GUIDELINES AND STANDARDS

Recommendations for the Evaluation of Left

Ventricular Diastolic Function by
Echocardiography and for Heart Failure With
Preserved Ejection Fraction Diagnosis: An
Update From the American Society of
Echocardiography
Sherif F. Nagueh, MD, FASE (Chair), Danita Y. Sanborn, MD, FASE (Co-Chair), Jae K. Oh, MD, FASE,
Bonita Anderson, MApplSc, DMU, ACS, FASE, FASA, Kristen Billick, BS, ACS, RCS, RDCS, FASE,
Genevieve Derumeaux, MD, PhD, Allan Klein, MD, FASE, Konstantinos Koulogiannis, MD, FASE,
Carol Mitchell, PhD, ACS, RDMS, RDCS, RVT, RT(R), FASE, Amil Shah, MD, Kavita Sharma, MD,
Otto A. Smiseth, MD, PhD, Honorary FASE, and Teresa S. M. Tsang, MD, FASE, Houston and Dallas, Texas;
Boston, Massachusetts; Rochester, Minnesota; Brisbane, Australia; San Diego, California; Creteil, France; Cleveland,
Ohio; Morristown, New Jersey; Madison, Wisconsin; Baltimore, Maryland; Oslo, Norway; and Vancouver, British
Columbia, Canada

Echocardiographic assessment of left ventricular (LV) diastolic function is an integral part of the routine eval-
uation of patients presenting with symptoms of dyspnea or clinical concerns for heart failure. Given the pres-
ence of diastolic dysfunction in many cardiovascular diseases, clinical reports should include comments on
diastolic function and/or left atrial (LA) pressure whenever possible. Since the publication of the 2016 ASE/
EACVI guidelines for assessment of LV diastolic function, new data on additional echocardiographic variables
as left atrial strain and their association with LV filling pressures have emerged. Moreover, prognostic data

From the Methodist DeBakey Heart and Vascular Center, Houston Methodist from and is a scientific advisory board member for Cardiol Therapeutics; Research
Hospital, Houston, Texas (S.F.N.); Massachusetts General Hospital, Boston, grant and scientific advisory board at Ventyx; has served on research grants and
Massachusetts (D.Y.S.); the Mayo Clinic, Department of Cardiovascular the scientific advisory board at Pfizer; has received book royalties from Wolters
Medicine, Rochester, Minnesota (J.K.O.); Cardiac Sciences Unit, The Prince Kluwer, and Elsevier. Konstantinos Koulogiannis, MD, FASE, is a consultant for Ab-
Charles Hospital, Brisbane, QLD, Australia and Faculty of Health, School of bott and the Edwards Lifesciences Mitral Advisory Board. Carol Mitchell, PhD,
Clinical Sciences, Medical Radiation Sciences, Queensland University of ACS, RDMS, RDCS, RVT, RT(R), FASE, is a member of the following professional
Technology, Brisbane, QLD, Australia (B.A.); the Scripps Clinic and Scripps La societies: the American Society of Radiologic Technologists, the American Institute
Jolla Memorial Hospital, La Jolla, California (K.B.); IMRB, Faculte
de Sante
de of Ultrasound in Medicine, the Society for Vascular Ultrasound, and the Society of
Creteil, Creteil, France (G.D.); the Section of Cardiovascular Imaging, Department Diagnostic Medical Sonography; is an author of textbook chapters for Elsevier; and
of Cardiovascular Medicine, Heart, Vascular and Thoracic Institute, Cleveland has received W. L. Gore & Associates contracted research grants to the University
Clinic, Cleveland, Ohio (A.K.); the Department of Cardiovascular Medicine of Wisconsin-Madison. Jae K. Oh, MD, FASE, FACC, FAHA, FESC: UpToDate
Morristown Medical Center, Gagnon Cardiovascular Institute, Morristown Medical Section Editor for Pericarditis, Anumana for potential royalty from AI ECG
Center, Morristown, New Jersey (K.K.); Division of Cardiovascular Medicine, Diastolic Function Algorithm, REDNVIA for research grant in EVOID AS trial, and
Department of Medicine, University of Wisconsin School of Medicine and Public Medtronic for consulting in valvular heart disease projects. Kavita Sharma, MD,
Health, University of Wisconsin-Madison, Madison, Wisconsin (C.M.); the Division Department of Cardiovascular Medicine Morristown Medical Center. Otto A. Smi-
of Cardiology, University of Texas Southwestern Medical Center, Dallas, Texas seth, MD, PhD, Honorary FASE: co-inventor of ‘‘Method for myocardial segment
(A.S.); the Division of Cardiology, Johns Hopkins University School of Medicine, work analysis’’, co-inventor of ‘‘Estimation of blood pressure in the heart’’ and
Baltimore, Maryland (K.S.); the Institute for Surgical Research, Division of has received one speaker honorarium from GE Healthcare. Teresa S. M. Tsang,
Cardiovascular and Pulmonary Diseases, Oslo University Hospital, Rikshospitalet MD, FRCPC, FACC, FASE: Executive board member of Canadian Society of Echo-
and University of Oslo, Oslo, Norway (O.A.S.); and the Division of Cardiology, cardiography
University of British Columbia, Vancouver, British Columbia, Canada (T.S.M.T.).
The following authors reported no actual or potential conflicts of interest in relation Attention ASE Members:
to this document: Bonita Anderson, MApplSc, DMU, ACS, FASE, FASA, Genevieve Login at www.ASELearningHub.org to earn CME credit through an online ac-
Derumeaux, MD, PhD, Sherif F. Nagueh, MD, FASE, Danita Y. Sanborn, MD, FASE, tivity related to this article. Certificates are available for immediate access
Amil M. Shah, MD, MPH, and Teresa S. M. Tsang, MD, FRCPC, FACC, FASE. upon successful completion of the activity and postwork. This activity is free
The following authors reported relationships with one or more commercial inter- for ASE Members, and $40 for nonmembers.
ests: Kristen Billick, BS, ACS, RCS, RDCS, FASE, is a clinical specialist at Lantheus
Medical Imaging; speaker at Bristol Myers-Squibb; member at Large ASE Board of Reprint requests: Sherif F. Nagueh, MD, FASE, Methodist DeBakey Heart and
Directors, the Cardiovascular Credentialing International Advisory Board, and the Vascular Center, Houston Methodist Hospital, 6565 Fannin Street, Houston,
Cardiovascular Credentialing International Advanced Cardiac Sonographer Exam TX 77030 (E-mail: [email protected]).
Writing Committee; and a Co-Chair ASE Nominations Committee, CCI Advisory 0894-7317/$36.00
Board, CCI Advanced Cardiac Sonographer (ACS) Exam Writing Committee, Trea-
Copyright 2025 Published by Elsevier Inc. on behalf of the American Society of
surer for the San Diego Society of Echo. Allan Klein, MD, FASE, is President-Elect
Echocardiography.
for the National Board of Echocardiography; has received research grants from and
https://doi.org/10.1016/j.echo.2025.03.011
is a scientific advisory board member for Kiniksa; has received research grants
537
538 Nagueh et al Journal of the American Society of Echocardiography
July 2025

from epidemiologic studies have demonstrated the association of echocardiographic measures with the
subsequent development of heart failure. This update provides a contemporary approach for the assessment
of LV diastolic function and the estimation of LA pressure in the general population of patients in sinus rhythm
referred for echocardiographic evaluation, and in special populations that require deviation from the general
approach. The update also discusses the application of echocardiography in the diagnosis of patients with
heart failure with preserved LV ejection fraction. (J Am Soc Echocardiogr 2025;38:537-69.)

Keywords: Diastole, Echocardiography, Doppler, Heart failure

TABLE OF CONTENTS
AR = Aortic regurgitation routine evaluation of patients pre-
1. Introduction 538 AS = Aortic stenosis senting with symptoms of dys-
Clinical and Technical Considerations 539 pnea or clinical concerns for
2. Invasive Assessment of LV Diastolic Function 539 ASE = American Society of heart failure (HF). Given the pres-
3. Normal Ranges for Diastolic Measurements 540 Echocardiography ence of diastolic dysfunction in
4. Reference Ranges Compared With Prognostic Values for Diastolic AV = Atrioventricular many cardiovascular diseases,
Function Measurements 551 clinical reports should include
5. Age-Independent Indices of Elevated LV Filling Pressures 552 CMR = Cardiac magnetic comments on diastolic function
6. LV Structure and LA Volume and Function 552 resonance
and/or LV filling pressures when-
7. LV Global Longitudinal Strain 552
CRT = Cardiac ever possible. The 2016
8. LA Strain 552
resynchronization therapy American Society of
9. Other Imaging Modalities for Assessment of LV Diastolic
Function 552 CT = Computed tomography Echocardiography (ASE) and
10. Definition of LV Diastolic Dysfunction Using Echocardiography 552 European Association of
11. Algorithm for Estimation of Mean LAP at Rest 553 CW = Continuous-wave Cardiovascular Imaging guide-
Supplemental Parameters 554 EF = Ejection fraction lines for diastolic function assess-
12. Reporting on Diastolic Function 555 ment sought to simplify the
13. Diastolic Exercise Echocardiography 555 ESC = European Society of clinical approach1 and have
A. Indications 555 Cardiology been shown to have good accu-
B. Performance 555 racy in a large multicenter study.2
GLS = Global longitudinal
C. Interpretation 556
strain Incomplete data, conflicting pa-
14. Application of AI to the Assessment of LV Diastolic Function 556
rameters, and/or uncertainty
15. Assessment of LV Diastolic Function and Estimation of LV Filling Pres- HCM = Hypertrophic
sures in Special Populations 556 cardiomyopathy
about inclusion and exclusion
A. Assessment of LV Diastolic Function in Patients With Valvular Heart criteria still result in an unaccept-
Disease 557 HF = Heart failure ably high frequency of unclassifi-
B. Heart Transplant Recipients 559 HFpEF = Heart failure with able or indeterminate cases
C. Pulmonary Hypertension 560 preserved left ventricular when using the 2016 guidelines
D. AV Block, Bundle Branch Block, and Electronic Pacing 560 ejection fraction algorithm. New data on addi-
E. Restrictive Cardiomyopathy 561 tional echocardiographic vari-
F. Pericardial Constriction 561 IVRT = Isovolumic relaxation
ables (e.g., left atrial [LA] strain)
G. HCM 562 time
and their associations with LV
H. Atrial Fibrillation 563
LA = Left atrial filling pressures have emerged.3
16. HFpEF Diagnosis 563
A. Clinical Diagnosis of HFpEF 563 LAP = Left atrial pressure Moreover, prognostic data from
B. Echocardiographic Imaging 565 epidemiologic studies have
C. Natriuretic Peptides 565 LARS = Left atrial reservoir demonstrated the association of
D. Role of HFpEF Prediction Scores 565 strain echocardiographic measures
E. Alternative Diagnoses 566 LAV = LA volume with the subsequent develop-
F. Exercise and Invasive Hemodynamic Testing 566 ment of HF.4 Therefore, this up-
G. Research Needs 566 LAVi = Left atrial volume date has two primary goals: (1)
Notice and Disclaimer 566 index
to provide a more contemporary
Reviewers 566
LBBB = Left bundle branch approach for the assessment of
Supplementary Data 567
block LV diastolic function and the esti-
LV = Left ventricular mation of LV filling pressures and
1. INTRODUCTION (2) to discuss the application of
Abbreviations LVEDP = Left ventricular echocardiography in patients
Echocardiographic assessment of end-diastolic pressure with HF with preserved LV ejec-
2D = Two-dimensional
left ventricular (LV) diastolic func- LVEF = Left ventricular tion fraction (HFpEF). As this
AI = Artificial intelligence tion is an integral part of the ejection fraction guideline serves as an update to
Journal of the American Society of Echocardiography Nagueh et al 539
Volume 38 Number 7

the 2016 diastolic function guide-

MAC = Mitral annular 4. The echocardiographer should have a solid understanding of
line document, it is important to
calcification
note that there are several images the physiologic rationale behind each variable, the situations
MR = Mitral regurgitation in the original guideline that that make any given variable less reliable, and the technical as-
NP = Natriuretic peptide show important pathologic or pects of acquisition and analysis of the variables.
abnormal echocardiographic
PA = Pulmonary artery findings that are not republished
PADP = PA diastolic pressure in this document but may be of in-
2. INVASIVE ASSESSMENT OF LV DIASTOLIC FUNCTION
terest to readers of this update.
PASP = Pulmonary artery The document has three main
systolic pressure The invasive assessment of LV diastolic function relies on the estima-
sections: one for the general pop-
tion of two fundamental parameters, the time constant of LV relaxa-
PCWP = Pulmonary capillary ulation, a second for specific pop-
tion and the chamber stiffness constant. These determine LV pressure
wedge pressure ulations that require deviation
throughout diastole. The time constant of LV relaxation, t, during the
PR = Pulmonary regurgitation from the general approach, and
isovolumic relaxation period (isovolumic relaxation time [IVRT]) can
a third focused on the diagnosis
RAP = Right atrial pressure be estimated from high-fidelity solid-state manometry of the left
of HFpEF. The document also
ventricle. The resulting time-pressure data can then be analyzed to es-
RV = Right ventricular contains a summary and recom-
timate t (abnormally prolonged >48 ms). The more common ones
mendations for artificial intelli-
SRIVR = Strain rate during the include monoexponential decay model to zero asymptote, where
gence (AI) applications in this
isovolumic relaxation period P(t) = Po e
t/t (Supplemental Figure 1), or to nonzero asymptote,
field.
where P(t) = Po e
t/t + PN, where Po is the LV pressure at peak
STE = Speckle-tracking

