Role of photon-counting computed tomography in patients with suspected cardiac amyloidosis
Y Ferrari Chen, V Castiglione, A Aimo, A Campora, D S Zai, P Morfino, F Mori, C De Gori, M Emdin, A Clemente, G VergaroAbstract
Background
The differentiation between cardiac amyloidosis (CA) and other hypertrophic phenotypes is crucial for the management of heart failure. Photon-counting computed tomography (PCCT) is a novel technology offering high spatial resolution and multi-energy spectral capabilities.
Purpose
This study aimed to evaluate the diagnostic performance and clinical value of PCCT, including extracellular volume (ECV) quantification and late iodine enhancement (LIE), in patients with suspected CA.
Methods
We conducted a prospective single-centre study enrolling 68 consecutive patients with suspected or recently diagnosed CA. All subjects underwent a comprehensive PCCT protocol including coronary angiography and spectral acquisitions at 3, 5, 7, and 10 minutes post-contrast. The final diagnosis was established by a multidisciplinary team according to international guidelines. We assessed the diagnostic accuracy of PCCT-derived ECV at different time points, the impact of acquisition timing on ECV quantification, the agreement between LIE and cardiac magnetic resonance (CMR) late gadolinium enhancement (LGE), and the prevalence of concomitant coronary artery disease.
Results
CA was confirmed in 36 patients (53%), predominantly the transthyretin subtype. The remaining patients were diagnosed with hypertrophic cardiomyopathy, hypertensive heart disease, non-dilated left ventricular cardiomyopathy, dilated cardiomyopathy, or Fabry disease. Scan duration was significantly shorter for PCCT than CMR (median 16 vs 34 minutes, p<0.001). PCCT-derived ECV at 3 minutes yielded the highest diagnostic accuracy (area under the curve 0.745), with an optimal cut-off of 35.9% for differentiating CA from other phenotypes, whereas later acquisitions showed progressively reduced discriminative power. LIE identified typical myocardial infiltration patterns, showing substantial agreement with CMR-LGE (Cohen’s κ = 0.70) and a positive predictive value of 98%. Higher ECV values and diffuse LIE patterns were associated with a more adverse clinical profile, including all-cause mortality and heart failure hospitalisation. Furthermore, the protocol ruled out obstructive coronary artery disease in 83% of the CA cohort, avoiding the need for invasive angiography. Multi-energy imaging also excluded intracardiac thrombus and enabled detection of clinically relevant extracardiac findings.
Conclusions
PCCT is a rapid and comprehensive imaging modality for the evaluation of patients with suspected CA. ECV quantification and LIE provide good diagnostic accuracy, comparable to CMR, while simultaneously assessing coronary and extracardiac pathology. Early post-contrast acquisition appears optimal for ECV calculation, and markers of extensive myocardial infiltration suggest a potential prognostic role. These findings support the integration of PCCT into the diagnostic algorithm, particularly in patients with contraindications to CMR.diagnostic performance and agreement analysismultimodal validation