Performance of Clinical Scoring Systems in the Diagnosis of Transthyretin Amyloid Cardiomyopathy in a Diverse Patient Cohort.

Background: Two diagnostic clinical scoring systems, the ATTR-CM Score and the T-AMYLO Score, have been proposed but not validated in diverse populations despite Black race being an important risk factor for transthyretin amyloid cardiomyopathy (ATTR-CM). The aim of this study was to evaluate their performance in diagnosing ATTR-CM in a diverse patient cohort.

Methods: This retrospective single-center study analyzed patients who underwent a 99mTc-pyrophosphate single-photon emission computed tomography scan (Tc-PYP) for workup of suspected ATTR-CM. ATTR-CM was considered present in those exhibiting Perugini scores of 2 or 3, confirmed by myocardial radiotracer uptake via single-photon emission computed tomography. The diagnostic performance of a multivariate regression model and the two scoring systems was tested against Tc-PYP as the gold standard.

Results: Our cohort included 476 patients, of which 308 (65%) were non-Hispanic Black, 93 (20%) were Hispanic, and 215 (45%) were female. A total of 164 (34%) had a positive Tc-PYP result. Age ≥74 years, male sex, history of carpal tunnel, left ventricular ejection fraction <55%, posterior wall thickness over 12 mm, and relative wall thickness over 0.57 were independent predictors of positive Tc-PYP results in our cohort, and hemoglobin level <10 mg/dL, glomerular filtration rate <30 mL/min/1.73 m², and coronary artery disease were independent predictors of negative Tc-PYP. The multivariate model had an area under the curve (AUC) of 0.92 (95% CI 0.90-0.95). The ATTR-CM Score (AUC, 0.86; 95% CI 0.83-0.90) had better diagnostic accuracy than the T-AMYLO Score (AUC, 0.75; 95% CI 0.71-0.80; P < .001).

Conclusions: Two simple clinical scoring systems, derived to identify patients at high risk of having ATTR-CM necessitating further diagnostic evaluation, showed good predictive accuracy in our diverse patient cohort. The ATTR-CM Score was superior to the T-AMYLO Score in our cohort.