dP/dt.5 Note that t is load dependent. The sensitivity of LV relaxa-
echocardiographic Clinical and Technical
tion to changes in afterload is most apparent in patients with systolic
Considerations
TDI = tissue Doppler imaging LV dysfunction. The relation between t and LVend-systolic pressure is
The application of the guidelines steeper in patients with systolic dysfunction than in hearts with
TEER = Transcatheter edge- starts with clinical data, including
to-edge repair normal LV systolic function6,7
age, heart rate, underlying Chamber stiffness affects the rate of LV pressure increase during LV
TR = Tricuspid regurgitation rhythm, blood pressure, two- filling. Chamber stiffness is most reliably obtained by analyzing LV
dimensional (2D) and Doppler end-diastolic volume and LVend-diastolic pressure (LVEDP) obtained
TTR = Transthyretin
echocardiographic findings with from conductance catheters from multiple cardiac cycles as filling is
respect to LV volumes and wall decreased by gradually inflating a balloon in the inferior vena cava.
thickness, ejection fraction (EF), It is determined primarily by the stiffness properties of the sarcomeres,
LA volumes, and presence and severity of mitral valve disease. The the interstitial space, and LV chamber geometry and wall thickness. It
guidelines are based on scientific work concerning diastolic function can be characterized by the LV chamber stiffness constant k. Table 1
and LV filling pressure in adults studied in ambulatory and acute hospital presents a summary of invasive measurements that indicate abnormal
care settings and thus should not be applied in children, normal pregnant LV diastolic function.5,7,8
women, or in intraoperative settings. The quality of the 2D, Doppler, The term LV filling pressures is frequently used but can refer to
and speckle-tracking echocardiographic (STE) signals as well as the lim- several different LV and LA diastolic pressures. These pressures
itations for each parameter should be carefully scrutinized. If a signal is include mean pulmonary capillary wedge pressure (PCWP), mean
suboptimal, it should not be used in formulating conclusions about LV
diastolic function. The presence of a single measurement that falls within
the normal range for a given age group does not necessarily indicate
normal diastolic function and conclusions should not be based on a sin- Table 1 Invasive measurement values that diagnose LV
gle measurement. In an individual patient, consistency among the diastolic dysfunction and HFpEF5,8
different indices is of great importance. For the most successful analysis
of diastolic function in any given case, the echocardiographer should Parameter Value
have a solid understanding of the physiologic rationale behind each var- 1. Time constant of LV relaxation (t), ms >48
iable, the situations that make any given variable less reliable, the proper 2. LV chamber stiffness constant >0.015*
acquisition technique, and the correct analysis of the echocardiographic
3. Rest mean PCWP, mm Hg >15
variables.
4. Rest LV end-diastolic pressure, mm Hg >16
Key Points 5. Exercise mean PCWP, mm Hg $25
6. Exercise LV end-diastolic pressure, mm Hg $23
1. The guidelines should not be applied to children, normal preg-
7. PCWP/Cardiac Output slope during supine exercise, >2
nant women, or the intraoperative setting.
mm Hg/L/min
2. The quality of the 2D echocardiographic images, Doppler
waveforms, and STE signals as well as the limitations for *Value based on the 90th percentile of control group without HFpEF,
where pressure and volume data were obtained by conductance cath-
each parameter should be carefully assessed. eter, from Kasner M, Westermann D, Steendijk P, et al. Utility of
3. The echocardiographic indices of diastolic function should al- Doppler echocardiography and tissue Doppler imaging in the estima-
ways be interpreted in the context of clinical status and other tion of diastolic function in heart failure with normal ejection fraction: a
echocardiographic parameters. comparative Doppler-conductance catheterization study. Circulation.
2007;116:637-647.
540 Nagueh et al Journal of the American Society of Echocardiography
July 2025

LA pressure (LAP), LV pre-A pressure, mean LV diastolic pressure, Table 2 Echocardiographic correlates with mean PCWP and
and LVEDP1 (Figure 1). In the early stages of diastolic dysfunction, with LVEDP
LVEDP is the only abnormally elevated pressure, while mean
PCWP and LAP remain normal. During tachycardia and/or increased PCWP, mean LAP, LV pre-A, and
LV afterload or intravascular volume, mean PCWP and LAP increase, mean LV diastolic pressure
which is the basis for diastolic stress testing (invasive and noninvasive). correlates LVEDP correlates

Some Doppler variables correlate with an increase in LVEDP, 1. Mitral peak E velocity 1. Mitral peak A velocity at tips level
whereas others reflect an increase in mean LAP or its surrogate 2. Mitral E/A ratio 2. A-wave duration at the annulus
(Table 2).
3. Mitral E velocity 3. Mitral A velocity (tips level)
Key Points deceleration time deceleration time
4. Mitral E/e0 ratio 4. Pulmonary vein peak Ar velocity
1. LV impaired relaxation is defined as a time constant of LV relax- 5. Pulmonary vein 5. Pulmonary vein Ar
ation >48 ms. systolic-to-diastolic duration
mitral A duration
2. Increased LV chamber stiffness is defined by a chamber stiff- velocity ratio
ness constant >0.015. 6. Peak TR velocity 6. LA minimum volume
3. Invasive criteria for HFpEF, in addition to the time constants of and PASP
LV relaxation and LV chamber stiffness, include mean PCWP 7. End-diastolic velocity 7. Tissue Doppler–derived mitral
of PR and PADP annular a0 velocity
at rest > 15 mm Hg, LVEDP at rest > 16 mm Hg, and mean
8. LARS 8. LA pump (or contractile) strain
PCWP with exercise $ 25 mm Hg.
4. Doppler measurements that correlate better with mean LAP
or PCWP include mitral E velocity, E/A ratio, and E/e0 ratio.
5. Echocardiographic measurements that correlate better with
3. NORMAL RANGES FOR DIASTOLIC MEASUREMENTS
LVEDP include mitral A velocity, pulmonary vein Ar velocity,
the time difference between Ar duration and that of mitral A Numerous studies have demonstrated the association between age
duration (Ar
A), and LA pump strain. and echocardiographic measurements of LV diastolic function (a list
6. Throughout this document, the term LV filling pressure refers to of these studies is included in the Supplemental Appendix). Some
mean LAP or its correlates (mean PCWP, LV pre-A pressure). It studies have further suggested that prognostic thresholds for dia-
should be noted that LVEDP and LAP are not the same, and stolic measurements may differ by age. The purpose of this section
although elevated LAP is always associated with elevated is to describe current estimates of normal ranges of diastolic mea-
LVEDP in patients with diastolic dysfunction, elevated LVEDP surements by age, on the basis of the observed range of values
may be present when LAP is normal. among subjects believed to be free of cardiovascular disease and
without known risk factors. Importantly, such normal ranges are
not necessarily equivalent to ‘‘optimal’’ values, as the aging process
Tables 3 and 4 summarize the technical aspects, hemodynamic itself may affect diastolic function. However, the use of age-specific
determinants, and clinical utility including advantages and limitations normal ranges can enhance clinical interpretation of diastolic indices
of each parameter used in the echocardiographic evaluation of LV and has become standard for echocardiographic measurements of
diastolic function. cardiac structure and function.

Figure 1 (Left) LV diastolic pressure recording. Arrows point to LV minimal pressure (min), LV rapid filling wave (RFW), LV pre-A pres-
sure (pre-A), A-wave rise with atrial contraction, and end-diastolic pressure (EDP). (Middle) LA pressure recording showing ‘‘V’’ and
‘‘A’’ waves marked along with Y and X descent. (Right) Simultaneous LV pressure and LAP recording showing early and late trans-
mitral pressure gradients. Notice that LA ‘‘A-wave’’ pressure precedes the late diastolic rise (LV A wave) in LV pressure.
 Volume 38 Number 7
Journal of the American Society of Echocardiography
Table 3 Parameters required for the assessment of LV diastolic function via transthoracic echocardiography

Parameter/variable Acquisition Measurements

Primary measurements
Transmitral Inflow 1. Apical four-chamber 6 color Doppler Peak E-wave velocity (cm/s): peak early
imaging to aid optimal alignment. diastolic modal velocity after ECG T wave
2. PW Doppler sample volume (1-3 mm) placed Deceleration time (ms): time interval from
at MV leaflet tips. peak E-wave velocity along the
3. Lower zero baseline and adjust velocity scale deceleration slope to the zero baseline
so signal above the zero baseline is as large Peak A-wave velocity (cm/s): peak late
as possible. diastolic modal velocity after ECG P wave
4. Use low wall filter setting (100-200 MHz) and E/A ratio: peak E-wave velocity divided by
low signal gain. peak A-wave velocity
5. Sweep speed at 100 mm/s. ± A duration (ms)*: Time interval from the
6. Optimal spectral waveforms should display onset to the offset of the A-wave signal at
minimal spectral broadening and not display zero baseline
spikes or feathering.
7. For A duration, when end of A-wave is not
well defined, sample volume may be lowered
a few millimeters toward the mitral annulus.
TDI at mitral annulus 1. Apical four-chamber with TDI preset e0 velocity (cm/s): peak early diastolic
(detection of low velocity, high amplitude modal velocity after ECG T-wave
signals). a0 velocity (cm/s): peak late diastolic modal
2. PW sample volume (5-10 mm) at septal and velocity after ECG P-wave
lateral insertion site of the mitral leaflets MV E/e0 ratio: MV peak E-wave divided by
(larger sample size required to ensure the TDI e0 velocity
sampling of annular excursion over systole Average E/e0 : MV peak E-wave divided by
and diastole). the average of the TDI septal e0 and lateral
3. Angle of interrogation should be as parallel as e0 velocities
much as possible to annular motion.
4. Adjust zero baseline and velocity scale to
display the full spectral signal above and
below the zero baseline as large as possible.
5. Sweep speed at 100 mm/s.
6. Optimal spectral waveforms should be sharp
and not display signal spikes, feathering or
ghosting.

(Continued )

Nagueh et al 541
 542 Nagueh et al
LAVi (mL/m2) 1. Apical four- and two-chamber views. From each view, trace the LA area (excluding
2. Each view optimized for left atrium. the PVs and LAA) and measure the LA
3. Avoid foreshortening by maximizing the length from the center of the MV annulus
width of the LA base (mitral annulus) and by to the center of the superior LA wall.
maximizing the LA long axis. This typically Ensure long-axis lengths are within 5 mm of
requires an anterior tilt, demonstrating the each other.
pulmonary veins entering the left atrium. LAV (mL): calculated via the method of disks
4. Acquire and freeze end-systolic frames (one or area-length method.
or two frames before MV opening). LAVi (mL/m2): LAV divided by BSA.

Peak TR velocity (m/s) 1. Acquired from any view that aligns TR jet Peak TR velocity (m/s): averaged over the
parallel with the ultrasound beam as noted respiratory cycle
via the color Doppler images.
2. CW Doppler cursor is aligned parallel to the
TR jet.
3. Adjust zero baseline and velocity scale to
ensure TR signal is displayed as large as

Journal of the American Society of Echocardiography

possible.
4. Optimize gain, compression and/or reject to
obtain a complete profile with minimal
spectral ‘‘bearding.’’
5. Sweep speed at 50-100 mm/s.

July 2025
 Volume 38 Number 7
Journal of the American Society of Echocardiography
PV inflow 1. Apical four-chamber with color Doppler Peak S-wave velocity (cm/s): peak systolic
imaging at a reduced Nyquist limit to aid velocity at ECG T wave. When two systolic
optimal alignment and identification of peaks (S1 and S2), the peak S2 should be
venous flow (anterior tilt may be required). measured for the S/D ratio
2. PW Doppler sample volume (3-5 mm) placed Peak D-wave velocity (cm/s): peak early
approximately 5-10 mm into the right upper diastolic velocity after ECG T wave
or right lower PV. S/D ratio: peak S-wave velocity divided by
3. Use low wall filter setting (100-200 MHz) and peak D-wave velocity
low signal gain. ± Peak AR velocity (cm/s)*: peak late
4. Adjust zero baseline and velocity scale to diastolic velocity after ECG P wave
display the full spectral signal above and ± AR duration (ms)*: Time interval from the
below the zero baseline as large as possible. onset to the offset of the AR-wave signal at
5. Sweep speed at 100 mm/s. zero baseline
6. Optimal spectral waveforms should not
display spikes or feathering.
7. Note: High PRF may be required when peak
velocities exceed the Nyquist limit. In this
instance, be aware of the second sample
volume position (if placed at the mitral valve
level, PV and MV signals will be
superimposed, and
PV S-wave and D-wave measurements will
not be accurate).
IVRT (ms) 1. Apical long-axis or five-chamber view. IVRT (ms): measured as the time interval
2. CW Doppler through LVOT to simultaneously between aortic valve closure and MV
display the end of aortic ejection and the opening.
onset of transmitral inflow.
3. Use low wall filter setting (100-200 MHz) and
low signal gain.
4. Adjust zero baseline and velocity scale to
display the full spectral signal above and
below the zero baseline as large as possible.
5. Sweep speed at 100 mm/s.
6. Optimal spectral waveforms should display
AV closing click and clear onset of transmitral
inflow in early diastole
(Continued )

Nagueh et al 543
 544 Nagueh et al
LV GLS (%) 1. Optimize the apical four-chamber, two- LV GLS (%): calculated using dedicated LV
chamber, and long-axis of the left ventricle, strain software to track the LV endocardial
avoiding foreshortening. wall and to calculate LV GLS. Ensure
2. Increase 2D gains to increase speckles. correct ECG gating for end-systole based
3. Narrow image sector and decrease image on aortic valve closure. Confirm tracking
depth for optimal frame rate (40-80 frames/ and adjust contour and region of interest if
s). Ensure sector wide enough to include full needed.
wall thickness and apex and depth to extend
beyond annulus to allow capture of entire left
ventricle throughout the cardiac cycle.
4. Region of interest should include 90% of the
myocardium but not the pericardium/
epicardium.
5. Confirm good-quality electrocardiogram.
6. Acquire three to five cardiac cycles for each
view ensuring similar heart rates for each
view.
Secondary measurements
Valsalva maneuver† 1. See above in primary measurements section Valsalva positive: E/A ratio < 1 or increase
for transmitral flow for the proper technique in A-wave velocity
of acquiring transmitral inflow signals. Valsalva negative: E/A ratio > 1
2. Patients should be instructed to bear down
against a closed glottis and practice this
technique before recording.
3. Transmitral inflow signal should be
continuously recorded for 10 to 12 s during
the strain phase of the maneuver.
4. Use a slower sweep speed (50 mm/s or
slower) to display the transmitral signal at
rest and during peak strain and/or during
peak strain and after release.
5. The acquired trace should be annotated to
indicate the use of the Valsalva maneuver.
6. An adequate Valsalva maneuver may be

Journal of the American Society of Echocardiography

defined as a >10% reduction in maximal E-
wave velocity from baseline.
Color M-mode Vp (cm/s)† 1. Apical four-chamber with color Doppler Vp (cm/s): measured from the level of the
imaging of transmitral inflow (variance mode mitral annulus to 4 cm into the LV cavity
off). along the early diastolic slope of first
2. M-mode cursor placed well aligned with the aliased velocity (red-blue interface)
path of transmitral inflow.
3. Lower the color Nyquist limit by either
decreasing the color velocity scale or by
moving the color baseline upward in the
direction of MV inflow to enhance the early
diastolic slope.

