The amyloidoses are a spectrum of disorders characterized by tissue deposition of various types of misfolded proteins. In light-chain (AL) amyloidosis, the most common subtype in the West, the culprit protein is the immunoglobulin light chain (either κ or λ), produced by a proliferating clone of plasma cells. Amyloid deposition in the heart, kidneys, liver and gastrointestinal tract, and peripheral nerves may lead to organ failure. Cardiac involvement with the AL subtype may have more grave functional manifestations compared with other types. This is attributed to direct toxicity of the circulating immunoglobulin light chains and/or microvascular myocardial ischemia caused by the perivascular amyloid deposits. Untreated, median survival is 12 months overall, 4 months if the patient has heart failure. There is no available treatment to reverse myocardial infiltration. Current treatment strategies aim to stop excessive production of light chains by suppressing neoplastic plasma cells. The most aggressive of these is high-dose chemotherapy, followed by autologous hematopoietic stem cell transplantation, an option associated with high treatment-related mortality in the sickest patients but ironically one that provides the best long-term results in those who can tolerate it.

In light of the high mortality associated with this disease, knowing which clinical, laboratory, and imaging markers identify high-risk patients is critical. Advanced heart failure is consistently the most important prognostic marker. Other variables found to be independently predictive of survival include hepatomegaly, concomitant multiple myeloma, elevated alkaline phosphatase, troponin T > 0.035 μg/L, troponin I > 0.1 μg/L, N-terminal pro–brain natriuretic peptide > 332 ng/L, and circulating free light chains > 152 mg/L.

In echocardiography, we have moved away from looking for the classic granular sparkling appearance of the myocardium because of the subjectivity of this finding as well as the advent of harmonic imaging, which may make even normal myocardium attain a sparkling appearance. The common findings of cardiac involvement include ventricular thickening, reduced cavity size, presence of diastolic dysfunction usually (but not always) with preserved systolic function, and concomitant pericardial effusion. More recently, advances in strain and 3-dimensional echocardiography have demonstrated abnormalities in myocardial strain and synchrony.

So in 2010, what is the role of echocardiography in AL amyloidosis? The present study by Bellavia et al points to the important and additive value of echocardiography in medium-term prognostication in patients with AL amyloidosis. In a large series of 249 patients with biopsy-proven AL amyloidosis (with 196 patients constituting the final model) followed for a median 18 months, the authors again found that heart failure class is the most important predictor of all-cause mortality, with age, pleural effusion, brain natriuretic peptide level, left ventricular ejection time, and peak longitudinal systolic strain of the basal anteroseptum having additive prognostic value. Using a simplified scoring system that may have practical clinical utility, especially with further validation, the investigators were able to stratify 4 groups on the basis of risk, including a high-risk group with >50% 1-year mortality.

The investigation is a relatively large study that allowed testing the independent prognostic value of multiple clinical, laboratory, and imaging markers. As a result, troponin, previously described as an independent marker, did not qualify in the final model. Similarly, prior echocardiographic findings that were associated with mortality, including ventricular thickness, mitral inflow deceleration time, and the ratio of early to late mitral inflow (E/A), were not found to be independent predictors. This may be due to the strong association of these variables with heart failure class or the other independent markers such as brain natriuretic peptide. In an interesting finding, the echocardiographic indices with additive prognostic value to heart failure class come from something old (ejection time) and something newer (Doppler-based strain). Left ventricular ejection time is an old index of global systolic performance that is rarely used nowadays. Yet it is a simple measurement obtained during routine echocardiographic examination. Our group similarly reported that left ventricular ejection time was a strong (and the only echocardiographic) variable predictive of mortality in patients with AL amyloidosis followed for a median of 29 months, with sensitivity and specificity of 73% and 90% in predicting 1-year cardiac mortality. Myocardial dysfunction leads to a prolonged pre-ejection period and reduced ejection time. Furthermore, amyloid-infiltrated myocardium is prevented from increasing end-diastolic volume, which also leads to reduced ejection time. Abnormal ventriculoarterial coupling is also associated with reduced ejection time. In addition to ejection time, Bellavia et al validated for the first time the prognostic use of Doppler myocardial imaging in amyloidosis. Strain echocardiography, a method for measuring myocardial mechanics directly, is superior to traditional cavity-based measures of systolic function such as ejection fraction or fractional shortening in the early detection of cardiomyopathy. Indeed, ejection fraction has not been found to be an independent prognostic factor in AL amyloidosis.

There are several important caveats the reader needs to discern from the study of Bellavia et al. The most important is that the investigators were not able to tease out the influence of therapy, specifically the use of high-dose chemotherapy and stem cell transplantation, on the survival outcome. As mentioned previously, high treatment-related mortality has prompted the adoption of a risk-adapted approach that offers this aggressive treatment only to low-risk patients and excludes patients with advanced heart failure or ≥3 organ involvement. Indeed, eligibility for stem cell transplantation by itself is associated with favorable survival. Therefore, the collinearity between favorable prognostic variables and the use of aggressive treatment makes it difficult to determine the independent prognostic role of treatment. A practical implication of this caveat is that the current approach of identifying low-risk patients (who likely have none or few of the prognostic variables) and treating them using the risk-adapted approach is associated with lower mortality, with the corollary true for the other extreme. A second caveat is that the proposed model reflects data from an advanced tertiary center that may not be applicable to primary care clinics and other hospitals. It is possible that patients presenting to their primary physicians may be sicker, with only the lucky ones surviving long enough to be referred to large tertiary care institutions such as the Mayo Clinic. It is reassuring that the model appears to hold true even in subjects recently (<3 months) diagnosed to have the disease, but widespread adoption of the proposed clinical risk stratification still requires further validation. A third caveat is the lack of magnetic resonance imaging data in the model, a modality that is gaining favor in light of its diagnostic and prognostic utility for this disease. A recent study demonstrated that diastolic echocardiographic parameters did not have additive value to late gadolinium enhancement on magnetic resonance imaging in the diagnosis or 1-year prognosis of patients with AL amyloidosis, although that study did not include the 2 echocardiographic parameters found by Bellavia et al to be prognostic. A study on a large number of subjects needs to be done to determine whether echocardiographic or magnetic resonance imaging findings have additive value compared with each other and whether either technique might provide other prognostic markers.

An important finding that should not be missed by readers is that even at an advanced tertiary center, AL amyloidosis is still associated with high mortality. With a median follow-up period of 18 months, overall mortality was 30%. A recent report also showed high mortality in a cohort of patients with AL amyloid with more advanced stages of the disease. This sobering reality should be a call to arms in our community to muster the focus, attention, resources, and multidisciplinary and multi-institutional collaborative efforts needed to fill the yawning knowledge gap we have regarding this disease in terms of basic underlying mechanisms, earlier diagnostic testing, and direct and definitive treatment options. Although identifying prognostic factors after making the diagnosis of AL amyloidosis is extremely important, as Bellavia et al have done, the high mortality in their patients tells us that we need to improve at identifying these patients in the early stage of the disease to favorably affect their outcomes. This is a task especially suited for the echocardiography community. We are likely to be the first to see ventricular thickening, diastolic dysfunction, and pericardial effusion on echocardiography; routinely adding the simple additional step of looking at an electrocardiogram to see whether there is associated low voltage would allow us to recommend further investigation leading to the earlier, and potentially lifesaving, diagnosis of AL amyloidosis. As echocardiographers, our challenge is to alert our referring colleagues to the possibility of this diagnosis, to avoid the tragedy that occurs all too commonly with this disease when a patient with AL amyloidosis remains undiagnosed for years.