Heart failure with preserved ejection fraction (HFPEF) is common and is a real public health issue. Diagnosis, especially when there are no congestive signs, is difficult. It has been shown that many patients with suspected HFPEF had left ventricular (LV) filling pressures elevated only at exercise (normal at rest).

Using stress echocardiography and taking into account left atrial (LA) remodeling at rest as a "memory" of chronic elevation of filling pressures. We believe that it is possible to improve the noninvasive diagnosis of exercise elevation of the LV end-diastolic pressure (LVEDP).

Purpose Heart failure with preserved ejection fraction (HFPEF) is common and is a real public health issue. Diagnosis, especially when there are no congestive signs, is difficult. It has been shown that many patients with suspected HFPEF had left ventricular (LV) filling pressures elevated only at exercise (normal at rest).

Hypothesis Using stress echocardiography and taking into account left atrial (LA) remodeling at rest as a "memory" of chronic elevation of filling pressures. We believe that it is possible to improve the noninvasive diagnosis of exercise elevation of the LV end-diastolic pressure (LVEDP).

Methods Prospective, monocentric and comparative study: catheterization versus echocardiography.

60 patients referred for coronary angiography will be recruited consecutively during their hospitalization.

Patients should perform a low intensity and short duration exercise, in both catheterization and echo labs (pedaling 3 minutes at 25Watts then, 3 minutes at 50W) The LVEDP will be measured invasively with a pigtail, at rest and at both levels of exercise.

Echocardiography will be performed within 24 hours after catheterization, after a full examination at rest, an identical exercise (same intensity, same duration, same position of the patient) than made in catheterization lab will be done. Following parameters will be recorded at both stress levels: trans mitral flow, mitral annular pulsed tissue Doppler imaging (both lateral and septal) and tricuspid regurgitation flow.

The doctor who will perform the echocardiographic acquisitions will not be informed of the results of catheterization. The acquisitions will be analyzed in a second time still blinded to the catheterization data.

The following echocardiographic parameters will be collected and compared to the invasive measurement of LVEDP:

• Ratio between pulsed Doppler peak E velocity and peak Ea velocity obtained with tissue Doppler imaging (E/Ea ratio) at rest and exercise,
• maximal LA volume indexed to body area
• (maximal LA volume) to (maximal LV volume) ratio.
• LA distensibility defined by: (maximal LA volume - minimal LA volume) / (minimal LA volume)
• LA Global longitudinal strain Finally, it will be investigated whether the combined use of E/Ea ratio at exercise with LA remodeling indices (of morphology and/or function) improves the performance characteristics of diagnostic test, compared to a separate use of these parameters.

Inclusion Criteria:

• Patients hospitalized for coronary angiography
• Sinus Rhythm
• Left ventricular Ejection Fraction > 50%

Exclusion Criteria:

• Age < 18years
• Hypertrophic Cardiomyopathy
• Cardiac transplantation
• Mitral stenosis
• Mitral insufficiency >2/4
• Severe calcification of mitral annulus
• Mitral prothesis or mitral repair
• aortic prothesis
• Severe aortic stenosis
• Atrial fibrillation
• Acute coronary syndrom < 3 months
• Left ventricular thrombus
• Severe renal Failure
• failure of radial way for coronary angiography
• coronary lesion indicating an angioplasty
• impossibility to perform an exercise
• refusal or inability to sign informed consent
• no French medical insurance