July 2025
 Volume 38 Number 7
Journal of the American Society of Echocardiography
TE-e0 time interval (ms)† See above in the primary measurements section TE (ms): time interval between the peak R
for transmitral flow for proper technique of wave on ECG and the onset of transmitral
acquisition of transmitral E-wave and TDI e0 E-wave velocity
velocities. Te0 (ms): time interval between the peak R-
wave on ECG and the onset of TDI e0
velocity.
R-R intervals should be matched.
TE-e0 (ms): Te0 minus TE

Peak PR end-diastolic 1. Acquired from any view that aligns PR jet Peak PR end-diastolic velocity (cm/s):
velocity (m/s) parallel with the ultrasound beam as noted measured at end-diastole
via the color Doppler images.
2. CW Doppler cursor is aligned parallel to the
PR jet.
3. Adjust zero baseline and velocity scale to
ensure PR signal is displayed as large as
possible.
4. Optimize gain, compression and/or reject to
obtain a complete profile with minimal
spectral ‘‘bearding.’’
5. Sweep speed at 50-100 mm/s.

(Continued )

Nagueh et al 545
 546 Nagueh et al
Advanced techniques
LA strain (%) 1. See above in the primary measurements LARS (%): peak positive strain value during
section for LA volumes for the proper ventricular systole.
technique for optimizing apical four-chamber LASct (%): measured in sinus rhythm as
and two-chamber dedicated LA views. 0 minus strain value at the onset of AC
2. Decrease gain and compression to optimize (pre-A wave on ECG), where 0 = strain
clean blood pool and LA tissue border, value at end-diastole (negative value).
minimizing artifact. LAScd (%): 0 minus strain value at AC
3. Narrow image sector for optimal frame rate (negative value).
(50-70 frames/s, preferably on higher end).
4. Confirm good quality ECG with a well-visible
P wave. Use R-R gating method to provide
reservoir, conduit, and contractile LA strain
values. Green dots, peak of the strain curve; red dots,
5. Acquire three to five cardiac cycles for each peak strain at onset of AC or at pre-A, S_R,
view ensuring similar heart rates for each LARS; S_CT, LASct; S_CD, LAScd.
view.
6. Use dedicated LA strain software to track LA
wall in both apical views (excluding
pulmonary veins and LAA).
7. Confirm tracking is on the underside of each
annular point, following the tissue boundary
and extrapolating the fossa ovalis, LAA and
PVs to the roof of the left atrium. Minimally
adjust contour if needed.
AC, Atrial contraction; AR, atrial reversal; BSA, body surface area; CD, LA conduit strain; ECG, electrocardiography; LARS = S_R, LA reservoir strain; LAScd = S_CD, LA conduit strain;
LASct = S_CT, LA contractile or pump strain; MV, mitral valve; PV, pulmonary vein; PW, pulsed-wave; Vp, early diastolic flow propagation velocity.
*Additive value when suspected elevated LVEDP and normal LAP (e.g., grade 1 diastolic dysfunction with elevated LVEDP and normal LAP).
†
May be attempted when pseudonormalization of the transmitral inflow profile is suspected. The Valsalva maneuver may also be performed to identify reversible and irreversible grade 3
diastolic dysfunction.

Journal of the American Society of Echocardiography

July 2025
Journal of the American Society of Echocardiography Nagueh et al 547
Volume 38 Number 7

Table 4 Utility, advantages, and limitations of variables used to assess LV diastolic function

Variable Utility and physiologic background Advantages Limitations

Mitral E-wave Reflects the LA-to-LV pressure gradient 1. Feasible and reproducible. 1. In patients with CAD and patients with
velocity during early diastole and is affected by 2. In patients with DCM and reduced HCM in whom LVEF is >50%,
changes in the rate of LV relaxation and LVEF, mitral velocities correlate better transmitral velocities correlate poorly
LAP. with LVFPs, functional class, and with LVFPs.
prognosis than LVEF. 2. More challenging to apply in patients
with arrhythmias.
3. Preload dependent.
4. Age dependent (decreasing with age).
Mitral A-wave Reflects the LA-to-LV pressure gradient Feasible and reproducible. 1. Sinus tachycardia, first-degree AV
velocity during late diastole (atrial contraction) block and paced rhythm can result in
and is affected by LV compliance and fusion of the E and A velocities. If E at
LA contractile function. onset of A is >20 cm/s, A velocity may
be higher than if diastole was longer.
2. Not applicable in AF/atrial flutter.
3. Age dependent (increases with aging).
Mitral E/A ratio Along with the DT, this ratio may be used 1. Feasible and reproducible. 1. Preload-dependent. A normal
to identify filling patterns: normal, 2. Increased ratio usually predicts transmitral profile may be difficult to
impaired relaxation, PN, and restrictive elevated LVFP in patients with differentiate from a PN transmitral
filling. myocardial disease but is not useful in profile, particularly with normal LVEF,
normal subjects. without additional variables.
3. Provides diagnostic and prognostic 2. When E velocity at onset of A is
information. >20 cm/s, E/A ratio will be reduced
4. In patients with DCM, LV filling pattern (see above).
correlates better with LVFPs, 3. Not applicable in AF/atrial flutter.
functional class, and prognosis than 4. Age dependent (decreases with
LVEF. aging).
5. A restrictive filling pattern in
combination with LA dilation in
patients with normal EFs is associated
with a poor prognosis similar to a
restrictive pattern in DCM.
Mitral DT Influenced by the rate of decline in LA-LV 1. Feasible and reproducible. 1. Preload-dependent.
pressure gradient after mitral valve 2. A short DT (<140 ms) in patients with 2. DT does not relate to LVEDP in normal
opening, and therefore LV relaxation reduced LVEFs indicates increased LVEFs.
and LV stiffness. LVEDP with high accuracy both in 3. Should not be measured with E and A
sinus rhythm and in AF. fusion or E at onset of A >20 cm/s
because of potential inaccuracy.
4. Age dependent (increases with aging).
5. Not applicable in atrial flutter.
Mitral A Reflection of LV compliance in late 1. In patients with cardiac disease, a 1. Cannot be reliably measured or used
duration diastole. longer A duration for age is usually when there is E and A fusion, sinus
Shortening occurs when there is reduced associated with normal LVFPs. arrhythmias, second- and third-
LV compliance resulting in a rise in LV 2. A shorter A duration (<120 ms) in degree AV block, or a short PR interval
pressure with atrial contraction, which patients with cardiac disease (<120 ms).
abruptly terminates transmitral inflow. indicates elevated LVFPs.
Best used in conjunction with pulmonary 3. When the ECG PR interval is normal,
venous AR duration (see below). termination of the A duration before
the peak ECG QRS complex is a
reliable indicator of elevated LVEDP.
Changes to Helps distinguish normal from PN When performed adequately under 1. Not every patient can perform this
mitral inflow transmitral patterns by reducing standardized conditions (keeping maneuver adequately. The patient
profile with preload. A decrease of E/A ratio of 40 mm Hg intrathoracic pressure must generate and sustain a sufficient
Valsalva $50% or an increase in A-wave constant for 10 s) accuracy in increase in intrathoracic pressure, and
maneuver velocity during the maneuver, not diagnosing increased LVFPs is good. the sample volume position needs to
caused by E and A fusion, are highly be maintained at the mitral leaflet tips
specific for increased LVFPs. during the maneuver.
2. It is difficult to assess if it is not
standardized.
(Continued )
548 Nagueh et al Journal of the American Society of Echocardiography
July 2025

Table 4 (Continued )
Variable Utility and physiologic background Advantages Limitations

MV L-wave Triphasic transmitral inflow profile with 1. When present in patients with known May rarely be seen with normal LV
velocity mid-diastolic flow indicates markedly cardiac disease (e.g., LVH, HCM), it is diastolic function when the subject has
delayed LV relaxation in the setting of specific for elevated LVFPs. However, bradycardia; however, when present,
elevated LVFPs and reflects a its sensitivity is overall low. the velocity is usually <40 cm/s.
continued LA-LV pressure gradient 2. Presence in AF may be associated
during diastasis. with increased LVFPs.
May also be seen on the TDI trace
between the e0 and a0 velocities.
TDI e0 velocity e0 velocity is an index of LV relaxation. 1. Feasible and reproducible. 1. Limited accuracy in patients with CAD
The hemodynamic determinants of e0 2. LVFPs have a minimal effect on e0 in and regional dysfunction in the
velocity include LV relaxation, restoring the presence of impaired LV sampled segments, significant MAC,
forces and filling pressure. e0 is relaxation. surgical mitral rings or prosthetic
decreased across all grades of 3. Less load dependent than other mitral valves and pericardial disease.
diastolic dysfunction. conventional PW Doppler parameters. 2. Need to sample at least two sites with
4. Helps distinguish PN from normal precise location and adequate size of
transmitral inflow profiles. sample volume.
3. Different cutoff values depending on
the sampling site.
4. Age dependent (decreases with
aging).
E/e0 ratio e0 velocity can be used to correct for the 1. Feasible and reproducible. 1. Not accurate in patients with heavy
effect increased preload on the 2. Values for average E/e0 ratio < 8 MAC, or prosthetic MV and pericardial
transmitral E velocity, E/e0 ratio can be usually indicate normal LVFPs, values disease.
used to predict increased LVFPs. >14 have high specificity for increased 2. ‘‘Gray zone’’ of values (E/e0 between 8
LVFPs. and 14) in which LVFPs are
indeterminate.
3. Accuracy is reduced in patients with
CAD and regional dysfunction at the
sampled segments.
4. Different cutoff values depending on
the sampling site.
LAVi LA volume reflects the cumulative effects 1. Feasible and reproducible. 1. LA dilatation, in the absence of
of increased LVFPs over time. It is 2. Provides diagnostic and prognostic diastolic dysfunction, may be seen in
directly but weakly related to LVFP. information about LV diastolic patients with high-output states, heart
Increased LA volume is an independent dysfunction and chronicity of disease. transplants with biatrial technique,
predictor of death, HF, AF, and 3. Apical 4-chamber view provides visual atrial flutter/fibrillation, significant MV
ischemic stroke. estimate of LA and RA size which disease, and in well-trained athletes.
confirms LA is enlarged. 2. Suboptimal image quality, including
4. A normal LA size suggests normal LA foreshortening, in technically
LVFPs. challenging studies precludes
accurate tracings.
3. It can be difficult to measure LA
volumes in patients with ascending
and descending aortic aneurysms and
large interatrial septal aneurysms.
TR velocity Can be used to estimate the PASP in the 1. PASP is passively elevated when the 1. Indirect estimate of LAP.
absence of PS or RVOT obstruction LAP is increased due to left heart PH; 2. Adequate recording of a full envelope
(RVSP = PASP). A significant correlation thus, an elevated PASP infers an is not always possible, though
exists between PASP and noninvasively elevated LAP. intravenous agitated saline or UEAs
derived LAP in group II PH. In the 2. Increased PASP indicates PH which increase yield.
absence of pulmonary disease, has prognostic implications. 3. Accuracy of calculation of PASP is
increased PASP suggests elevated LAP. dependent on the reliable estimation
of RAP.
4. With very severe TR and a low systolic
RV-RA pressure gradient, PASP
cannot be accurately estimated.
5. RVSP does not equal PASP when
there is PS or RVOT obstruction.
(Continued )
Journal of the American Society of Echocardiography Nagueh et al 549
Volume 38 Number 7

Table 4 (Continued )
Variable Utility and physiologic background Advantages Limitations

PR end- Can be used to estimate the PAEDP when 1. PAEDP is closely related to PCWP. 1. Adequate recording of a full PR jet
diastolic TR velocity cannot be accurately 2. Increased PAEDP indicates PH which envelope is not always possible
velocity measured. A significant correlation has prognostic implications. though UEAs increases yield.
exists between PAEDP and invasively, 2. Accuracy of calculation of PAEDP is
as well as noninvasively, derived dependent on the reliable estimation
PCWP. In the absence of pulmonary of RAP.
disease, the PAEDP approximates the 3. PAEDP overestimates PCWP by
PCWP. >5 mm Hg when there is increased
PVR.
Pulmonary S-wave velocity (sum of S1 and S2) is 1. Reduced S velocity, an S/D ratio <1 1. Feasibility of recording PV inflow can
veins: S influenced by LAP, LA contractility and and systolic filling fraction (systolic be suboptimal, particularly in ICU
velocity, D relaxation, LA stiffness, and LV and RV VTI/total forward flow VTI) < 40% patients.
velocity and contractility. D-wave velocity is indicate increased LAP in patients with 2. The relationship between PV systolic
S/D ratio influenced by changes in LAP in early reduced LVEFs. filling fraction and LAP has limited
diastole and LV relaxation and it 2. In patients with AF, DT of the D- accuracy in patients with normal EF,
changes in parallel with the transmitral velocity can be used to estimate mean AF, MV disease, and HCM.
E-wave velocity. AR velocity is PCWP. 3. AR velocity is absent in patients
determined by LA contractility and LV 3. AR velocity > 35 cm/s indicates an with AF.
compliance. increased LVEDP.
Decrease in LV compliance and increase
in LAP is associated with decrease in S
velocity, increase in D velocity, and an
increase in AR velocity and a longer AR
duration. The S/D is inversely related to
LAP.
Ar-A duration The time difference between durations of 1. AR duration > mitral A duration by 30 1. Adequate recordings of AR duration
PV flow and transmitral inflow during ms indicates an increased LVEDP. may not be feasible in many patients
atrial contraction is associated with the 2. Independent of age and EF. (comparing the end of AR and
LVEDP. The longer the time difference, 3. Accurate in patients with MR and transmitral A signals to the ECG QRS
the higher LVEDP. patients with HCM. may be helpful in identifying a
shortening of the transmitral A
duration).
2. Not applicable in patients with atrial
arrhythmias, sinus tachycardia or
heart block.
3. Transmitral A duration cannot be
reliably measured in certain cases (see
above).
IVRT Reflects the time interval aortic valve 1. Overall feasible and reproducible. 1. Limited use in isolation
closure and MV opening and the 2. IVRT can be combined with other 2. Age-dependent (shorter in young
crossover between LA and LV transmitral inflow parameters such as patients with rapid LV filling and
pressures. E/A ratio to estimate LVFPs in patients lengthens as relaxation slows with
Duration is directly related to LV with HFrEF. age)
relaxation and inversely related to LAP 3. Helpful in identifying elevated LVFPs in 3. Preload-dependent (normalizes with
(i.e., prolonged in patients with patients with MAC. increasing LAP)
impaired LV relaxation and normal 4. When IVRT markedly prolonged (>110 4. IVRT is in part affected by heart rate
LVFPs, shortened with increased LAP). ms), LAP likely normal. and arterial pressure.
5. A short IVRT (<70 ms) has a high 5. More challenging to measure and
specificity for elevated LAP in patients interpret with tachycardia.
with cardiac disease. 6. Identification of the onset of MV
6. In patients with MS or MR, IVRT can be opening can be challenging.
combined with TE-e0 to estimate
LVFPs.
(Continued )
550 Nagueh et al Journal of the American Society of Echocardiography
July 2025

