Prognostic Value of LGE in AL Cardiac Amyloidosis
What is the incremental prognostic value of late gadolinium enhancement (LGE), a marker of amyloid infiltration of the myocardium, on cardiac magnetic resonance imaging (CMR) in patients with amyloid light chain (AL) amyloidosis?
A total of 76 patients with definite histologically proven AL amyloidosis and CMR were retrospectively reviewed, following an earlier cross-sectional paper with outcomes data. Approximately one bhalf (n = 75) of all patients who had AL amyloid and CMR (n = 151) were excluded due to prior myocardial infarction or myocarditis, or stem cell or heart transplantation. LGE was visually scored as “global,” “focal patchy,” or “none.” Serum troponin T and N-terminal pro–B-type natriuretic peptide were measured. Patients with normal levels of both biomarkers were classified as stage I. Patients with either biomarker elevated were classified as stage II and those with both biomarkers elevated were classified as stage III. The primary outcome was death.
Of 76 patients, 32 (42%) had global LGE, 24 (32%) had focal patchy LGE, and 20 (26%) had none. Among the population, there were 40 deaths over 34.4 months. Compared to focal patchy involvement, global enhancement was associated with a 2.9-fold increased risk of death (p < 0.001), while no enhancement was associated with a risk of death only 0.37 times that of focal (p < 0.01). After adjustment for biochemical marker stage, global LGE was associated with increased mortality (hazard ratio, 2.03; p = 0.05).
The authors concluded that global LGE is associated with a doubling of mortality in patients with AL amyloidosis.
Cardiac MRI appears to be prognostically useful in patients with AL amyloidosis. This adds to a growing body of literature that has demonstrated the value of MRI techniques in amyloidosis. This study specifically demonstrates that global LGE has particular prognostic importance beyond traditional biomarker staging of cardiac involvement in amyloidosis. As a retrospective, observational study, several caveats must be noted. First, it is likely that there is some degree of referral bias, as patients with low likelihood of cardiac involvement based on clinical history and symptoms may have been less likely to be referred. Second, nearly one half of all potential patients were excluded, potentially limiting generalizability. Nonetheless, this important result demonstrates the potential clinical application of this technique.
< Back to Listings