ESC Heart Failure (Oct 2019)

Diagnostic performance of imaging investigations in detecting and differentiating cardiac amyloidosis: a systematic review and meta‐analysis

  • Jack Brownrigg,
  • Massimiliano Lorenzini,
  • Matthew Lumley,
  • Perry Elliott

DOI
https://doi.org/10.1002/ehf2.12511
Journal volume & issue
Vol. 6, no. 5
pp. 1041 – 1051

Abstract

Read online

Abstract Aims The study aims to systematically assess the diagnostic performance of cardiac magnetic resonance (CMR) and nuclear scintigraphy (index tests) for the diagnosis and differentiation of subtypes of cardiac amyloidosis. Methods and results MEDLINE and Embase electronic databases were searched for studies evaluating the diagnostic performance of CMR or nuclear scintigraphy in detecting cardiac amyloidosis and subsequently in differentiating transthyretin amyloidosis (ATTR) from immunoglobulin light‐chain (AL) amyloidosis. In this meta‐analysis, histopathological examination of tissue from endomyocardial biopsy (EMB) or extra‐cardiac organs were reference standards. Pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio were calculated, and a random effects meta‐analysis was used to estimate diagnostic odds ratios. Methodological quality was assessed using a validated instrument. Of the 2947 studies identified, 27 met the criteria for inclusion. Sensitivity and specificity of CMR in diagnosing cardiac amyloidosis was 85.7% and 92.0% against EMB reference and 78.9% and 93.9% with any organ histology reference. Corresponding sensitivity and specificity of nuclear scintigraphy was 88.4% and 87.2% against EMB reference and 82.0% and 98.8% with histology from any organ. CMR was unable to reliably differentiate ATTR from AL amyloidosis (sensitivity 28.1–99.0% and specificity 11.0–60.0%). Sensitivity and specificity of nuclear scintigraphy in the differentiation of ATTR from AL amyloidosis ranged from 90.9% to 91.5% and from 88.6% to 97.1%. Pooled negative likelihood ratio and positive likelihood ratio for scintigraphy in this setting were 0.1 and 8, with EMB reference standard. Study quality assessed by QUADAS‐2 was generally poor with evidence of bias. Conclusions Cardiac magnetic resonance is a useful test for diagnosing cardiac amyloidosis but is not reliable in further classifying the disease. Nuclear scintigraphy offers strong diagnostic performance in both the detection of cardiac amyloidosis and differentiating ATTR from AL amyloidosis. Our findings support the use of both imaging modalities in a non‐invasive diagnostic algorithm that also tests for the presence of monoclonal protein.

Keywords