Table 4 (Continued )
Variable Utility and physiologic background Advantages Limitations

LV GLS Measure of LV systolic function. Impaired 1. A more sensitive index of myocardial 1. Requires dedicated software package
LV GLS is common in some patients systolic performance than EF. which is not available in all institutions.
with HFpEF indicating subclinical LV 2. Provides earlier detection of 2. Suboptimal image quality in
dysfunction. myocardial disease in the setting of a technically challenging studies
normal EF. precludes accurate measurements.
3. Can be more reproducible than EF. 3. Load-dependent.
4. May be a superior discriminator of 4. Values may vary between vendors;
outcomes in patients with HFrEF and hence, results may not be
HFpEF. interchangeable.
Color M-mode Vp indirectly related to the time constant 1. Relatively load-independent. 1. In patients with normal LV volumes
Vp: Vp, and of LV relaxation (t); (the longer it takes 2. Vp is reliable as an index of LV and EF but elevated LVFPs, Vp can be
E/Vp ratio for the LV to relax, the slower the Vp). relaxation in patients with depressed misleadingly normal.
Vp can be used to correct for the effect LVEFs and dilated LV but not in 2. Vp pseudonormalization may also be
increased preload on the transmitral E patients with normal EFs. seen in patients with small,
velocity, E/Vp ratio is directly related to 3. Helps distinguish PN from normal hypertrophied ventricles.
the LAP. transmitral inflow profiles. 3. Lower feasibility and reproducibility.
4. The ratio of the transmitral E to Vp (E/ 4. Suboptimal alignment between M-
Vp) $ 2.5 predicts PCWP >15 mm Hg mode cursor and transmitral inflow
with reasonable accuracy in patients results in erroneous measurements.
with depressed EFs.
TE-e0 time Can identify patients with diastolic 1. Helps distinguish PN from normal 1. Non-simultaneous measurements
interval dysfunction due to delayed onset of e0 transmitral inflow profiles. (important to match R-R intervals).
velocity compared with onset of the 2. Ratio of IVRT to TE-e0 (IVRT/TE-e0 ) can 2. More challenging to acquire
transmitral E velocity. be used to estimate LVFPs in normal satisfactory signals with close
subjects and patients with MV attention needed to sampling location,
disease. gain, and filter settings.
3. Can be used to differentiate patients 3. Time intervals are numerically quite
with restrictive cardiomyopathy who small so any error in measurement
have a prolonged time interval from may prove significant.
those with pericardial constriction in
whom it is not usually prolonged.
LAS LAS in diastolic function primarily 1. LARS provides an estimate of LVFPs 1. Requires dedicated LAS software
focused on LARS. LARS shows a direct in patients with reduced EF. package which is not available in all
correlation with the degree of diastolic 2. LARS may be helpful in distinguishing institutions.
dysfunction (LARS worsens as the between degrees of diastolic 2. Suboptimal image quality in
degree of diastolic dysfunction dysfunction. technically challenging studies
worsens) and is inversely related to 3. LARS may be used as a substitute for precludes accurate measurements.
LVFP (the lower the LARS, the higher missing standard variables. 3. Values are age dependent (LARS
the LVFP). 4. LARS provides prognostic value in decreases with age).
As a late diastolic parameter, LASct patients with HF, AF, ischemic heart 4. LARS dependent on LV systolic
(pump strain) is inversely related to disease, and valvular heart disease. function (may be normal despite
LVEDP (the less negative the LASct, 5. LA dysfunction identified with LARS elevated LVFP in patients with normal
the higher the LVEDP). may show abnormalities before EF with preserved LV GLS).
anatomic changes occur. 5. R-R gating may be an inaccurate
substitution for end-diastole (e.g.,
when there is a BBB).
6. May be inaccurate when there is a
mobile atrial septum or a thin-walled
LA or when tracking does not follow
the mitral annulus or follows speckles
outside of the imaging plane.
7. LARS should not be used to assess
LVFP in patients with AF, significant
MR, heart transplant recipients,
patients with normal EF and
GLS > 18%, or suspected LA
stunning.

AF, Atrial fibrillation; AR, atrial reversal velocity in pulmonary veins; BBB, bundle branch block; CAD, coronary artery disease; DCM, dilated cardio-
myopathy; E@A, mitral velocity at the start of atrial contraction; HFrEF, HF with reduced EF; ICU, intensive care unit; LARS, LA reservoir strain; LAS,
LA strain; LASct, LA contractile strain; LVFP, LV filling pressure; Mitral DT, E-wave deceleration time; MV, mitral valve; PAEDP, PA end-diastolic pres-
sure; PH, pulmonary HTN; PN, pseudonormal; PS, pulmonary stenosis; PV, pulmonary vein; PVR, pulmonary vascular resistance; RA, right atrial;
ROI, region of interest; RVOT, RV outflow tract; RVSP, RV systolic pressure; UEA, ultrasound enhancing agent; Vp, early diastolic flow propagation
velocity; VTI, velocity time integral.
Journal of the American Society of Echocardiography Nagueh et al 551
Volume 38 Number 7

Table 5 Normal reference ranges for diastolic measurements by age category

Age, y

Diastolic measure 20-39 40-60 60-80

E wave, m/s 0.54 (0.52-0.57) to 1.11 (1.07-1.16) 0.47 (0.46-0.49) to 1.02 (0.99-1.05) 0.39 (0.37-0.42) to 0.92 (0.88-0.96)
A wave, m/s 0.24 (0.21-0.27) to 0.68 (0.63-0.72) 0.33 (0.32-0.35) to 0.82 (0.80-0.84) 0.43 (0.40-0.45) to 0.97 (0.93-1.00)
E/A ratio 0.88 (0.82-0.94) to 2.73 (2.66-2.81) 0.69 (0.66-0.73) to 2.07 (2.03-2.11) 0.50 (0.45-0.56) to 1.40 (1.34-1.47)
e0 lateral (cm/s) 9.9 (9.4-10.4) to 22.1 (21.5-22.8) 7.5 (7.3-7.8) to 17.5 (17.1-17.9) 5.2 (4.8-5.6) to 13.0 (12.4-13.5)
e0 septal (cm/s) 7.2 (6.8-7.7) to 16.4 (16.0-16.9) 5.7 (5.4-5.9) to 13.5 (13.2-13.8) 4.1 (3.7-4.5) to 10.6 (10.1-11.0)
e0 average (cm/s) 8.7 (8.2-9.2) to 19.1 (18.6-19.7) 6.7 (6.4-7.0) to 15.4 (15.1-15.7) 4.7 (4.3-5.1) to 11.7 (11.2-12.2)
E/e0 lateral 2.5 (2.0-3.0) to 6.3 (5.3-7.2) 3.6 (3.4-3.9) to 9.4 (8.9-10.0) 4.8 (4.5-5.0) to 12.6 (12.0-13.2)
E/e0 septal 4.0 (3.3-4.7) to 9.1 (8.2-9.9) 4.9 (4.6-5.3) to 12.1 (11.7-12.6) 5.9 (5.5-6.3) to 15.2 (14.7-15.7)
E/e0 average 4.0 (3.8-4.3) to 9.1 (8.5-9.7) 4.6 (4.4-4.8) to 11.5 (11.2-11.9) 5.2 (4.9-5.4) to 14.0 (13.4-14.5)
LAVi, mL/m2 12.1 (10.9-13.2) to 39.4 (34.6-44.2) 12.9 (12.2-13.5) to 38.3 (35.4-41.1) 13.7 (12.7-14.6) to 37.1 (33.0-41.3)
LAVi, Simpson, mL/m2 12.5 (12.0-13.0) to 41.9 (38.1-45.6) 13.3 (13.0-13.6) to 41.0 (38.5-43.4) 14.2 (13.7-14.6) to 40.0 (36.5-43.6)
2
LAVi, A-L, mL/m 8.9 (3.9-13.9) to 20.9 (12.9-28.8) 11.0 (8.9-13.0) to 27.1 (24.0-30.3) 13.0 (9.9-16.0) to 33.4 (28.6-38.2)
TR velocity, m/s 1.3 (1.1-1.5) to 2.7 (2.6-2.7) 1.5 (1.4-1.6) to 2.7 (2.7-2.7) 1.7 (1.5-1.8) to 2.8 (2.7-2.8)
LA strain, % 29.5 (27.6-31.3) to 63.2 (59.9-66.5) 26.8 (25.6-28.0) to 57.7 (55.6-59.9) 24.1 (22.2-26.0) to 52.3 (48.9-55.7)
LAS, TomTec, % 29.9 (27.0-32.9) to 60.5 (57.6-63.4) 27.5 (25.7-29.4) to 55.4 (53.6-57.2) 25.1 (22.6-27.6) to 50.3 (47.9-52.7)
LAS, EchoPAC, % 29.5 (27.9-31.1) to 64.9 (59.7-70.2) 25.3 (24.0-26.5) to 61.5 (57.4-65.6) 21.1 (18.7-23.4) to 58.1 (50.3-65.8)
Reference values are based of fifth and 95th percentile values derived from regression equations to fit summary data from persons free of cardio-
vascular disease or risk factors (see Supplemental Appendix). Values displayed are: fifth percentile limit (95% confidence limits) to 95th percentile
limit (95% confidence limit).

Summary-level data were extracted from publications of several velocities but not for E/A ratio or the other measurements assessed.
population-based studies reporting the distribution of diastolic For LAVi, differences were also observed by method of measure-
function measurements among subjects known to be free of ment: biplane method of disks vs area-length method
cardiovascular diseases within specified age ranges and, when (Supplemental Figures 27 and 28). For LARS, differences were
available, stratified by sex. These studies generally excluded indi- noted on the basis of the speckle-tracking strain software vendor
viduals with known cardiovascular disease (e.g., coronary artery used (Supplemental Figures 31 and 32).
disease, atrial fibrillation, HF) or cardiovascular risk factors (e.g.,
hypertension, diabetes, obesity, renal dysfunction). Not all mea-
surements were available in all studies. Using summary data on 4. REFERENCE RANGES COMPARED WITH PROGNOSTIC
values by age group, we identified the normal ranges for diastolic VALUES FOR DIASTOLIC FUNCTION MEASUREMENTS
measurements on the basis of their fifth and 95th percentile limits.
Details of the data extraction, harmonization, and statistical anal- The values provided in this section are based on an estimate of
ysis used to derive these limits are provided in the Supplemental the most extreme 10% of values from individuals of varying
Appendix. A list of studies used for data extraction is provided ages free of known cardiovascular disease or traditional risk fac-
in Supplemental Table 2. tors. It is important to recognize that prognostically relevant alter-
Supplemental Figures 5 to 16 display the resulting estimated fifth ations in diastolic measurements can occur within the range of
and 95th percentile limits of diastolic measurements by age and normal values defined in this way. Among >5,700 older adults
sex, while Table 5 provides the estimated normal ranges for each dia- (>65 years of age) in the ARIC (Atherosclerosis Risk in
stolic measurement by decade of life. Expanded data are provided in Communities) study, increases in the incidence of HF or death
the Supplemental Appendix. Supplemental Table 3 provides fifth and were observed at values of septal e0 velocity < 6 cm/s (lower
95th percentile reference limits for each diastolic measurement by 10th percentile limit 4.6 cm/s) and lateral e0 velocity < 7 cm/s
age decade. Supplemental Figures 17 to 32 provide additional plots (lower 10th percentile limit 5.2 cm/s). Additionally, in this study
of the fifth and 95th percentile limits of each diastolic measure by E/e0 ratio and LAVi demonstrated monotonic and near linear as-
age, including 95% CIs and the associated scatterplot of the source sociations with incidence of HF or death, without evidence of a
data. clear threshold with respect to prognosis. Additional studies
Differences in normal ranges by age were observed for measure- with greater power and inclusion of individuals of broader age
ments on the basis of transmitral flow velocities (E velocity, A veloc- range will be necessary to determine if similar patterns are seen
ity, E/A ratio), tissue Doppler imaging (TDI) early diastolic velocity in younger individuals, and to refine estimates of prognostically
(e0 ), E/e0 ratio, tricuspid regurgitation (TR) peak velocity, and LA vol- relevant thresholds for diastolic measurements. As noted above,
ume (maximum LA volume index [LAVi]) and function (LA reser- these data also do not address ‘‘optimal’’ values for these diastolic
voir strain [LARS]; P < .001 for all). Statistical differences in fifth measurements, which may be obscured by their population-level
and 95th percentile limits were observed by sex for E and A age-related changes.
552 Nagueh et al Journal of the American Society of Echocardiography
July 2025

Key Points
Supplemental Appendix for technical aspects of acquisition and mea-
1. Numerous studies have demonstrated the association between surement).
age and echocardiographic measurements of LV diastolic func-
tion.
2. Normal ranges are not necessarily equivalent to ‘‘optimal’’ 8. LA STRAIN
values, as the aging process itself may affect diastolic function.
3. E/e0 ratio and LAVi have near linear associations with inci- LA strain has emerged as a useful parameter for estimating LV filling
dence of HF or death, without evidence of a clear threshold pressures. Additional details pertaining to acquisition and measure-
with respect to prognosis. ments can be found in Supplemental Table 1 and Supplemental
Figures 3 and 4. Obtained by STE imaging, LA strain is available on
most ultrasound systems, and offline analysis is also possible.
Accuracy for LAP estimation is highest in patients with depressed
5. AGE-INDEPENDENT INDICES OF ELEVATED LV FILLING
LVEF.10 In addition, it is possible to divide LARS by the E/e0 ratio to yield
PRESSURES
a noninvasive index of LA stiffness.11 This index has the highest accu-
racy in comparison with other echocardiographic measurements in
Some indices of LV diastolic pressures are age independent. These
identifying patients with HFpEF12 and in identifying patients with
indices are changes in mitral inflow velocities with Valsalva maneuver,
HFpEF who are most likely to be hospitalized for HF management.13
and the difference in duration between pulmonary vein Ar velocity and
mitral A velocity.1 The Valsalva maneuver can help distinguish normal
LV filling from pseudonormal filling (and whether restrictive LV filling is
9. OTHER IMAGING MODALITIES FOR ASSESSMENT OF LV
reversible or not) because a decrease in E/A ratio of $50% is highly
DIASTOLIC FUNCTION
specific for increased LV filling pressures.1 The procedure should be
standardized by continuously recording mitral inflow using pulsed-
Radionuclide angiography, cardiac magnetic resonance (CMR), and
wave Doppler for 10 seconds during the strain phase of the maneuver.
cardiac computed tomography (CT) can be used to measure LV filling
An increase in Ar-A duration is consistent with increased LVEDP.
rates. In addition, CT and CMR have been applied to measure mitral
Pulmonary artery systolic pressure (PASP) identifies patients with
annular diastolic velocities, albeit with consistent underestimation in
increased LV filling pressures, provided pulmonary vascular disease
comparison with TDI. As a result, the other imaging modalities are
is excluded.1 The presence of a triphasic transmitral inflow profile
not routinely applied for evaluation of LV diastolic function.
with mid-diastolic flow (L-wave) velocity $ 50 cm/s occurs in patients
However, they can provide valuable insights into LV structural changes
with markedly delayed LV relaxation and increased LAP.1 A similar
including more precise measurement of LV mass and volumes. Using
pattern may be seen on mitral annular velocity recordings and with
CMR, it is possible to identify and to quantify the extent of replacement
mitral valve M-mode tracings.
and interstitial fibrosis.14,15 Importantly, the fibrosis burden adds incre-
mental value to echocardiographic assessment of LV diastolic function
6. LV STRUCTURE AND LA VOLUME AND FUNCTION for risk stratification of patients with, or at risk for, HF.16

