Peak atrial longitudinal strain to replace atrial volume in diastolic function grading: associations with long-term outcomes in a community-based sample with normal left ventricular ejection fraction

Abstract Background Diastolic dysfunction (DD) often precedes symptomatic heart failure and is a predictor of adverse outcomes; thus, accurate assessment and grading are crucial for early interventions. Current guidelines recommend combining multiple echocardiographic parameters, including left atrial volume index (LAVi), for diagnosing DD in patients with normal left ventricular ejection fraction (LVEF). Recently, peak atrial longitudinal strain (PALS) emerged as a more sensitive marker of left atrial function, although its added value in the grading of DD and associated adverse outcomes is scarcely explored. Purpose We aimed to investigate the impact of replacing LAVi with PALS on DD grading by its associations with long-term all-cause mortality in a low-risk, community-based sample. Methods We retrospectively identified 1319 volunteers from a population-based screening program (mean age 53.8±14.9 years, 59.1% female) with LVEF>50%. Subjects underwent echocardiography to measure mitral annular early diastolic velocities (e’), the ratio between mitral inflow early diastolic velocity (E) and e' (E/e'), tricuspid regurgitation peak velocity, LAVi, and PALS by speckle tracking. Diastolic function was initially classified according to the 2016 ASE/EACVI guidelines into normal, indeterminate, and DD groups and then reclassified by replacing LAVi>34 ml/m² with a standard PALS cut-off of <24%. The primary endpoint was all-cause mortality. Results During the median follow-up time of 11 years, 135 (10,2%) subjects met the primary endpoint. According to guideline definitions, 59 subjects were diagnosed with DD, 128 were indeterminate, and 1132 showed normal function. Kaplan-Meier analysis (Figure 1) showed significant differences in mortality between normal versus indeterminate and DD groups (log-rank p<0.001), with a three-fold (HR [95% CI]: 3.363 [1.802 - 6.274], and nearly five-fold increased risk (HR [95% CI]: 4.935 [1.919 - 12.690], respectively. Notably, no significant difference was observed between the outcomes of the indeterminate and DD groups. When subjects were reclassified using PALS instead of LAVi, 35 had DD, 75 were indeterminate, and 1209 had normal function. This classification resulted in different outcomes between groups (Figure 2, log-rank p=0.003). Compared to the normal, the indeterminate group exhibited a two-fold increase in mortality risk (HR: 2.277 [95% CI: 1.038-4.997]), while the DD group had a ten-fold increased risk (HR: 10.180 [95% CI: 2.870-36.130]). DD was associated with a nearly five-fold increase in mortality risk (HR: 4.724 [95% CI: 2.177-10.250]) compared to the indeterminate group. Conclusions In a community-based sample with normal LVEF, replacing LAVi with PALS in diastolic function classification separated groups with significant differences in their long-term all-cause mortality. This adjustment also reduced the size of the indeterminate group and refined its mortality risk.Figure 1Figure 2