Refining diastolic function assessment: replacing left atrial volume with reservoir strain and its associations with long-term outcomes in a cohort with normal left ventricular systolic function

Left atrial reservoir strain (LASr) has emerged as a sensitive marker of LA function and elevated filling pressures, even though its role in the assessment of diastolic dysfunction (DD) and the subsequent risk stratification has remained relatively underexplored.

Accordingly, we aimed to investigate the prognostic implications of replacing left atrial volume index (LAVi) with LASr in the ASE/EACVI algorithm for diagnosing DD in subjects with normal left ventricular (LV) systolic function.

We retrospectively identified 1377 volunteers from a population-based screening program with normal LV systolic function and no evidence of myocardial disease. Echocardiographic measurements comprised recommended parameters of diastolic function and LASr by speckle tracking. Diastolic function was assessed using the 2016 ASE/EACVI algorithm and its modified version in which LAVi >34 ml/m² was replaced with LASr <23%. The primary endpoint was the composite of all-cause mortality and heart failure hospitalization.

During a median follow-up of 11 years, 179 (13%) subjects met the primary endpoint. Compared to having normal function based on the 2016 ASE/EACVI algorithm, being classified as indeterminate and DD had threefold and nearly fivefold (HRs: 3.575 [95% CI: 2.497 – 5.117], p<0.001, and 4.703 [95% CI: 2.955 – 7.484], p<0.001) increased risks of meeting the primary endpoint, respectively. Interestingly, the survival curves of the indeterminate and DD groups did not separate (Figure 1, *: log-rank p<0.05). After reclassification using the modified algorithm, the size of the indeterminate group decreased by 45% and the survival curves of the function groups separated (Figure 2). It also resulted in a fourfold risk for indeterminate and an eightfold (HRs: 3.936 [95% CI: 2.586-5.991], p<0.001, and 8.102 [95% CI: 4.948-13.269], p<0.001) for DD compared to normal function. DD was associated with a twofold (HR: 2.058 [95% CI: 1.128-3.757], p=0.019) risk compared to the indeterminate group. Using univariable significant predictors of the primary outcome (sex, age, systolic blood pressure, body mass index), adjusted analysis was performed. Compared to normal function based on the 2016 ASE/EACVI algorithm, those classified as having DD had a significantly higher risk of meeting the primary endpoint (adjusted HR: 1.662 [95% CI: 1.025 – 2.696], p=0.040), whereas those classified as indeterminate did not (adjusted HR: 1.345 [95% CI: 0.924 – 1.958], p=0.122). Using the modified algorithm, reclassification resulted in both the indeterminate and DD groups exhibiting significantly higher risk than the normal group (adjusted HRs: 1.562 [95% CI: 1.004 – 2.430], p=0.048, and 2.741 [95% CI: 1.620 – 4.638]), p<0.001, respectively).

In a community-based cohort with normal LV function, replacing LAVi with LASr in the ASE/EACVI algorithm for diagnosing DD resulted in improved classification and subsequent risk stratification.Figure 1Figure 2