Look-Locker Inversion recovery; TTE, transthoracic echocardiography; TTR, transthyretin.

Table 9. Key Literature Summarizing the Utility of CMR for Risk Assessment in Cardiac Amyloidosis

AL, light-chain amyloidosis; ATTR, transthyretin-related amyloidosis; LGE, late gadolinium enhancement; PSIR, phase sensitive inversion recovery; LL, look locker; IR, inversion recovery; GE, gradient echo.

* OR from Raina et al.¹⁷⁰

Table 10. Key Literature Summarizing the Diagnostic Value of ^99mTc-PYP Radionuclide Imaging for Cardiac Amyloidosis

AL, light chain amyloidosis; ATTR, transthyretin amyloidosis; AA, Apo serum amyloid A; CHF, congestive heart failure; CL, contralateral; DPD, 3,3-diphosphono-1,2- propanodicarboxylic acid; HMDP, hydroxymethylene diphosphonate; H, heart; H/CL, heart/contralateral lung; LVH, left ventricular hypertrophy; FAP, familial amyloid polyneuropathy; PYP, ^99mTc pyrophosphate; DPD, ^99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid (DPD); HMDP, ^99mTc-hydroxymethylene diphosphonate; MDP, ^99mTc-methylene diphosphonate.

Table 11. Key Literature Summarizing the Diagnostic Value of ^99mTc-DPD/HMDP Radionuclide Imaging for Cardiac Amyloidosis

AL, light chain amyloidosis; ATTRv, transthyretin amyloidosis mutant; ATTRwt, transthyretin amyloidosis wild type; AF, atrial fibrillation; CMR, cardiac magnetic resonance imaging; DPD, 3,3-diphosphono-1,2-propanodicarboxylic acid; FPA, fat pad aspiration; IHC, immunohistochemistry; HMDP, hydroxymethylene diphosphonate; HCM, hyper- trophic cardiomyopathy; HR, heart retention; H, heart; N, number; TTR, transthyretin; WBR, whole-body retention; WB, whole body.

Table 12. Key Literature Summarizing the Role of Radionuclide Imaging in Screening for Cardiac Amyloidosis in Selected Settings

AL, light chain amyloidosis, ATTRv, transthyretin amyloidosis mutant; ATTRwt, transthyretin amyloidosis wild type; AVR, aortic valve replacement; AV, atrioventricular; CAD, coronary artery disease; AS, aortic stenosis; EMB, endomyocardial biopsy; eval, evaluated; FAP, familial amyloid polyneuropathy; IHC, immunohistochemistry; HR,, heart retention; H, heart; HFpEF, heart failure with preserved ejection fraction; HMDP, hydroxymethylene diphosphonate; DPD, 3,3-diphosphono-1,2-propanodicarboxylic acid; PYP, pyrophosphate; LV, left ventricle; LVEF, left ventricular ejection fraction; MS, mass spectrometry; N, number; Pts, patients; RV, right ventricle; SPECT, single photon emission computed tomography; TTR, transthyretin; WBR, whole-body ratio; WB, whole body.

* red flags=increased thickness of AV valves, interatrial septum or RV free wall, pericardial effusion, myocardial granular sparkling.

Table 13. Key Literature Summarizing the Prognostic Value of ^99mTc-DPD/HMDP Radionuclide Imaging for Cardiac Amyloidosis

AA, amyloid A amyloidosis; AF disease, Anderson–Fabry disease; AL, light-chain amyloidosis; ATTR, transthyretin-related amyloidosis; ATTRv, amyloidosis due to transthyretin gene mutations; ATTRwt, amyloidosis due to deposition of wild-type transthyretin (senile amyloidosis); AV, atrio-ventricular; AVR, aortic valve replacement; CMP cardiomyopathies; echo, echocardiogram; FPA, fat pad aspiration; GFR, glomerular filtration rate; HR, heart retention; H/WB, heart/whole body; HCM, hypertrophic cardiomyopathy; HFpEF, heart failure with preserved ejection fraction; IHC, immunohistochemistry; LVH, left ventricle hypertrophy; LVEF, left ventricular ejection fraction; LVM/V, left ventricle mass/volume ratio; mBMI, modified body mass index; MS, mass spectrometry; NYHA, New York Heart Association; MACE, major adverse cardiac events; RV, right ventricle; SBP, systolic blood pressure; TAVR, transcatheter aortic valve replacement; WBR, whole-body retention.

