Abstract Body (Do not enter title and authors here): Background
Coronary microvascular dysfunction (CMD) has been found common in patients with heart failure (HF) with preserved ejection fraction (HFpEF). The simplest and most feasible method for evaluating CMD is to measure coronary flow reserve (CFR) using echocardiography.

Hypothesis
Whether the echocardiography-assessed CFR could predict the outcome in HFpEF patients is still unknown.

Aims
We designed a prospective cohort study to investigate the prognostic effect of echocardiography-assessed CFR in HFpEF.

Methods
Between December 2021 and April 2024, confirmed hospitalized HF patients with left ventricular ejection fraction (LVEF) ≥40% in Zhongshan Hospital Fudan University (Shanghai, China) were prospectively and consecutively enrolled. During index hospitalization, patients underwent the adenosine-based transthoracic Doppler echocardiography to evaluate the CFR of the left anterior descending branch. Treatments were given according to guidelines. Follow-up after discharge was carried out at outpatient or by telephone call. The primary endpoint was all-cause mortality.

Results
A total of 176 HFpEF patients (111 males, mean age 66.8 ± 11.8 years old) were included. Median follow-up time was 290 days. Totally 93 patients (52.8%) demonstrated CFR <2.5, which is the classical CMD diagnostic threshold. However, CFR=2.5 had no significant discriminative ability in predicting mortality in this HFpEF cohort. Then the receiver operating characteristic curve combined with Youden's index were performed and indicated CFR=1.72 as the cut-off value to predict all-cause mortality. So patients were divided into two groups by CFR <1.72 (n=28) and CFR ≥1.72 (n=148). Compared with patients with CFR ≥1.72, HFpEF patients with CFR <1.72 were more likely to have atrial fibrillation, had higher level of high sensitive Troponin T, NT-proBNP, urinary albumin to creatinine ratio, and lower level of reactive hyperemia index measured by EndoPAT. During follow-up, 4 and 3 patients died in CFR < or ≥ 1.72 group, respectively (14.3% vs. 2.0%, P=0.013). Kaplan-Meier curves showed that patients with CFR <1.72 had poorer survival rate (Log-rank P=0.003). HFpEF patients had much higher risk in all-cause mortality if CFR <1.72 (hazard ratio 7.873, 95% CI 1.683-36.840) after adjusted for age, sex, NYHA class and LVEF.

Conclusions
Lower level of echocardiography-assessed CFR obtained in hospitalization was associated with higher risk of all-cause mortality after discharge in HFpEF patients.