External validation of three multiparametric echocardiographic scoring systems for AL-CM and ATTR-CM in a single amyloidosis referral center
J Di Palma-Grisi, Y Weiss, S Hazaveh, C Faris, B Osei-Frimpong, A Cardiero, H Parmar, M Vandyck-AcquahAbstract
Introduction
Cardiac amyloidosis is a form of infiltrative cardiomyopathy and an increasingly prevalent driver of heart failure with preserved ejection fraction, typically diagnosed by cardiac MRI (cMRI) or endomyocardial biopsy (EMB) [1]. Multiparametric echocardiographic scores for light-chain amyloid cardiomyopathy (AL-CM) and transthyretin amyloid cardiomyopathy (ATTR-CM) have proliferated [2-3]. We assessed the AL, Increased Wall Thickness (IWT) and ATTR-CM scores in a cohort of patients seen at our AL amyloidosis referral center.
Methods
Retrospective chart review was conducted under an IRB-approved protocol for initial referrals. 437 patients with AL amyloidosis were referred between 1 January 2010 and 31 August 2025. Of these, 98 had transthoracic echocardiography (TTE) with the parameters (E/E’, RWT, GLS, TAPSE) to calculate an AL score. An additional 188 patients without AL amyloidosis underwent workup for ATTR-CM; 84 had IVSd or PWTd ≥ 12mm, 78 had TTEs with the parameters (E/E’, RWT, GLS, TAPSE, SAB) to calculate an IWT score and 109 had TTEs with the parameters (EF, PWTd, RWT) to calculate an ATTR-CM score.
Results
In the AL amyloidosis group (n=98), the AUC was 0.72. At predefined cutoffs, there was a sensitivity of 97% at AL ≥ 1 and a specificity of 85% at AL ≥ 5. In the cMRI or EMB-confirmed AL-CM group (n=65), median age was 65, median NT-proBNP was 1367 pg/mL (IQR 553 pg/mL - 5828 pg/mL), and median hsTnT was 38 ng/L (IQR 16 ng/L - 65 ng/L). In the negative group (n=33), median age was 70, median NT-proBNP was 375 pg/mL (IQR 98 pg/mL - 640 pg/mL) and median hsTnT was 15 ng/L, (IQR 7 ng/L - 27 ng/L). AL scores ≥ 3 or NT-proBNP ≥ 332 pg/mL had a sensitivity of 0.93 and a specificity of 0.19. In the suspected ATTR-CM group (n=78), 63 were confirmed to have ATTR-CM by pyrophosphate (PyP) scan (grade 2/3), cMRI or EMB; in the other 15, ATTR-CM was ruled out by PyP scan (grade 0), cMRI or EMB. The IWT score had an AUC of 0.76. At predefined cutoffs, sensitivity was 0.81 at IWT ≥ 2 and specificity was 1.0 at IWT ≥ 8. For the ATTR-CM score, 0-4 points are assigned by age (2 for 60-69, 3 for 70-79, 4 for 80+), 2 for male sex, 1 for EF < 60%, 1 for PWTd ≥ 12mm, 2 for RWT > 0.57 and 1 removed for hypertension. Scores of ≥ 6 are high-risk with sensitivity of 0.73 and specificity of 0.47. With cutoffs of ≥ 2 and ≥ 8, sensitivity was 0.94 at ≥ 2 and specificity was 0.93 at ≥ 8.
Conclusions
Multiparametric scores were reliable in our study, the first of its kind to externally validate the AL and IWT scores in an AL amyloidosis referral center in the United States. The AL score had high sensitivity and specificity at its predefined cutoffs; when combined with NT-proBNP it can be a useful screening tool at a single point (≥ 3). The IWT score was less reliable than the ATTR-CM score, which was validated in a community population, when using comparable cutoffs of ≥ 2 and ≥ 8.Table 1For image description, please refer to the figure legend and surrounding text.Table 2For image description, please refer to the figure legend and surrounding text.