Table 14. Key Literature Summarizing the Utility of ^99mTc-PYP Radionuclide Imaging for Risk Assessment in Cardiac Amyloidosis

AS, aortic stenosis; ATTR, transthyretin amyloidosis; PYP, pyrophosphate; SPECT, single photon emission computed tomography.

Table 15. Key Literature Summarizing the Diagnostic Value of ¹²³I-MIBG Radionuclide Imaging for Cardiac Amyloidosis

AA, amyloid A; AL, amyloid light chain; ATTRv, hereditary transthyretin amyloidosis; FU, followup; HMR, heart-to-mediastinal ratio; HR, hazard ratio; Pt, patient.

Table 16. Key Literature Summarizing the Utility of ¹²³I-MIBG Radionuclide Imaging for Risk Assessment in Cardiac Amyloidosis

CI, confidence interval; HR, hazard ratio; H/M, heart to mediastinal ratio.

Table 17. Recommendations for Standardized Acquisition of ¹²³I-MIBG for Cardiac Amyloidosis

For details on ¹²³I-MIBG imaging, readers are referred to Ref. 235.

Table 18. Recommendations for ¹²³I-MIBG Imaging Reporting

H, heart; LEHR, low energy high resolution; M, mediastinum; ME, medium energy; LV, left ventricular; SPECT, single photon emission computed tomography; WR, wash out rate.

Tables 7 through 18 are located after the References section, beginning on page 31.

There are 2 primary reasons for an addendum. First, since the original publication of this article,²³⁶ the Acknowledgments now include Dr. Richard Cheng and Dr. Roy John, who critically reviewed the article but were inadvertently not listed as reviewers.

Second, since the original publication²³⁶ and introduction of approved therapies for transthyretin cardiac amyloidosis (ATTR-CA), the clinical use of bone tracer cardiac scintigraphy has been extended to populations with lower prevalence of ATTR-CA. Numerous observations have raised concerns about (1) incorrect diagnosis of ATTR-CA based on ^99mTc-pyrophosphate (PYP) planar imaging and heart-to-contralateral lung (H/CL) ratio without confirmation of diffuse myocardial uptake on SPECT imaging at some sites; (2) excess blood pool activity on the 1-hour planar and SPECT images being interpreted as positive scans; and (3) missed diagnosis of light chain (AL) amyloidosis, as serum-free light chain studies and serum and urine immunofixation electrophoresis studies may not be recommended in the ^99mTc-PYP/-3,3-diphosphono-1,2-propanodicarboxylic acid/hydroxymethylene diphosphonate (^99mTc-PYP/DPD/HMDP) report. Incorrect diagnosis leads to inappropriate therapy and worse patient outcomes. SPECT and planar imaging performed at 3 hour maximize specificity.^114,237,238 Additionally, technical parameters have been updated.

This addendum clarifies the protocols, interpretation, and reporting of ^99mTc-PYP imaging:

1. Acquisition (Table 4):

a. The time between injection of ^99mTc-PYP and scan is revised: 2- or 3-hour imaging is recommended, and 1-hour imaging is optional (Table 4). If excess blood pool activity is noted, 3-hour imaging is recommended. The timing between injection and scanning is now consistent for ^99mTc-PYP, -DPD, and -HMDP. We recognize some experienced centers that have become proficient at 1-hour scanning; the recommendation for 2- or 3-hour imaging is particularly important for centers starting new Tc-PYP programs.

b. SPECT imaging is required in all studies (irrespective of time between injection and scan) to highlight the importance of directly visualizing tracer uptake in the myocardium.

c. 1-hour planar-only imaging is not recommended.

d. Emerging literature suggests that cadmium zinc telluride (CZT) SPECT can also be used for ^99mTc-PYP/DPD/HMDP imaging.^239,240

2. Interpretation (Table 5):

a. Planar imaging and H/CL ratio alone are insufficient for diagnosis of ATTR cardiac amyloidosis. SPECT imaging is necessary to identify myocardial uptake of ^99mTc-PYP/DPD/HMDP.

b. Repeat imaging is recommended at 3 hours if excess blood pool activity is noted.

c. The steps in Table 5 clarify that visual grading on planar and SPECT imaging is the primary method for diagnosis of ATTR cardiac amyloidosis.

d. Recommendations are clarified for ease of interpretation.

