RESUMO
AIMS: Systemic amyloidosis represents a heterogeneous group of diseases resulting from amyloid fibre deposition. The purpose of this study is to establish a differential diagnosis algorithm targeted towards the two most frequent subtypes of CA. METHODS AND RESULTS: We prospectively included all consecutive patients with ATTR and AL evaluated between 2018 and 2022 in two centres in a score derivation cohort and a different validation sample. All patients had a complete clinical, biomarker, electrocardiographic, and imaging evaluation. Confirmation of the final diagnosis with amyloid typing was performed according to the current international recommendations. The study population included 81 patients divided into two groups: ATTR (group 1, n = 32: 28 variant and 4 wild type) and AL (group 2, n = 49). ATTR patients were younger (50.7 ± 13.9 vs. 60.2 ± 7.3 years, P = 0.0001), and significantly different in terms of NT-proBNP [ATTR: 1472.5 ng/L (97-4218.5) vs. AL 8024 ng/L (3058-14 069) P = 0.001], hs-cTn I [ATTR: 10 ng/L (4-20) vs. AL 78 ng/L (32-240), P = 0.0002], GFR [ATTR 95.4 mL/min (73.8-105.3) vs. AL: 68.4 mL/min (47.8-87.4) P = 0.003]. At similar left ventricular (LV) wall thickness and ejection fraction, the ATTR group had less frequently pericardial effusion (ATTR: 15% vs. AL: 33% P = 0.0027), better LV global longitudinal strain (ATTR: -13.1% ± 3.5 vs. AL: -9.1% ± 4.3 P = 0.04), RV strain (ATTR: -21.9% ± 6.2 vs. AL: -16.8% ± 6 P = 0.03) and better reservoir function of the LA strain (ATTR: 22% ± 12 vs. AL: 13.6% ± 7.8 P = 0.02). Cut-off points were calculated based on the Youden method. We attributed to 2 points for parameters having an AUC > 0.75 (NT-proBNP AUC 0.799; hs-cTnI AUC 0.87) and 1 point for GFR (AUC 0.749) and TTE parameters (GLS AUC 0.666; RV FWS AUC 0.649, LASr AUC 0.643). A score of equal or more than 4 points has been able to differentiate between AL and ATTR (sensitivity 80%, specificity 62%, AUC = 0.798). The differential diagnosis score system was applied to the validation cohort of 52 CA patients showing a sensitivity of 81% with specificity of 77%. CONCLUSIONS: CA is a complex entity and requires extensive testing for a positive diagnosis. This study highlights a series of non-invasive checkpoints, which can be useful in guiding the decision-making process towards a more accurate and rapid differential diagnosis.