The incidence of Heart failure with preserved ejection fraction (HFpEF) in Heart failure patients increases rapidly. However, the current clinical awareness is insufficient, and the cardiac structural and functional injury are not well understood. It is difficult to recognize the subclinical changes of the cardiac in the early stage with conventional imaging techniques, and it is common to ignore the existence of the clinical alterations. This study aimed to investigate the cardiac features, early diagnosis and risk factors of HFpEF patients, based on the multi-modality (Magnetic resonance imaging- nuclear medicine imaging- echocardiography) imaging and multicenter study, combined with large data and artificial intelligence. This study will provide deep insights into the HFpEF in multicenter population.

Heart Failure with Preserved Ejection Fraction (HFpEF) is a special subtype of Heart Failure (HF), and the incidence of HF cases is rising to 4.5 million every year, according to "Chinese cardiovascular disease report 2018" and "China Heart Failure and diagnostic guidelines 2018". In 2000, the incidence of patients with chronic Heart Failure is as high as 0.9%, and faces significantly increase with the increase of age. Moreover, HFpEF patients accounted for over 50% of heart failure, presenting normal left ventricular ejection fraction (LVEF), and nonspecific HF clinical performance. In addition, as a heterogenous disease, HFpEF is often associated with various comorbidities, including hypertension (~ 75%), diabetes (~ 40%), obesity (> 80%), aging (~ 75 years), renal dysfunction (25-50%), pulmonary hypertension (~ 50%), and other diseases. There is still much confusion about the pathophysiology of the disease, and no effective treatment was confirmed, therefore the diagnosis and treatment HFpEF has some challenges. With the increase of cardiovascular risk factors such as hypertension (morbidity: 23.2% in 2018), diabetes (morbidity:10.9% in 2018, treatment rate 32.2%) and the aging trend, the morbidity and mortality of HFpEF are still on the rise, posing a threat to the life quality of more and more patients. Early identification and intervention of HFpEF is an important method to reduce mortality and improve prognosis. Yet, many studies have explored the role of different biochemical and inflammatory markers in the diagnosis and prognosis assessment of HFpEF, limited for mixed indicators and low sensitivity.

Cardiac Magnetic Resonance imaging (CMR) is a non-invasive "one-stop" examination, including cardiac structure, function, tissue characteristics, blood perfusion examination. In particular, the emerging T1 mapping and Feature Tracking (FT) techniques enable the early and quantitive identification of cardiac dysfunction prior to abnormal LVEF. It has been found that the Extracellular Volume Fraction (ECV) based on T1 mapping and the myocardial strain parameters based on FT have the ability to diagnose and predict the prognosis of HFpEF patients. Echocardiography takes advantages in early identification of HFpEF patients and reveals the diastolic dysfunction. Nuclear medicine imaging shows priorities in blood perfusion and myocardial viability verification. Magnetic resonance imaging - echocardiography - nuclear medicine multimodal imaging complements and promotes each other, for example, molecular nuclear medicine imaging (recognition of metabolism), echocardiography (primary selection and determination of diastolic dysfunction), as well as the noninvasive high-resolution magnetic resonance and new emerging molecular imaging (identification of macroscopic, microscopic structure and function). The multimodel imaging overcomes the limits of single imaging method, greatly improves the accuracy of early diagnosis ability and diagnosis. However, large studies are based on single-center studies and meta analysis of small samples, and the comprehensive markers derived from multimodel study and multicenter study are lacked. Domestic relevant studies are in the initial stage.

To sum up, this study attempts to achieve early diagnosis and intervention of HFpEF and improve life quality of HFpEF patients through a multicenter, large sample cooperative study based on multimodel imaging (CMR imaging, echocardiography, nuclear medicine imaging). This study is expected to deepen the understanding of the pathogenesis and pathophysiological characteristic of HFpEF, providing a set of parameters based on multimodel imaging, hence assisting in early identification of cardiac structure and function change, early diagnosis of HFpEF and achieving risk stratification. In other way, the marker derived from this study may help target treatment of HFpEF.

Inclusion Criteria:

• left ventricular ejection fraction (LVEF)≥50%；
• N-terminal pro-b type natriuretic peptide (NT-proBNP)>220pg/ml or b type natriuretic peptide (BNP) >80 pg/ml;
• symptoms and syndromes of heart failure;
• At least one criteria of cardiac structure (left ventricular hypertrophy, or left atrial enlargement) and function abnormalities (based on tissue doppler, color doppler).

Exclusion Criteria:

• Special types of cardiomyopathy, including hypertrophic cardiomyopathy, restricted cardiomyopathy, etc.
• Infarction, myocardial fibrosis caused by ischemic cardiomyopathy and acute coronary syndrome ;
• Severe arrhythmia;
• Severe primary cardiac valvular disease;
• Restrictive pericardial disease;
• Refuse to participate in the study.

This study is a prospective, multicenter study with HFpEF patients, diagnosed based on 2016 and 2019 European Society of Cardiology (ESC) concensus. The inclusion criteria include left ventricular ejection fraction (LVEF)≥50%；N-terminal pro-b type natriuretic peptide (NT-proBNP)>220pg/ml or b type natriuretic peptide (BNP) >80 pg/ml; symptoms and syndromes of heart failure; and at least one criteria of cardiac structure (left ventricular hypertrophy, or left atrial enlargement) and function abnormalities. We exclude patients with special cardiomyopathy, ; Infarction, myocardial fibrosis caused by ischemic cardiomyopathyand acute coronary syndrome; Severe arrhythmia; Severe primary cardiac valvular disease; Restrictive pericardial disease; Refuse to participate in the study.

Our study data is applicable to other researchers with permmsion.