3. Reporting (Table 6):

a. Diffuse myocardial uptake should be visualized to report a positive scan.

b. The criterion for H/CL ratio >1.5 as strongly positive has been removed (consistent with diagnostic criteria listed in the “ASNC/AHA/ASE/EANM/HFSA/ISA/SCMR/SNMMI Expert Consensus Recommendations for Multimodality Imaging in Cardiac Amyloidosis: Part 2 of 2—Diagnostic Criteria and Appropriate Utilization,”²⁴¹ where H/CL ratio was not listed).

c. Conclusions have been clarified.

Tables 4, 5, and 6 reflect these clarifications.

Abbreviations

Amyloid immunoglobulin light chain

Amyloid transthyretin

^99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid

Extracellular volume

Ejection fraction

Hydroxymethylenediphosphonate

Late gadolinium enhancement

Left ventricular

Pyrophosphate

^99mTechnetium

We would like to thank the reviewers of this document for their input, which has significantly improved the quality of this document, including Renée P. Bullock-Palmer, MD, FACC, FASNC, FASE, FSCCT; Dennis A. Calnon, MD, FASNC; Richard Cheng, MD; Marcelo F. Di Carli, MD; Martha Grogan, MD; Phillip Hawkins, PhD, FMedSci; Wael A. Jaber, MD, FACC, FAHA; Roy John, MD; Prem Soman, MD, FASNC; James E. Udelson, MD, FACC; Ashutosh D. Wechalekar, DM, MRCP, FRCPath.

Disclosures

All other contributors have nothing relevant to disclose.

The American Heart Association requests that this document be cited as follows: Dorbala S, Ando Y, Bokhari S, Dispenzieri A, Falk RH, Ferrari VA, Fontana M, Gheysens O, Gillmore JD, Glaudemans AWJM, Hanna MA, Hazenberg BPC, Kristen AV, Kwong RY, Maurer MS, Merlini G, Miller EJ, Moon JC, Murthy VL, Quarta CC, Rapezzi C, Ruberg FL, Shah SJ, Slart RHJA, Verberne HJ, Bourque JM. ASNC/AHA/ASE/EANM/HFSA/ISA/SCMR/SNMMI expert consensus recommendations for multimodality imaging in cardiac amyloidosis: part 1 of 2—evidence base and standardized methods of imaging. Circ Cardiovasc Imaging. 2021;14:e000029. DOI: 10.1161/HCI.0000000000000029

This document was approved for publication by the governing body of the American Society of Nuclear Cardiology (ASNC) and was endorsed by the American College of Cardiology (ACC), the American Heart Association (AHA), American Society of Echocardiography (ASE), the European Association of Nuclear Medicine (EANM), the Heart Failure Society of America (HFSA), the International Society of Amyloidosis (ISA), the Society for Cardiovascular Magnetic Resonance (SCMR), and the Society of Nuclear Medicine and Molecular Imaging (SNMMI). This document was approved by the American Heart Association Science Advisory and Coordinating Committee on July 8, 2019.

This article has been copublished in the Journal of Nuclear Cardiology and the Journal of Cardiac Failure.

“Part 2—Diagnostic Criteria and Appropriate Utilization” is available at: https://www.ahajournals.org/doi/10.1161/HCI.0000000000000030

A copy of the document is available at https://professional.heart.org/statements by using either “Search for Guidelines & Statements” or the “Browse by Topic” area. To purchase additional reprints, call 215-356-2721 or email Meredith.[email protected]com.

The expert peer review of AHA-commissioned documents (eg, scientific statements, clinical practice guidelines, systematic reviews) is conducted by the AHA Office of Science Operations. For more on AHA statements and guidelines development, visit https://professional.heart.org/statements. Select the “Guidelines & Statements” drop-down menu, then click “Publication Development.”

Permissions: Multiple copies, modification, alteration, enhancement, and/or distribution of this document are not permitted without the express permission of the American Heart Association. Instructions for obtaining permission are located at https://www.heart.org/permissions. A link to the “Copyright Permissions Request Form” appears in the second paragraph (https://www.heart.org/en/about-us/statements-and-policies/copyright-request-form).