In many patients with diastolic dysfunction, LV and LA structural

changes are present. LA enlargement, in the absence of chronic atrial 10. DEFINITION OF LV DIASTOLIC DYSFUNCTION USING
arrhythmia and mitral valve disease, is a marker of chronic elevation ECHOCARDIOGRAPHY
of LAP.1,7,9 However, this conclusion should be reached only after
exclusion of other reasons for LA enlargement, including anemia, heart LV diastolic function is assessed on the basis of LV relaxation and
transplant recipients with biatrial technique, hyperdynamic state, or myocardial stiffness. Patients with diastolic dysfunction can have
athletic status. Hence it is important to pay attention to the clinical normal or elevated LV filling pressures. The existing echocardio-
setting. graphic surrogates of LV relaxation and LV chamber stiffness are not
Pathologic LV hypertrophy is usually associated with increased LV perfect and thus patients with diastolic dysfunction can be missed
chamber stiffness. The presence of increased LV mass index and LA by echocardiography if one were to rely on a single variable.
enlargement are among the criteria for the diagnosis of HFpEF, as Accordingly, the working group recommends a combination of echo-
will be discussed later.5,7-9 cardiographic measurements for diagnosis of diastolic dysfunction.
These include an index of LV relaxation, echocardiographic variables
of reduced early diastolic LV filling relative to late diastolic filling, and
7. LV GLOBAL LONGITUDINAL STRAIN functional and structural changes related to elevated LAP and LV
diastolic pressures (Figure 2).
In some patients with diastolic dysfunction and normal EF, including For LVrelaxation, the best indices are mitral annular e0 velocity and LV
those with and those without HFpEF, LV global longitudinal function diastolic strain rate during isovolumic relaxation and early diastole.9 Of
is reduced.7-9 Abnormal LV longitudinal systolic function can be the three indices, e0 velocity has the highest feasibility and reproducibility
detected by using speckle-tracking to measure global longitudinal for daily application. Ideally, cutoff values that are associated with clini-
strain (GLS, Supplemental Figure 2). Although not an index of LV dia- cally relevant events should be sought and considered in the definition
stolic function, reduced LV GLS is one of the criteria used for HFpEF of normal function. However, data with respect to association with out-
evaluation and is associated with worse outcomes in many cardiovas- comes is available only for individuals >65 years of age and not younger
cular diseases associated with diastolic dysfunction, including cardio- subjects. Thus, the working group chose normal ranges for subjects
myopathies and left sided valvular heart disease (see the <65 years of age and prognostically low-risk features in subjects
Journal of the American Society of Echocardiography Nagueh et al 553
Volume 38 Number 7

STEP 1 STEP 2 Diastolic dysfunc on

Assess e’ as marker of impaired Assess markers of LA/LV present if:
LV relaxa on remodeling and elevated LAP •e’ reduced & 1 or more
markers in Step 2 present or
•e’ preserved, but 2 or more
•e’ septal ≤6 cm/s or •Average E/e’ > 14 markers in Step 2 present
•lateral ≤ 7 or •LARS ≤ 18%
•Average ≤ 6.5* •E/A≤0.8*, or ≥2
•LAVI > 34 ml/m2 ¶

* : can also consider age specific cutoff values to iden fy abnormally reduced e’ velocity or abnormally reduced E/A ra o
: a er excluding LA enlargement in athletes, or due to anemia, atrial fibrilla on or flu er, and mitral valve disease
¶ : another finding consistent with diastolic dysfunc on: LV mass index >95 g/m2 in women or 115 g/m2 in men, a er
exclusion of increased LV mass in athletes
Figure 2 Steps for diagnosing LV diastolic dysfunction.

>65 years of age. Of note, the cutoff values based on normal ranges in
subjects >65 years of age are very similar to the ones that are associated difference in duration between pulmonary vein Ar velocity
with worse outcomes. The recommended cutoffs for e0 velocity, which and mitral A velocity.
are based on reported normal values, are shown in Table 6. 4. Patients with diastolic dysfunction and HFpEF frequently have
Mitral annular e0 velocity is determined by LV relaxation, restoring abnormalities in LV structure and systolic function as well as in
forces, and lengthening load.9,17,18 On a cellular level, the rate of LV LA volume and function.
relaxation reflects the decay of active force developed during systole. 5. LV filling rates and LA volume and function can be assessed us-
The restoring forces account for diastolic suction and can be repre-
ing cardiac CT and CMR. CMR determination of replacement
sented by the behavior of an elastic spring that is compressed to a
and interstitial fibrosis provides incremental prognostic value
dimension less than its resting length during systole, and recoils
back to the resting length during diastole when the compression is to the echocardiographic assessment of LV diastolic function.
released. The lengthening load is the pressure in the left atrium at
the mitral valve opening, which ‘‘pushes’’ blood into, and thereby
lengthens the LV. The dependence of e0 on LAP is most prominent 11. ALGORITHM FOR ESTIMATION OF MEAN LAP AT REST
in the presence of normal LV relaxation, whereas the effect of LAP
on e0 velocity is absent or reduced in the presence of diastolic dysfunc- Echocardiography can estimate mean LAP using several parameters.
tion.17 This is due to the delay in e0 velocity such that it occurs when During echocardiographic determination of diastolic function and
LV pressure equals or exceeds LAP.19,20 filling pressure, the patient’s rhythm, heart rate, and blood pressure
For surrogates of LAP, E/A and E/e0 ratios and LARS are the recom- should be recorded as they affect both LV diastolic function and
mended measurements. For structural surrogates, LAVi and LV mass in- the Doppler indices used in the algorithm (Figure 3). No single
dex are the recommended indices. Given the direct dependence of e0 approach can estimate mean LAP in all clinical situations. Figure 3
velocity on LV relaxation, it is the first index in the algorithm. If abnor- presents the validated main algorithm (21), irrespective of LVEF, in
mally reduced, only one additional variable is needed from the func- a practical approach that can be applied to most patients. The algo-
tional and structural variables discussed above. If e0 velocity is not rithm in Figure 3 applies to all patients in sinus rhythm except for cases
reduced, then two abnormal variables as shown in Figure 2 are needed described in the special population sections. In subsequent sections,
to diagnose the presence of diastolic dysfunction. A reduced E/A ratio specific recommendations for special groups are discussed.
for age is due to impaired LVrelaxation leading to reduced early diastolic In clinical studies, LAVi is often inconsistent with other indices of
LV filling relative to late diastolic LV filling, whereas E/A $ 2, average E/ LAP. True LA volume is often overlooked, and in some patients,
e0 > 14, and LARS < 18% are consistent with elevated LAP. this measurement is technically challenging. Furthermore, the correla-
tion between LAP and LA volumes is not strong and LA volumes fail
to track changes in LAP.2,21 In addition, there are other causes for LA
Key Points enlargement including anemia, athletic heart, high cardiac output
states, atrial arrhythmias, and mitral valve disease. Therefore, unlike
1. LV diastolic dysfunction is identified on the basis of mitral
annulus e0 velocity measurements, reduced early diastolic LV
filling relative to late diastolic filling, and structural and func- Table 6 Mitral annular e0 velocity values for diagnosis of
tional surrogates of LAP and LV diastolic pressures. impaired LV relaxation
2. Septal e0 velocity < 6 cm/s, lateral e0 velocity < 7 cm/s, or
average e0 velocity < 6.5 cm/s indicates abnormal LV relaxa- 20-39 y 40-65 y >65 y

tion irrespective of age. 1. Septal e0 , cm/s <7 <6 <6

3. Age-independent indices of LV filling pressure include changes 2. Lateral e0 , cm/s <10 <8 <7
in mitral inflow velocities with the Valsalva maneuver and the 3. Average e0 , cm/s <9 <7 <6.5
554 Nagueh et al Journal of the American Society of Echocardiography
July 2025

Except in
LV Diastolic Function Grading & LAP Estimation MAC, MR, MS¶
Atrial Fibrillation
LVAD
1. Reduced e’ velocity: septal ≤6 or lateral ≤ 7 or average ≤ 6.5 cm/s * Non-cardiac PH
2. Increased E/e’: septal ≥ 15 or lateral ≥ 13 or average ≥ 14 HTX
Pericardial
3. Increased TR velocity ≥ 2.8 m/s or PASP ≥ 35 mm Hg constriction

All normal Reduced e’ only Increased TR/PASP only or 3 of the above

Increased E/e’ only or
Any 2 abnormal variables

E/A≤ 0.8 E/A > 0.8

Pulmonary Vein S/D ≤0.67 or
LARS ≤ 18% or LAVi > 34 mL/m2
None ≥1 present
Normal LAP Alternatively Increased LAP
IVRT ≤ 70 ms

If none available or reliable use

Supplemental methodsȚ E/A < 2 E/A ≥ 2

Normal DF Grade 1 If symptomatic Diastolic Exercise Grade 2 Grade 3

Echo (Mild/Mod ↑ LAP) (Marked ↑ LAP)

Figure 3 Algorithm for estimation of mean LAP for patients in sinus rhythm and who do not have severe primary MR, any degree of
mitral stenosis (MS), or moderate or severe MAC. The algorithm should also not be applied to patients in atrial fibrillation, heart trans-
plant (HTX) recipients, noncardiac PH, pericardial constriction or LV assist device (LVAD). *For annular e0 velocity, age-adjusted lower
limits of normal values shown in Table 6 can be applied in place of the values shown in this figure. {The algorithm should also not be
applied to patients with mitral valve repair, mitral valve replacement, or mitral-transcatheter edge-to-edge repair. DF, Diastolic func-
tion; Ț, PR end-diastolic velocity $ 2m/s, PA diastolic pressure $ 16 mm Hg, mitral inflow L-wave velocity $ 50 cm/s, Ar-A
duration > 30 ms, and/or a decrease in mitral E/A ratio of $50% with Valsalva maneuver.

the 2016 guidelines, the algorithm in Figure 3 includes LAVi in the and LAVi do not meet the cutoff threshold for elevated LAP, then
second stage, if needed. LAP is likely normal. In symptomatic patients, diastolic exercise echo-
The algorithm in Figure 3 begins with mitral annular e0 velocity, E/ cardiography is recommended when LAP at rest is normal to increase
0
e ratio, and peak TR velocity or PASP. When the right atrial pressure the sensitivity of detecting patients with HFpEF when there are indi-
(RAP) can be estimated, decisions about LAP should be based on cators of abnormal LA/LV morphology and/or function (Figure 3).
whether PASP is $35 mm Hg or not. If RAP cannot be estimated, LARS can be readily obtained in most patients with satisfactory 2D
then peak TR velocity should be relied on, and a peak apical views (see the Supplemental Appendix for technical details). If
velocity $ 2.8 m/s in the absence of pulmonary parenchymal or LARS is #18%, then LAP is elevated. LARS # 18% has high specificity
vascular disease supports the conclusion that LAP is elevated. The but can have low sensitivity in patients with normal LVEF for detecting
TR jet should be obtained from multiple windows and the use of elevated LAP.3,10,22 On the other hand, relying on LARS cutoff values
intravenous saline or ultrasound enhancing agents is encouraged in of 19% to 24%, which is in the low normal range,10 leads to higher
cases with incomplete TR jet, to enhance the jet and obtain a com- sensitivity and lower specificity for detecting elevated LAP.
plete TR envelope. Caution should be exercised to avoid measuring For pulmonary venous flow velocities, previous work showed that
the peak velocity from blooming artifacts. a systolic filling fraction (systolic velocity-time integral [VTI]/systolic
For estimation of LAP at rest, the presence of all three findings of VTI + diastolic VTI) of #40% has good accuracy in identifying pa-
reduced e0 velocity, increased E/e0 ratio, and PASP $ 35 mm Hg or tients with elevated mean LAP.1 The corresponding value for pulmo-
peak TR velocity $ 2.8 m/s when RAP cannot be estimated, supports nary vein systolic velocity–to–diastolic velocity ratio of #0.67 is equal
the conclusion of elevated LAP. Subsequently, diastolic function is to #40% systolic filling fraction and is therefore recommended. The
classified as grade 2 if the E/A ratio is <2 or grade 3 if the E/A ratio ratio is most reliable in patients with LV systolic dysfunction and
is $2. If all three variables do not meet the cutoff values for elevated should not be considered in normal subjects with normal echocardio-
LAP in Figure 3, then LAP is normal, and the patient has normal dia- graphic results, when the ratio can be # 0.67. In patients with normal
stolic function. LVEF, the ratio can be >0.67 despite elevated LV filling pressures. In
Grade 1 diastolic dysfunction is present when e0 velocity is reduced these patients, confirmation should be sought with IVRT, and if IVRT
with normal E/e0 ratio and normal PASP, along with E/A ratio # 0.8. is not available, then other parameters discussed in the ‘‘Supplemental
When e0 velocity is reduced and the E/A ratio is >0.8, then additional Parameters’’ section below should be analyzed to reach a conclusion
variables should be considered as LAP can be elevated in some of about LAP. IVRT # 70 ms is consistent with elevated LAP. If the find-
these patients. Likewise, if only PASP or only E/e0 ratio, or any two ings indicate elevated LAP, then grading diastolic function should be
of the three variables (e0 velocity, E/e0 ratio, and PASP) are consistent pursued on the basis of E/A ratio (Figure 3).
with elevated LAP, then additional variables should be evaluated.
The additional variables recommended are LARS, pulmonary vein Supplemental Parameters
systolic to diastolic velocities ratio, LAVi, or alternatively IVRT. If These can be relied on in the absence of the three parameters in the
LARS, pulmonary vein systolic to diastolic velocities ratio, IVRT, section above.
Journal of the American Society of Echocardiography Nagueh et al 555
Volume 38 Number 7

