Cardiac amyloidosis is characterised by impaired left ventricular global longitudinal strain with apical sparing, yet its diagnostic accuracy has varied considerably across studies. This large-scale international study, spanning 12 centres across 4 countries, set out to determine the ability of the apical sparing ratio and the most common echocardiographic parameters to differentiate patients with confirmed cardiac amyloidosis from those with clinical or echocardiographic suspicion of the condition but in whom the diagnosis had been ruled out.

Using Us2.ai's automated deep learning platform to analyse echocardiograms from 544 confirmed cardiac amyloidosis patients and 200 clinically similar controls, the findings challenge a widely held diagnostic assumption. Apical sparing pattern, long considered a hallmark feature of cardiac amyloidosis, showed only modest accuracy in distinguishing confirmed cases from patients without the diagnosis. Even at the optimal cut-off, apical sparing was present in one third of controls, highlighting the risks of over-reliance on this single marker and underscoring the need for a more comprehensive, multi-parameter approach to cardiac amyloidosis screening.

 

Key Findings

  • At the optimal cut-off of 1.67, the apical sparing ratio (ASR) achieved only 72% sensitivity and 66% specificity for cardiac amyloidosis, and was present in 32% of controls in whom the diagnosis had been ruled out, challenging its reliability as a standalone marker.
  • ASR had an AUC of 0.74 for differentiating CA from phenotypic controls, compared to 0.77 for relative wall thickness, indicating only modest diagnostic accuracy for either parameter alone.
  • Us2.ai's automated deep learning platform analysed echocardiograms from 544 confirmed CA patients and 200 clinically similar controls across 12 centres in 4 countries, one of the largest and most geographically diverse CA echo studies to date.
  • The findings call for multi-parameter screening strategies that go beyond ASR, as no single echocardiographic marker provides sufficient accuracy to reliably identify or exclude cardiac amyloidosis.

 

 

Cotella, J., Randazzo, M., Maurer, M. S., Helmke, S., Scherrer-Crosbie, M., Soltani, M., Goyal, A., Zareba, K., Cheng, R., Kirkpatrick, J. N., Yogeswaran, V., Kitano, T., Takeuchi, M., Fernandes, F., Hotta, V. T., Campos Vieira, M. L., Elissamburu, P., Ronderos, R., Prado, A., Koutroumpakis, E., … Lang, R. M. (2024). Limitations of Apical Sparing Pattern in Cardiac Amyloidosis: A Multicenter Echocardiographic Study. European Heart Journal. Cardiovascular Imaging, 25(6), 754-761. https://doi.org/10.1093/ehjci/jeae021