Cardiac Magnetic Resonance-Derived Extracellular Volume Mapping for the Quantification of Hepatic and Splenic Amyloid

Background: Systemic amyloidosis is characterized by amyloid deposition that can involve virtually any organ. Splenic and hepatic amyloidosis occurs in certain types, in some patients but not others, and may influence prognosis and treatment. SAP (serum amyloid P component) scintigraphy is uniquely able to identify and quantify amyloid in the liver and spleen, thus informing clinical management, but it is only available in 2 centers globally. The aims of this study were to examine the potential for extracellular volume (ECV) mapping performed during routine cardiac magnetic resonance to: (1) detect amyloid in the liver and spleen and (2) estimate amyloid load in these sites using SAP scintigraphy as the reference standard.

Methods: Five hundred thirty-three patients referred to the National Amyloidosis Centre, London, between 2015 and 2017 with suspected systemic amyloidosis who underwent SAP scintigraphy and cardiac magnetic resonance with T1 mapping were studied.

Results: The diagnostic performance of ECV to detect splenic and hepatic amyloidosis was high for both organs (liver: area under the curve, -0.917 [95% CI, 0.880-0.954]; liver ECV cutoff, 0.395; sensitivity, 90.7%; specificity, 77.7%; P<0.001; spleen: area under the curve, -0.944 [95% CI, 0.925-0.964]; spleen ECV cutoff, 0.385; sensitivity, 93.6%; specificity, 87.5%; P<0.001). There was good correlation between liver and spleen ECV and amyloid load assessed by SAP scintigraphy (r=0.504, P<0.001; r=0.693, P<0.001, respectively). There was high interobserver agreement for both the liver and spleen (ECV liver intraclass correlation coefficient, 0.991 [95% CI, 0.984-0.995]; P<0.001; ECV spleen intraclass correlation coefficient, 0.995 [95% CI, 0.991-0.997]; P<0.001) with little bias across a wide range of ECV values.

Conclusions: Our study demonstrates that ECV measurements obtained during routine cardiac magnetic resonance scans in patients with suspected amyloidosis can identify and measure the magnitude of amyloid infiltration in the liver and spleen, providing important clues to amyloid type and offering a noninvasive measure of visceral amyloid burden that can help guide and track treatment.