Beyond conventional measures: left atrial reservoir strain as a superior prognostic marker for all-cause mortality in HFpEF

Considerable data are available concerning improvements in diagnostic approaches over recent years for the treatment of heart failure with reduced ejection fraction (HFrEF). Conversely, regarding heart failure with preserved ejection fraction (HFpEF), there has been relatively little change in terms of therapy and prognostic stratification. Many efforts have been made to phenotype HFpEF spectrum, aiming to identify simple and reliable tools for recognizing patients at high risk for mortality.

The aim of the present study is to identify a prognostic tool for all-cause mortality in patients with HFpEF.

From 2013 to 2024 thirty-eight HFpEF patients who underwent both transthoracic echocardiography at rest and cardiopulmonary exercise test (CPET) combined with echocardiography have been enrolled. Patients with established cardiomyopathy, such as hypertrophic cardiomyopathy and cardiac amyloidosis, have been excluded from the analysis. Mean follow-up period was 3,7 years. Primary outcome was all-cause mortality.

Sixteen patients (42%) died during the study's follow-up period. No statistically significant differences were found for clinical, demographic, and therapeutic variables. Among all echocardiographic parameters, both at rest and during exertion, E/e' at rest (12 [8.9-19] vs. 17 [12-33], p-value 0.023), SPAP at rest and during exertion (31 ± 7 vs. 43 ± 16, p-value 0.021 and 51 ± 10 vs. 66 ± 17, p-value 0.019), TAPSE/SPAP at rest and during exertion (0.67 ± 0.15 vs. 0.47 ± 0.20, p-value 0.005 and 0.48 ± 0.15 vs. 0.33 ± 0.12, p-value 0.019), and left atrial reservoir function (22 [15-29] vs. 9.4 [6-15], p-value 0.001) were significantly associated with all-cause mortality (Table 1). Among CPET variables, respiratory exchange rate and oxygen pulse were associated with all-cause mortality (1.1 ± 0.13 vs. 1.18 ± 0.15, p-value 0.021 and 10 [7-11] vs. 8.1 [5.4-8.7], p-value 0.008) (Table 1). In multivariate analysis, the only independent predictor of mortality was left atrial reservoir function (p-value 0.026, OR 1.17 [1.02-1.35]). The optimal cutoff was identified using the Youden Index analysis (cut-off: 16.4%, sensitivity 85%, specificity 68%; AUC 0.839 [0.710-0.969]) (Figure 1). Survival curves based on the optimal cutoff are reported in Figure 1.