DTA

Digital Theses Archive

 

Tesi etd-11122021-230453

Type of thesis
Corsi integrativi
Author
MACCARANA, AGNESE
URN
etd-11122021-230453
Title
A comprehensive echocardiographic score of valve disease in cardiac amyloidosis
Structure
Cl. Sc. Sperimentali - Medicina
Course
SCIENZE MEDICHE - SCIENZE MEDICHE
Committee
Tutor Prof.ssa ANGELONI, DEBORA
Relatore Prof. EMDIN, MICHELE
Membro Prof. RECCHIA, FABIO ANASTASIO
Membro Prof. COCEANI, FLAVIO
Membro Prof. PASSINO, CLAUDIO
Membro Dott. GIANNONI, ALBERTO
Membro Dott.ssa CASIERI, VALENTINA
Membro Prof. LIONETTI, VINCENZO
Keywords
  • AL cardiac amyloidosis
  • ATTR cardiac amyloidosis
  • cardiac amyloidosis
  • echocardiographic score
  • echocardiograpy
  • valve disease
Exam session start date
20/12/2021;
Availability
parziale
Abstract
Background: Cardiac amyloidosis (CA) is caused by the myocardial accumulation of amyloid fibrils that in the vast majority of cases are composed of light-chains (AL-CA) or transthyretin (ATTR-CA). The most known manifestations of CA are left ventricular pseudohypertrophy and diastolic dysfunction, often with a restrictive pattern. The consequences of amyloid deposition in the left atrium, the right heart sections and in the valves have received limited attention so far. In this study we identified the echocardiographic features of valvular involvement, and we created a score that provides a synthetic estimate of the global burden of valve disease. <br>Methods: From the database of patients undergoing a diagnostic workup for suspected CA at the Fondazione Toscana Gabriele Monasterio, Pisa, Italy, from 2010 to 2020 (n=423) we randomly selected 2 samples of patients with ATTR-CA or AL-CA (n=20 each), and we matched them by age and sex with 2 samples of patients with CA excluded. For each patient we re-evaluated the baseline echocardiographic exam, considering the mitral valve, the tricuspid valve and the aortic valve. We analyzed 31 items related to the structure and function of valve leaflets or cusps, the anulus and subvalvular apparatus of the atrioventricular valves. We assigned score points to each item and we created a score system named Amyloid VAlve (AVA). <br>Results: Patients with ATTR-CA displayed more often a shortened or hidden and retracted posterior mitral valve leaflet than patients with AL-CA, as well as thickened mitral chordae tendineae and a stenotic aortic valve. The only significant difference between ATTR-CA and their matched controls was represented by less frequent calcification of the posterior mitral valve leaflet. Furthermore, patients with AL-CA displayed more often a short or hidden posterior mitral valve leaflet than their matched controls. Overall, patients with ATTR-CA had higher mitral score values than those with AL-CA or than ATTR-CA controls. Global AVA score values were 15.8 (13.6-17.4) in patients with ATTR-CA, 11.0 (9.3-14.9) in those with AL-CA, 12.8 (11.1-14.4) in ATTR-CA controls, and 11.0 (9.1-13.0) in AL-CA controls (p values: 0.004 for ATTR-CA vs. AL-CA, 0.009 for ATTR-CA vs. matched controls, and 0.461 for AL-CA vs. matched controls). Given the higher AVA score in patients with ATTR-CA than those with AL-CA, we investigated the possibility to discriminate between these two conditions based on score values. The area under the curve was 0.765, and the best cut-off was 14 (sensitivity 75%, specificity 70%, positive predictive value 71%, negative predictive value 74%). Among all clinical, laboratory and echocardiographic characteristics, only age, interventricular septum and posterior wall thickness, left ventricular mass index, and continuous score values or values higher than or equal to 14 emerged as univariable predictors of ATTR-CA. None of these variables independently predicted ATTR-CA. The AVA score values had an area under the curve (AUC) of 0.739 to distinguish ATTR-CA from matched controls; the best cut-off was again 14, with the same sensitivity, specificity, PPV and NPV than for the differentiation between ATTR- and AL-CA. Absolute AVA score values and the 14 cut-off were both univariable predictors, while they did not reach independent prognostic significance. <br>Eleven patients with ATTR-CA (55%) died over 2.1 years (1.0-3.7); six patients out of 19 with available data (32%) died for cardiovascular causes. Ten patients out of 18 (55%) were hospitalized for HF over 1.3 years (0.6-2.3). Among patients with AL-CA, 9 died (45%), 7 of whom for cardiovascular causes (35%), over 3.1 years (0.7-7.4); 9 patients out of 16 (56%) were hospitalized for HF over 0.4 years (0.2-1.2). The AVA score was not a good predictor of fatal outcomes in ATTR-CA (AUC values: 0.611 for all-cause death, 0.519 for cardiovascular death), while it was more predictive in AL-CA (AUC values: 0.687 for all-cause death, 0.725 for cardiovascular death). AUC values for HF hospitalization were quite similar in ATTR-CA (0.700) and AL-CA (0.635). <br>Conclusions: The AVA score is the first attempt of global evaluation of echocardiographic valve features in CA. Patients with ATTR-CA have a particularly prominent impairment of mitral valve structure and function, conditioning higher score values. Accordingly, the AVA score is quite effective in discriminating patients with ATTR-CA from AL-CA or matched controls, with a fair diagnostic performance despite the small sample sizes in this study. Conversely, the prognostic value of AVA score values seems more limited. <br>
Files