Derivation and validation of a cut-off value for the TAPSE/PASP ratio for predicting long-term all-cause mortality in heart failure with preserved ejection fraction

Introduction

Right ventricular (RV) – pulmonary artery (PA) uncoupling represents the critical moment when the RV systolic force is surpassed by the RV afterload, an essential point in the trajectory of heart failure (HF) patients, of utmost importance for the progression of congestion and decompensation of these patients, linked to vital prognosis. A surrogate non-invasive method of quantifying the RV-PA coupling is the tricuspid annular plane systolic excursion (TAPSE) to pulmonary artery estimated systolic pressure (PASP) ratio, evaluated by echocardiography, a parameter linked to mortality in patients with HF with reduced ejection fraction (EF).

Scarce data is available regarding the prognostic role of the TAPSE/PASP ratio in patients with HF with preserved EF (HFpEF).

Purpose

Our aim was to derive and validate a cut-off level for the TAPSE/PASP ratio for all-cause long-term mortality of HFpEF patients after hospitalization.

Methods

We conducted a retrospective cohort study including two samples of HFpEF patients admitted to our cardiology department between January-December 2019 for the derivation cohort and between April-October 2022 for the validation cohort.

We excluded patients with acute coronary syndrome, pulmonary embolism, repeated admissions, in-hospital death, and absence of necessary echocardiographic data from the initial hospitalization.

All-cause mortality for the derivation cohort was assessed after a mean follow-up duration of 4 years and for the validation cohort, after a mean follow-up of 3.4 years.

Receiver operating characteristic (ROC) analysis was used to determine the TAPSE/PASP cut-off value based on the Youden index associated criterion, and the optimal sensitivity (Se), and specificity (Sp).

Results

The derivation cohort included 121 patients, 38.22% male, mean age of 72.21±9.68 years, mean TAPSE/PASP ratio of 0.65±0.32 mm/mmHg. Long-term all-cause mortality was 29.26%. In ROC analysis TAPSE/PASP was associated with all-cause mortality with an AUC 0.76, 95% CI 0.68–0.83, cut-off ≤0.46 mm/mmHg, Se 69.44%, Sp 80.23%, p<0.001. Patients with TAPSE/PASP ≤0.46 mm/mmHg had an OR 8.75 (95%CI 3.63 – 21.11, p< 0.001) for all-cause long-term mortality. In multivariable analysis, TAPSE/PASP≤0.46 mm/mmHg was an independent predictor of the outcome (HR 5.59, 95% CI 2.02–15.45, p<0.001).

The validation cohort included 86 patients, 46.51% male, mean age of 70.71 ± 12.88 years, mean TAPSE/PASP of 0.70 ± 0.30 mm/mmHg. Long-term all-cause mortality was 14.63%. TAPSE/PASP had a strong predictive power for all-cause mortality with an AUC 0.82, 95%CI 0.72 – 0.89, p<0.001. Patients with TAPSE/PASP≤0.46 had an OR of 15.01 (95%CI 3.67 – 61.36, p<0.001) for all-cause mortality.

Conclusion

TAPSE/PASP is an independent predictor of all-cause long-term mortality of HFpEF patients after hospitalization for decompensation. We suggest the validated cut-off of TAPSE/PASP ≤0.46 for stratifying the vital prognosis of these patients.