Diastolic Dysfunction Evaluation in Cardiac Amyloidosis: Beyond Traditional Parameters to Discriminate Between True Heart Failure With Preserved Ejection Fraction and Mimics

Abstract Background Diastolic dysfunction in cardiac amyloidosis (CA), according to current guidelines, can be classified using mitral flow velocities, mitral annular e’ velocity, E/e’ ratio, tricuspid regurgitation jet and left atrial (LA) maximal volume. However, LA minimal volume and the ratio between LA and left ventricular (LV) volumes at end–diastole, known as left atrio–ventricular coupling index (LACI), have recently been proposed to refine prognosis and peak atrial longitudinal strain (PALS) has been demonstrated to better stratify diastolic dysfunction. Aim of the Study To evaluate, in patients with CA: 1) the role of LACI as marker of LV diastolic dysfunction; 2) whether LACI could discriminate cardiomyopathy from normal cardiac function; 3) whether LACI and/or PALS could help discriminate between true heart failure with preserved ejection fraction (HFpEF) and amyloidotic cardiomyopathy (HFpEF–mimic). Methods We retrospectively analyzed 63 patients with CA and compared them with 80 true HFpEF patients and 70 healthy people; echocardiography was performed at the time of the diagnosis and diastolic function was characterized according to current guidelines. Results In patients with CA, LACI significantly increased as diastolic dysfunction worsened (P for trend 0.0001) and significantly correlated with LV filling pressures estimated by E/e’ ratio (beta = 0.65, P 0.0001) and pulmonary artery systolic pressure (beta = 0.63, P 0.0001). LACI, after adjustment for age, gender and body mass index (BMI), was significantly higher in CA 0.62 (95%CI 0.52–0.70) and in HFpEF 0.61 (95%CI 0.55–0.68) than in controls 0.29 (95%CI 0.19–0.39), while the difference between the two diseased populations was not statistically significant; a cut–off value of 0.45 was able to discriminate between controls and diseased patients (sensitivity = 95%, specificity = 94%, area under ROC curve = 0.98). PALS, after adjustment for age, gender and BMI, was significantly lower in CA 11.1 (95%CI 8.4–13.8%) and 8.4 (95%CI 3.5–13.3%) respectively in ATTR and AL than in HFpEF 17.4 (95%CI 15.6–19.3%), and a cut–off value of 9.9% was able to discriminate between them (sensitivity = 89%, specificity = 73%, area under ROC curve = 0.86). Conclusions LACI is a novel index able to discriminate between normal and diseased patients. In CA, it shows good correlation with diastolic dysfunction and LV filling pressures. PALS can be used to discriminate between true HFpEF and CA.