a. Pulmonary regurgitation (PR) end-diastolic velocity: An end-diastolic veloc- creases rapidly with exercise or even leg raising.23 Echocardiography
ity $2 m/s or pulmonary artery (PA) diastolic pressure $16 mm Hg is can determine the status of LV relaxation by measuring e0 velocity.9,17,18
consistent with elevated LAP in the absence of pulmonary disease. Ultra- In normal subjects, mitral e0 increases about 3 to 5 cm/s on average with
sound enhancing agents can facilitate recording of a complete PR jet. exercise,24 but in subjects with diastolic dysfunction, e0 velocity does
b. Mitral inflow L-wave velocity: This is a mid-diastolic velocity that occurs af-
not increase as much as in a normal subject, or not at all. In patients
ter the mitral E wave. It was initially described in normal healthy young in-
with diastolic dysfunction, mitral inflow E velocity increases with exer-
dividuals with bradycardia with peak velocity <40 cm/s. Later, it was also
noted in patients with marked delay in myocardial relaxation and increased cise and e0 does not change as much, such that E/e0 ratio increases.
filling pressure. Because of delayed relaxation, LV diastolic pressure falls af- Normal E/e0 values have been published for middle-age and young sub-
ter the initial rise from LV filling, resulting in a mid-diastolic LA-LV pressure jects using treadmill or supine bicycle exercise with remarkably similar
gradient, leading to blood flow into the left ventricle with peak values of 6 to 8 at rest and with exercise. It rarely becomes higher than
velocity $ 50 cm/s. Therefore, the presence of an L-wave velocity 10. Several studies have shown a good correlation between E/e0 ratio
$50 cm/s is an indicator of both impaired LV relaxation and elevated LAP. and invasively obtained PCWP, LAP, or LV mean diastolic pressure
c. Premature termination of mitral inflow before QRS complex and diastolic with variable levels of effort, including day-to-day activity as well as dur-
mitral regurgitation (MR): Diastolic MR is recognized as diastolic flow ing supine bicycle exercise in the catheterization laboratory.23,25-27
from the left ventricle to the left atrium and can be indicative of increased
LVEDP. It is analogous to the ‘‘B’’ bump on a mitral valve M-mode trace.
However, this can also happen when the electrocardiographic PR interval
is >200 ms with sufficient atrial relaxation before systole. Therefore, one
A. Indications
should be careful to exclude advanced atrioventricular (AV) block, atrial
flutter, and organized atrial fibrillation activity before attributing diastolic Diastolic stress testing is most valuable when resting echocardiography
MR to increased LV diastolic pressures. does not explain the symptoms of HF or exertional dyspnea. In general,
d. Pulmonary vein atrial reversal velocity: In the evolution of diastolic dysfunc- patients with completely normal diastolic function at rest with pre-
tion, there is an early stage when LVEDP is elevated but mean LAP is served e0 velocity need not undergo stress testing as it is highly unlikely
normal. This stage is diagnosed on the basis of a high amplitude that they will develop elevated filling pressures with exercise. Likewise,
(>35 cm/s) and long duration of pulmonary vein Ar velocity, along with patients with abnormal findings at baseline consistent with elevated LV
an Ar-A duration >30 ms.
filling pressures should not be referred for stress testing as the cardiac
e. Changes in mitral inflow with Valsalva: A decrease in E/A ratio of $50%
etiology for dyspnea is already established and their filling pressures
with the Valsalva maneuver is specific for elevated LAP.
will almost certainly increase further with exercise. The most appro-
The evaluation of LV diastolic function should not stop at the algo- priate patient population for diastolic exercise testing is the group of pa-
rithm above in symptomatic patients with grade 1 diastolic dysfunc- tients with grade 1 diastolic dysfunction, which indicates the presence
tion. These patients should be referred for diastolic exercise of delayed myocardial relaxation and normal mean LAP at rest. The
echocardiography. diagnostic evaluation of symptomatic patients with indeterminate dia-
stolic function or filling pressure at rest can also benefit from diastolic
stress testing. LARS has been reported to identify patients who are
12. REPORTING ON DIASTOLIC FUNCTION more likely to develop increased filling pressure with exercise.28

All echocardiographic reports should include an assessment of dia-

stolic function grading and filling pressure, indicating whether diastolic
function is normal or abnormal with grade 1, 2 (mild to moderate in- B. Performance
crease in LAP), or 3 (marked increase in LAP). If the interpreting cardi- Diastolic stress testing can be performed using either supine bicycle or
ologist is unable to determine the grade of diastolic dysfunction, the treadmill exercise echocardiography. The use of dobutamine stress
status of LV filling pressure should be mentioned in the report testing for the assessment of diastolic function is strongly discouraged.
(whether LAP is normal or elevated). An isolated increase in For supine bicycle exercise, there is sufficient time during each stage
LVEDP should be reported, as it predisposes patients to elevated of exercise to acquire 2D images, mitral inflow velocities, annular veloc-
mean LAP with exercise, faster heart rate, or increased afterload. ities, and peak TR velocity. For treadmill exercise, 2D images are ob-
Whenever possible, the report should mention whether a change tained first to assess wall motion for myocardial ischemia. Mitral
has occurred in LV diastolic function grade in comparison with previ- inflow velocities are usually fused during exercise, and it would be
ous studies. most helpful to obtain TR velocity first, immediately after wall motion
Measurements that should be included in the report are mitral analysis, followed by mitral inflow and annular velocities when E and A
inflow velocities, mitral annular e0 velocity, peak TR velocity, E/A ratio, velocities are not fused. When LV filling pressure is elevated with exer-
and average E/e0 ratio (unless only one side is acquired or satisfactory). cise, it usually remains elevated for several minutes, providing sufficient
LARS, pulmonary vein S/D ratio or systolic filling fraction, mitral inflow time to acquire Doppler velocities after acquisition of parasternal and
A duration, pulmonary vein Ar duration, and/or IVRT should be apical views for LV wall motion and volume analysis. As soon as mitral
included in the report if relied on to arrive at the conclusions pertaining E and A velocities are separated, mitral annular and mitral inflow veloc-
to diastolic function grade and mean LAP. ities (in this order) are acquired. In patients with atrial fibrillation, dia-
stolic function parameters can be obtained immediately after
recording 2D views. Recently, lung scanning to detect ‘‘B’’ lines, which
13. DIASTOLIC EXERCISE ECHOCARDIOGRAPHY
indicates water in the lung or pulmonary venous congestion, was intro-
duced as a part of diastolic exercise stress echocardiography.29 In some
LV filling pressure at rest may not be sufficient to evaluate a patient with
patients, similar information can be obtained with simple leg raising for
dyspnea that happens mostly with exertion. Importantly, up to 50% of
1 minute using a regular echocardiographic examination table.
patients with HFpEF have normal LV filling pressure at rest that in-
However, this simple method lacks sufficient sensitivity.30
556 Nagueh et al Journal of the American Society of Echocardiography
July 2025

C. Interpretation learning models to automate view classification, view segmentation,

The results are considered definitely abnormal, indicating increased LV and input of the machine-derived measurements into a rule-based de-
filling pressure when the following conditions are met: average E/e0 cision tree algorithm for severity classification using the 2016 guide-
ratio $ 14 or septal E/e0 ratio $ 15 and peak TR velocity > 3.2 m/s. lines as a reference.33-35 Single-view approaches have also been
The test indicates normal filling pressure when average (or septal) E/e0 studied. These often involve the segmentation of LA and LV volumes,
is <10 and peak TR velocity is <2.8 m/s. HFpEF diagnosis is not sup- measurement of LVEF and longitudinal strain, and training the model
ported by an E/e0 ratio <14, even with an increase in peak TR velocity to identify which combination of LA and LV metrics provides the most
to >3.2 m/s. HFpEF diagnosis is considered likely when E/e0 ratio is accurate grading of diastolic function.33,35-37 Other single-view ap-
>14 and the peak TR velocity is >2.8 but <3.2 m/s. One should be proaches use a deep-learning neural network to train models to iden-
cautious in drawing conclusions on the basis of an isolated increase in tify and grade diastolic function directly from the 2D apical four-
exercise peak TR velocity, as normal subjects can have significant in- chamber videos without segmentation.33-35 All these approaches
creases in peak TR velocity related to increased pulmonary blood have used the 2016 guideline as a reference, and thus, they are
flow. After analysis of >14,000 diastolic exercise echocardiograms, it similarly prone to grading several cases as indeterminate.
was found that 17% of patients develop increased LV filling pressure Machine learning models based on clinical outcomes38,39 offer a
with exercise, while 28% have evidence of myocardial ischemia.31 more evidence-based approach to optimizing the classification of dia-
Patients with increased filling pressure after exercise but no ischemia stolic function with both reduction of ‘‘indeterminate’’ cases and
have a worse prognosis than patients with isolated ischemia.31 In addi- improvement of the clinical relevance of diastolic function classifica-
tion, PASP > 50 mm Hg or TR velocity > 3.2 m/s portends a worse tion. Recognizing that diastolic function represents a continuum
outcome,31 hence the recommendation for this cutoff value. An inva- from normal function to the most severe diastolic dysfunction grade,
sive hemodynamic investigation, including the use of exercise, may be the ability to accurately assess where a patient is in that continuum is
necessary when exercise echocardiographic assessment is negative or important for prognostication and management purposes. To that
indeterminate in a patient with a clinical presentation concerning for end, a continuous diastolic function score has been developed on
HFpEF. the basis of inputs of traditional diastolic parameters on echocardiogra-
phy, even for cases that would have been labeled indeterminate ac-
cording to the reference guidelines.32 Further validation of this
Key Points model and other models using invasive hemodynamics and clinical
1. Diastolic exercise stress testing is indicated in patients with dys- outcomes is needed to establish precision, reproducibility, and clinical
relevance. Vendor reporting systems based on AI that incorporate
pnea and grade 1 diastolic dysfunction at rest and in patients
several echocardiographic measurements to arrive at a conclusion
with indeterminate LV filling pressure at rest. It is performed us-
regarding LV diastolic function are needed as they have the potential
ing supine bicycle or treadmill exercise stress testing. of increasing the incorporation of diastolic function assessment results
2. At rest, mitral E and e0 velocities should be recorded, along in clinical reports.
with the peak velocity of TR, using agitated saline if needed.
The same parameters are recorded during exercise or 1 to 2 mi-
Key Points
nutes after termination of exercise when E and A velocities are
not merged, because increased filling pressures usually persist 1. The application of AI can enhance the noninvasive assessment
for a few minutes. of LV diastolic function by automating parameter measure-
3. The result is considered positive when during exercise, average ment and severity grading.
E/e0 ratio is $14 (or septal E/e0 ratio is $15) and peak TR ve- 2. There are multiple- and single-view approaches. Single-view
locity is $3.2 m/s. approaches involve the segmentation of LA and LV volumes,
measurement of LVEF and longitudinal strain, and training
the model to identify which combination of LA and LV metrics
14. APPLICATION OF AI TO THE ASSESSMENT OF LV provides the most accurate grading of diastolic function.
DIASTOLIC FUNCTION 3. Machine learning models based on clinical outcomes offer a
more evidence-based approach to optimize the classification
Given the complexity of diastolic function assessment, both in terms of of diastolic function.
the need to measure numerous parameters and the requirement of 4. Validation of AI models using invasive hemodynamics and
data integration to grade the severity, the application of AI to streamline clinical outcomes is needed to establish their precision, repro-
these processes through automating both parameter measurement and
ducibility, and clinical relevance.
severity grading would be highly valuable. This type of application has
been demonstrated in few studies that used a rule-based algorithm to
grade diastolic function on the basis of measurements obtained by ex-
perts or by machine learning models, including some studies that used 15. ASSESSMENT OF LV DIASTOLIC FUNCTION AND
electrocardiographic findings relying on data from close to 100,000 ESTIMATION OF LV FILLING PRESSURES IN SPECIAL
paired echocardiograms and electrocardiograms.32,33 Interestingly, POPULATIONS
electrocardiographic findings showed good accuracy in identifying
the three grades of diastolic dysfunction, with similar prognostic value In the following sections we discuss the pathophysiology of disorders
to echocardiography.33 with abnormal cardiac structure, valve disease, and atrial arrhythmias,
Several approaches have been tested for AI-assisted diastolic func- which modify the relationship between indices of diastolic function
tion assessment. Multiple-view approaches involve the use of machine and LV filling pressure (Table 740-58).
Journal of the American Society of Echocardiography Nagueh et al 557
Volume 38 Number 7

Table 7 Indicators of elevated LV filling pressures in special populations

Disease Echocardiographic measurements indicative of elevated LV filling pressure

1. Atrial fibrillation40-42 1. DT < 160 ms in patients with depressed LVEF

2. Peak acceleration rate of mitral E velocity ($1,900 cm/s2)
3. IVRT (#65 ms)
4. DT of pulmonary venous diastolic velocity (#220 ms)
5. E/Vp ratio ($1.4)
6. Septal E/e0 ratio ($11)
7. Peak TR velocity > 2.8 m/s
2. Sinus tachycardia43,44 1. Predominant early LV filling pattern with depressed LVEF
2. IVRT # 70 ms is specific (79%)
3. Pulmonary vein systolic filling fraction #40% is specific (88%)
4. Average E/e0 ratio > 14 (high specificity but low sensitivity)
5. When E and A velocities are partially or completely fused, the presence of a compensatory period after premature
beats often leads to separation of E and A velocities which can be used for assessment of diastolic function
3. HCM45 1. Average E/e0 (>14)
2. Ar-A ($30 ms)
3. Peak TR velocity (>2.8 m/s)
4. LA maximum volume index (>34 mL/m2)
4. Restrictive 1. Average E/e0 (>14)
cardiomyopathy46-49 2. DT < 140 ms*
3. E/A ratio > 2.5*
4. IVRT < 50 ms*
5. PH50,51 1. E/A $ 2 favors postcapillary PH
2. E/A # 0.8 favors precapillary PH
3. When E/A ratio is >0.8 but <2, lateral E/e0 ratio > 13, LA maximum volume index > 34 mL/m2, and LARS < 18%
favor the diagnosis of postcapillary PH.
6. Mitral stenosis52 1. IVRT < 60 ms*
2. Mitral A peak velocity > 1.5 m/s
3. IVRT/TE-e0 ratio < 4.2
7. MR52-54 1. IVRT < 60 ms*
2. Ar-A ($30 ms)
3. IVRT/TE-e0 ratio < 5.6
4. Average E/e0 ratio > 14 in patients with depressed EF
8. Moderate/severe 1. LV filling pressure normal when mitral E/A ratio is <0.8
MAC55 2. LV filling pressure elevated when mitral E/A ratio is >1.8
3. E/A ratio >0.8 but <1.8, IVRT should be measured. LV filling pressure normal when IVRT is $80 ms, whereas it is
elevated if IVRT <80 ms.
9. LV assist device56,57 1. E/A ratio > 2
2. RAP > 10 mm Hg
3. PASP > 40 mm Hg
4. Average E/e0 ratio > 14 or septal E/e0 ratio $ 15
5. LA maximum volume index > 33 mL/m2
6. Interatrial septum position†
10. Cardiac transplant 1. Average E/e0 ratio < 7 denotes normal LV filling pressures
recipients58 2. Average E/e0 ratio > 14 denotes elevated LV filling pressures
3. For E/e0 ratio > 7 and <14, SRIVR, from all three apical views, is measured, and the ratio of mitral E velocity to SRIVR
is derived. A ratio # 200 cm is consistent with normal LV filling pressure, but >200 cm denotes elevated LV filling
pressure.
4. In patients in whom SRIVR cannot be measured, peak TR velocity is relied on. Peak TR velocity # 2.8 m/s is
consistent with normal LV filling pressures, but >2.8 m/s denotes elevated LV filling pressures.

*Specific but not sensitive indicators of elevated LV filling pressure.

†
LAP = RAP if the interatrial septum position is neutral. If septum bulges to right, then LAP is 5 mm Hg higher than RAP. If septum bulges to left, then
LAP is 5 mm Hg lower than RAP.

A. Assessment of LV Diastolic Function in Patients With with other etiologies of LV inflow obstruction, such as a large LA tumor ob-
Valvular Heart Disease structing LV inflow, or cor triatriatum. A short IVRT (<60 ms) and mitral A
velocity > 1.5 m/s are usual findings in patients with increased LAP. The
a. Mitral stenosis: Usually, patients with mitral stenosis have normal or time interval between the onset of mitral E and annular e0 velocities (TE-
reduced LV diastolic pressures, except for the rare occurrence of coexisting e0 ) can be applied to estimate LV filling pressures in patients with mitral valve
myocardial disease. The same hemodynamic findings are present in patients disease. In the presence of impaired LV relaxation, e0 velocity is reduced and
558 Nagueh et al Journal of the American Society of Echocardiography
July 2025

delayed such that it occurs at the second LA-LV pressure crossover point. In d. LAP estimation after TEER: Flow across an iatrogenic atrial septal
comparison, mitral E velocity occurs earlier with elevated LAP. Thus, TE-e0 is defect after TEER can be detected from multiple views using color
prolonged and can correct for the effect of LV relaxation on IVRT. IVRT/TE- Doppler. When flow is aligned with the ultrasound beam (usually
e0 correlates well (inversely) with mean PCWP and LAP in patients with from the subcostal view), CW Doppler can be used to record the
mitral stenosis.1,52 However, E/e0 ratio is not useful. peak flow velocity across the interatrial septum in late diastole in pa-
b. MR: Primary MR leads to LA and LV enlargement and an increase in the tients in sinus rhythm, corresponding to the LA ‘‘a’’ wave, and the
compliance of both chambers, which attenuates the increase in LAP. Later, peak velocity at end-systole corresponding to the LA ‘‘v’’ wave. The
with increased LA stiffness, mean LAP and PA pressures increase, which is modified Bernoulli equation is then used to estimate the corresponding
related to MR, not LV myocardial disease. However, with LV diastolic LAP as follows:
dysfunction, the increased LV diastolic pressures contribute to the increase
in LAP. The sequence is opposite to that seen in primary myocardial dis- LAP (mm Hg) = 4V2 + RAP,
ease such as dilated cardiomyopathy, which leads to increased LV diastolic where V is the peak velocity of flow across the interatrial septum in
pressures initially and later to functional MR. Therefore, in patients with meters per second. RAP is estimated on the basis of inferior vena cava
secondary MR, echocardiographic correlates of increased filling pressures diameter, its change with respiration and sniffing, and hepatic vein
reflect the combination of both myocardial and valvular disorders. Moder- flow.1 When interatrial left-to-right shunt velocity is >1.7 m/s, LAP
ate and severe MR usually lead to an increase in mitral E velocity and a is usually elevated. However, in patients with biventricular HF, the in-
decrease in pulmonary vein systolic velocity, and the S-to-D ratio. In severe teratrial velocity may not be helpful in estimating LAP. The aforemen-
MR, systolic pulmonary venous flow reversal can be seen. Thus, MR can tioned approach can also be applied in other situations with residual
induce changes in transmitral and pulmonary venous flow patterns resem-
shunt across the interatrial septum after ablation procedures for atrial
bling advanced LV diastolic dysfunction. Irrespective of MR severity, Ar-A
duration remains a good indicator of increased LVEDP.53 A continuous-
fibrillation, LA appendage occlusion, other procedures involving
wave (CW) Doppler MR velocity profile showing early peaking and transseptal access, and also congenital atrial septal defects.
reduced late LV-LAP gradient is a highly specific, though not sensitive,
sign of increased LAP. The utility of E/e0 ratio in predicting LV filling pres- e. Aortic stenosis and aortic regurgitation: The presence of diastolic dysfunc-
sures in the setting of moderate or severe MR is more complex. In patients tion, including elevated LV filling pressures, predicts worse outcomes in pa-
with depressed EF, an increased E/e0 ratio has a direct significant relation tients with moderate or severe aortic stenosis (AS).61 Furthermore, the
with LAP and predicts hospitalizations and mortality.54 However, E/e0 ra- improvement of diastolic dysfunction after surgical and transcatheter aortic
tio does not appear to be useful in patients with primary MR and normal valve replacement is associated with lower rates of adverse outcome
EF,59 though some investigators have noted a good correlation between E/ events.62,63 It is therefore recommended that assessment of LV diastolic
e0 ratio and mean PCWP as well as PASP in this population. Likewise, function status be carried out in all patients with AS and be included in
LARS has been shown to have no significant relationship with LA ‘‘v’’ the report. There are usually no major challenges to the application of the
wave or mean pressure in patients with significant MR and could not guidelines in patients with AS. If moderate or severe MAC is present,
detect changes in ‘‘v’’ pressure after transcatheter edge-to-edge repair then the recommendations pertaining to patients with significant MAC
(TEER).60 In comparison, IVRT/TE-e0 ratio correlates reasonably well should be applied.
with mean PCWP, regardless of EF. An IVRT/TE-e0 ratio <3 readily predicts
For patients with severe aortic regurgitation (AR), the AR jet can
PCWP > 15 mm Hg in this patient population.52 In patients with atrial
interfere with the recording of mitral inflow velocities and careful
fibrillation and MR, the use of matched intervals (necessitating acquisition
of many cycles) is necessary. It is challenging to assess LV relaxation and LV positioning of the sample volume is needed to avoid contamination
filling pressures after mitral valve repair or replacement, although time in- with the AR jet. In severe acute AR, the presence of abbreviated LV
tervals and PA pressures could be of value for drawing inferences about LV diastolic filling period, premature closure of the mitral valve, and dia-
filling pressures.52 stolic MR indicate elevated LV filling pressures. In chronic severe AR,
c. Mitral annular calcification (MAC): MAC frequently accompanies hyper- the mitral inflow pattern often shows predominant early diastolic
tensive heart disease, aortic sclerosis, coronary artery disease, and chronic filling with short deceleration time of mitral E velocity,64 but there
kidney disease and is prevalent in elderly patients. In patients with moderate are limited data on the accuracy of estimation of LV filling pressures
to severe MAC, mitral orifice area is decreased, leading to increased dia- in patients with chronic severe AR. In patients with AR, the presence
stolic transmitral velocities, while lateral e0 velocity may be decreased
of LA enlargement, average E/e0 > 14, LARS < 18%, and peak TR
because of reduced annular excursion. Thus, an increase in E/e0 ratio can
velocity > 2.8 m/s support the presence of increased LV filling
occur because of the mechanical effects of MAC. Although data are limited,
LV filling pressures are usually normal when mitral E/A ratio is <0.8 but pressures.
elevated when the ratio is >1.8. When E/A is 0.8 to 1.8, IVRT should be
measured (Figure 4). LV filling pressure is usually normal when IVRT is Key Points
$80 ms, whereas it is elevated if IVRT is <80 ms.55
1. Mitral stenosis renders assessment of LV diastolic function more
challenging, but IVRT, TE-e0 , and mitral E and A velocities can be
Mitral E/A Ra o
of value in the semiquantitative prediction of mean LAP.
2. Ar-A duration > 30 ms and IVRT/TE-e0 < 5.6 may be applied
for prediction of LV filling pressures in patients with MR and
<0.8 0.8-1.8 >1.8 normal LVEF, whereas E/e0 ratio may be considered only in pa-
tients with MR and depressed EF.
IVRT 3. In patients with moderate and severe MAC, LV filling pressures
are usually normal when mitral E/A ratio is <0.8 but elevated
≥80 ms <80 ms
when the ratio is >1.8. When E/A is 0.8 to 1.8, IVRT should be
Normal LA pressure Elevated LA pressure
measured. LV filling pressure is usually normal when IVRT is
Figure 4 Algorithm for estimation of mean LAP in patients with $80 ms, whereas it is elevated if IVRT <80 ms.
moderate or severe MAC.
Journal of the American Society of Echocardiography Nagueh et al 559
Volume 38 Number 7

some studies noted that they were reduced 1 year after transplantation
4. The guidelines in patients without valvular heart disease can be compared with a normal population. LV diastolic dysfunction has often
applied to patients with AS, irrespective of severity of valvular been described as a sensitive sign of early graft rejection as myocardial
stenosis. This excludes patients with moderate or severe MAC. edema causes increased diastolic stiffness and filling pressures in the
5. Flow across atrial septal defect (congenital or iatrogenic after presence of a normal EF.1 Later, myocardial fibrosis seen with chronic
TEER and other procedures) can be recorded by CW Doppler. graft rejection can lead to a restrictive LV filling pattern and markedly
The modified Bernoulli equation is used to estimate the corre- reduced annular velocities. However, no single diastolic parameter is
sponding LAP as LAP (mm Hg) = 4V2 + RAP, where V is the reliable enough to predict graft rejection.
A recently validated simplified approach may be used in transplant
peak velocity of flow across the interatrial septum in meters
recipient patients in sinus rhythm, whereby average E/e0 < 7 denotes
per second.
normal LV filling pressures, and E/e0 > 14 denotes elevated LV filling
6. In patients with acute or chronic severe AR, premature closure pressures.58 For average E/e0 between 7 and 14, LV strain rate during
of the mitral valve, diastolic MR, LA enlargement, average E/ the isovolumic relaxation period (SRIVR), from all 3 apical views, is
e0 > 14, and TR peak velocity > 2.8 m/s are consistent with measured, and the ratio of mitral E velocity to SRIVR is derived. A ratio
elevated LV filling pressures. #200 cm is consistent with normal LV filling pressures, but >200 cm
denotes elevated LV filling pressures (Figure 5). In patients in whom
SRIVR cannot be measured, peak TR velocity is used. A peak TR ve-
B. Heart Transplant Recipients locity #2.8 m/s is consistent with normal LV filling pressures, but
>2.8 m/s denotes elevated LV filling pressures.58
The transplanted heart is affected by many factors that influence LV
diastolic function, making the interpretation of diastolic function
more difficult. The donor heart is denervated, leading to sinus tachy- Key Points
cardia, which can lead to fusion of mitral E and A velocities.
Pulmonary venous flow is usually not helpful in estimating LV filling 1. Predominant early diastolic filling in patients with preserved EF
pressures as the S-to-D ratio is reduced given the young age of is a common finding after heart transplantation and is observed
most donor hearts. in patients with normal LV diastolic function as donor hearts
For mitral inflow, an E/A ratio $2 in patients with preserved EF is a are usually obtained from healthy individuals.
common finding after heart transplantation and can be observed in pa- 2. No single diastolic parameter is reliable enough to predict graft
tients with normal LV diastolic function as donor hearts are usually ob- rejection.
tained from healthy individuals. It is most pronounced in the early
3. A simplified approach may be used whereby average E/e0 < 7
weeks after surgery, and in some patients is likely related to myocardial
denotes normal LV filling pressures, but E/e0 > 14 supports the
edema. This inflow pattern changes at follow-up. Although LV diastolic
pressures can be normal at rest, a large increase in LV minimal pressure conclusion of elevated LV filling pressures. For E/e0 between 7
and LVEDP has been noted during exercise in heart transplant recipi- and 14, SRIVR, from all three apical views, is measured, and the
ents with diastolic dysfunction.65 This is due to lack of shortening or ratio of mitral E velocity to SRIVR is derived. A ratio # 200 cm
prolongation of the time constant of LV relaxation with exercise in addi- is consistent with normal LV filling pressures, but >200 cm de-
tion to increased LV chamber stiffness.65 Mitral annular e0 velocity is notes elevated LV filling pressures. In patients in whom SRIVR
influenced by the pronounced translational motion of the heart and cannot be measured, peak TR velocity is used. A peak TR ve-
may not detect the changes in LV relaxation status with exercise. Of locity #2.8 m/s is consistent with normal LV filling pressures,
note, myocardial tissue velocities are lowest early after surgery and but >2.8 m/s denotes elevated LV filling pressures.
tend to increase during the following weeks and months, though

Figure 5 Algorithm for estimation of mean LAP in heart transplant recipients in sinus rhythm.
560 Nagueh et al Journal of the American Society of Echocardiography
July 2025

C. Pulmonary Hypertension Abnormalities of the cardiac conduction system due to disease, aging,
Evaluation of LV filling pressure in patients with moderate or severe drugs, or pacing can adversely affect AV synchrony and synchronous
pulmonary hypertension (PH) group I and groups III to V requires a LV contraction and relaxation, which may reduce functional capacity
different approach than in patients with PH due to left-sided heart dis- by altering LV systolic and diastolic function, and thus the variables
ease or group II PH.1 Because PH is associated with elevated right ven- used to assess diastolic function. If the PR interval is too short, atrial
tricular (RV) systolic and diastolic pressures, there is often septal filling is terminated early by ventricular contraction, thus reducing
flattening, and therefore lateral E/e0 rather than the average of septal mitral A duration, LV end-diastolic volume, and cardiac output. A
and lateral E/e0 should be used to evaluate LV filling pressure. In pa- first-degree AV block of 200 to 280 ms is usually well tolerated if
tients with PH, lateral E/e0 > 13 reflects elevated LV filling pressure, LVEF and heart rate are normal. However, in patients with shortened
and values <8 are consistent with normal filling pressure.50 In a recent diastolic filling periods due to markedly impaired LV relaxation, faster
study looking at noninvasive evaluation of LV filling pressure in PH, heart rates, bundle branch block, or ventricular pacing, a first-degree
these observations were confirmed, and it was observed that E/e0 in AV block of >280 ms usually results in fusion of E and A velocities.
the range of 8 to 13 had weak or no association with LV filling pres- If atrial contraction occurs before early diastolic mitral flow velocity
sure.51 Therefore, the use of lateral E/e0 as a marker of LV filling pres- has decreased toward zero (defined as #20 cm/s), E/A is reduced
sure can result in indeterminate cases. However, the combination of because of a higher A-wave velocity.66 This fusion of early and late
lateral E/e0 with either mitral E/A ratio or LARS (<16%) readily iden- diastolic filling waves with an E/A <1 can be misinterpreted as
tified patients with elevated PCWP with good accuracy and high feasi- impaired relaxation filling pattern. In addition, with mitral E and A
bility (Figure 6). fusion, the larger atrial stroke volume increases the mitral A-wave
duration. Diastolic fusion of filling waves can limit exercise capacity
Key Points because LV end-diastolic volume is reduced, lowering maximal car-
diac output. At PR values >320 ms, E and A velocity fusion leads
1. Reliable variables for estimation of LV filling pressure in pa- to filling only with atrial contraction (uniphasic A velocity), and dia-
tients with PH include lateral E/e0 , mitral E/A, LAVi, and LARS. stolic MR is seen.67 In these patients, maximal exercise capacity is
2. An E/A value #0.8 is usually seen in patients with noncardiac almost always limited because of the inability to increase LV filling
PH, whereas a ratio $2 is seen in patients with group II PH. with increasing heart rate. If there is complete fusion, peak TR velocity
3. LAVi > 34 ml/m2, lateral E/e0 > 13, and LARS < 18% favor the and LA volume and strain can be used to draw inferences about LV
presence of group II PH. filling pressures. Albeit the accuracy of LARS has not been critically
4. The recommended algorithm in patients with PH begins with examined in this setting.
Right bundle branch block results in delayed activation of the RV
mitral E/A, where a ratio #0.8 favors the diagnosis of precapil-
myocardium as electrical depolarization must spread through myo-
lary PH, and a ratio $2 favors the diagnosis of postcapillary
cytes instead of the specialized conduction system. Although minor
PH. For E/A > 0.8 but <2, lateral E/e0 > 13, LAVi > 34 mL/ changes in LV and RV synchrony are observed, no studies have
m2, and LARS # 16% favor the diagnosis of group II PH. convincingly shown that this leads to clinically meaningful changes
in LV diastolic variables or exercise capacity. This is also true of left
anterior or right posterior hemiblock. In contrast, left bundle branch
D. AV Block, Bundle Branch Block, and Electronic Pacing block (LBBB) can be associated with prolongation of IVRT, which
Normal subjects have near simultaneous contraction and relaxation leads to shortening of LV filling time,68,69 which in turn limits stroke
of all ventricular segments, which is demonstrated as synchronous sys- volume. This can be a challenge during exercise when the rapid heart
tolic shortening and diastolic lengthening as seen with strain imaging. rate further reduces diastolic filling time. LBBB can be associated with

LAP Es ma on in Pa ents with Pulmonary Hypertension

E/A ≤0.8, and E/A ≤0.8, and E > 50 cm/s or E/A≥2, and
E ≤ 50 cm/s E/A >0.8 to <2 e’ is reduced

LARS
LARS >18% LARS ≤18%

LARS not available

Lateral E/e’ ra o
E/e’ <8 E/e’>13

Normal LAP LAP indeterminate for ra o >8 and <13 Elevated LAP

Figure 6 Algorithm for estimation of mean LAP in patients with PH.

Journal of the American Society of Echocardiography Nagueh et al 561
Volume 38 Number 7

abnormal function of the interventricular septum leading to a reduc- pressure changes with time, mitral inflow E/A > 2.5, deceleration
tion in septal e0 velocity. Therefore, septal and average E/e0 may not time of E velocity < 150 ms, IVRT < 50 ms,48,49 decreased septal
reflect LV filling pressure reliably in patients with LBBB. Lateral E/e0 and lateral e0 velocities (3-4 cm/s), decreased GLS with preserved
may still be a valid indicator of filling pressure. There is a need for radial and circumferential strain,72 and reduced LA strain.47 In
further studies to determine how lateral E/e0 may be used in patients advanced cardiac amyloidosis, mitral annular velocity tracings may
with LV dyssynchrony caused by LBBB, RV pacing, and in patients show the ‘‘5-5-5’’ sign with systolic, early diastolic, and late diastolic
with cardiac resynchronization therapy (CRT). velocities all <5 cm/s.47 Grade 3 diastolic dysfunction is associated
RV pacing often leads to LV mechanical dyssynchrony resembling with a poor outcome.73 For cardiac amyloidosis (TTR or light chain
the contraction pattern in LBBB. Therefore, RV pacing can impair LV type), there are ‘‘red flags’’ suggestive of cardiac involvement, such as
filling and limits utility of septal e0 as a marker of diastolic dysfunction. increased LV and RV wall thickness, biatrial enlargement, preserved
Patients who mostly need atrial pacing have no alterations in LV sys- EF with low stroke volume index, an association with paradoxical
tolic and diastolic function. In patients who are not dependent on ven- low-flow, low-gradient AS, and diastolic dysfunction (E/A ratio > 2
tricular pacing to maintain heart rate, pacemaker settings are often set with increased filling pressures and markedly reduced annular veloc-
with a long AV delay to encourage fusion or native QRS beats to mini- ities.22 In addition, speckle-tracking of the LV myocardium in pa-
mize RV pacing. There are few studies that have looked at the utility tients with cardiac amyloidosis has shown a distinctive phenotype
of mitral annular velocities in this setting and it appears that their ac- of apical sparing using a polar plot of LV longitudinal strain
curacy is less in the presence of LBBB, RV pacing, and in patients who compared with hypertensive heart disease, HCM, and AS.74 There
have received CRT.21,70 have been several ratios to evaluate apical sparing, including a ratio
of apical strain to mid and basal strain (>1),74 the septal apical-to-
basal ratio using the four-chamber septal apical and basal segmental
Key Points longitudinal strain values with a value >2.1,75 as well as EF/strain ra-
1. In patients with first-degree AV block, the variables used to tio >4.1.76
evaluate diastolic function and filling pressures are valid if there
is no fusion of mitral E and A velocities. F. Pericardial Constriction
2. The accuracy of mitral annular velocities and E/e0 is less in the Pericardial constriction is characterized by dissociation of intratho-
presence of LBBB, RV pacing, and in patients who have racic and intracardiac pressures as well as interventricular depen-
received CRT. dence due to the effect of the constricting pericardium on the LV
3. If only mitral A velocity is present, peak TR velocity (>2.8 m/s) and the RV. This results in respirophasic shift of the ventricular
can be used as an indicator of LV filling pressures. The accuracy septum, septal bounce, mitral and tricuspid inflow variation with
of LARS in this setting has not been examined. respiration (>25% and 40%, respectively), and expiratory reversal
of end-diastolic flow within the hepatic veins (end-diastolic reversal
velocity/forward flow velocity $ 0.8).77,78 In addition, mitral septal
e0 velocity is often >7 cm/s, whereas it is usually #5 cm/s in restric-
tive cardiomyopathy. Tethering of the LV lateral and RV free walls
E. Restrictive Cardiomyopathy contributes to the constrictive physiology and is demonstrated by
Restrictive cardiomyopathies are composed of a heterogeneous an increased ratio of mitral septal to lateral e0 (annulus reversus).79
group of heart muscle diseases including but not limited to such dis- Similarly, the lateral LV and RV free wall peak systolic strain is dimin-
eases as idiopathic restrictive cardiomyopathy, cardiac amyloidosis, ished compared with the septal peak systolic strain (strain rever-
and sarcoidosis. These diseases are often characterized by restrictive sus).80 Figure 7 shows a validated algorithm based on data from
pathophysiology, which is defined as a rapid rise in diastolic ventricu- the Mayo Clinic comparing pericardial constriction with restrictive
lar pressure with only a small increase in LV volume due to increased cardiomyopathy.78,81 This algorithm was corroborated in a similar
myocardial stiffness.71 It is important to make the distinction between study.82 The presence of a normal septal (or medial) annular e0 ve-
restrictive LV filling, which can occur with other diseases such as cor- locity in a patient referred with HF diagnosis should raise suspicion
onary artery disease, dilated cardiomyopathy, and hypertrophic car- of pericardial constriction.
diomyopathy (HCM), and restrictive cardiomyopathy.
A common restrictive cardiomyopathy is cardiac amyloidosis, an
infiltrative cardiomyopathy most commonly caused by either immu-
Key Points
noglobulin light chain deposition or by misfolding in the hepatic- 1. Patients with early disease restrictive cardiomyopathy usually
derived transthyretin (TTR) protein into amyloid fibrils. In the earlier have grade 1 diastolic dysfunction that progresses to grade 2
stages of cardiac amyloidosis, diastolic function can vary from grade
and grade 3 as the severity of the disease increases.
1 diastolic dysfunction with impaired relaxation and normal LV
2. In patients with advanced disease, grade 3 diastolic dysfunc-
filling pressures to grade 2 (pseudonormalization). In later stages,
grade 3 diastolic dysfunction occurs with markedly elevated LV tion is present and is characterized by mitral inflow E/
filling pressures.46 In contemporary diagnostic approaches, there A > 2.5, deceleration time of E velocity < 150 ms,
has been an evolution of the diastolic function techniques applied IVRT < 50 ms, and decreased septal and lateral e0 velocities
in studying these patients, initially using mitral inflow and pulmo- (3-4 cm/s), as well as decreased LV GLS, RV strain, and LA
nary vein flow, to TDI of the mitral annulus and now STE imaging, reservoir and pump strain.
including LV GLS, LA strain, and RV strain.47 The advanced stages of 3. Strain imaging of the LV myocardium in patients with cardiac
restrictive cardiomyopathy are characterized by typical restrictive amyloidosis can have a distinctive phenotype of apical sparing.
physiology with a dip-and-plateau pattern for early diastolic LV
562 Nagueh et al Journal of the American Society of Echocardiography
July 2025

velocity ratio. However, MR does not affect pulmonary vein Ar veloc-

4. Patients with pericardial constriction usually have respirophasic ity. Ar peak velocity and duration can be used to draw inferences
shift of the interventricular septum, septal bounce, mitral and about LVEDP in the absence of atrial myopathy and first-degree AV
tricuspid inflow variation with respiration (>25% and 40%, block.45,53 In addition, LA reservoir and pump strains in patients
respectively), normal to increased medial annular early diastolic with HCM have been associated with functional capacity and devel-
velocity (>7 cm/s), increased expiratory reversal of end-diastolic opment of atrial fibrillation, though there is need to study their specific
flow within the hepatic veins (end-diastolic reversal velocity/for- application for LAP estimation. A restrictive LV filling pattern with
ward flow velocity $ 0.8), as well as annulus and strain reversus. increased E/e0 is associated with HF hospitalizations, reduced exercise
tolerance in children and adults, and sudden cardiac death.45

G. HCM Key Points

Impaired LV relaxation is an early finding in subjects carrying patho-
genic mutations for HCM that occurs before the development of 1. A comprehensive approach is recommended for the evalua-
LV hypertrophy.45 Furthermore, diastolic dysfunction is a ubiquitous tion of diastolic function in patients with HCM.
finding in HCM patients irrespective of the hypertrophy pattern and 2. The Doppler variables that are recommended are mitral inflow
contributes to their symptoms even in the absence of dynamic velocities, pulmonary vein velocities, mitral annular velocities,
obstruction.45 The Doppler variables that are recommended are peak TR velocity by CW Doppler, and biplane LA volumes.
mitral inflow velocities, pulmonary vein velocities, mitral annular ve- 3. Restrictive LV filling with increased E/e0 is associated with
locities, peak TR velocity by CW Doppler, and biplane LA volumes. reduced functional capacity and HF hospitalizations in patients
Significant MR can lead to elevated LAP, LA volumes, and peak TR with HCM.
velocity with a decrease in the pulmonary vein systolic to diastolic

Figure 7 (A) Algorithm for differentiation of pericardial constriction from restrictive cardiomyopathy. (B) Sensitivity, specificity and
predictive values of the algorithm. Reprinted with permission from Klein et al.78 NPV, Negative predictive value; PPV, positive predic-
tive value.
Journal of the American Society of Echocardiography Nagueh et al 563
Volume 38 Number 7

H. Atrial Fibrillation 16. HFPEF DIAGNOSIS

Atrial fibrillation is a common finding in patients with diastolic dysfunc-
tion and HFpEF. Atrial fibrillation poses several challenges due to tachy- HFpEF constitutes half of all HF hospitalizations, with a growing prev-
cardia in some cases, the absence of organized atrial contraction, the alence relative to HF with reduced EF.86-88 This is multifactorial
variability in cycle length, and the frequent presence of LA enlargement. because of an aging population with an increasing burden of
Several echocardiographic parameters have been suggested as markers comorbidities such as hypertension, diabetes, obesity—
of LV filling pressure in patients with atrial fibrillation and smaller single- predominantly metabolic syndrome–associated comorbidities—that
center studies have shown promising results.40,41,83 In a recent multi- contribute to the development of HFpEF.89 The guidelines present
center study, it was found that no single echocardiographic parameter a stepwise guide to the diagnosis of HFpEF, including clinical diag-
had sufficiently strong association with LV filling pressure to be recom- nosis, and guide for noninvasive and invasive testing (Figure 9).
mended as a stand-alone marker. When using a multiparametric
approach, the accuracy in differentiating between normal and elevated
A. Clinical Diagnosis of HFpEF
LV filling pressure was moderate (Figure 8).42
When assessing LV filling pressure in atrial fibrillation, one should use The clinical diagnosis of HFpEF is reached by establishing the pres-
average values from several cardiac cycles, and the selected heart cycles ence of signs and symptoms of congestive HF as well as an echocar-
should be reflective of the average heart rate.40 For E/e0 and ratios that diographic determination of normal LVEF (generally accepted to be
rely on timing of mitral E onset and e0 velocity onset, the use of a dual $50%) in the absence of other cardiac or noncardiac causes for the
Doppler probe can enable the recording of both velocities from the patient’s symptoms. The American College of Cardiology/
same cardiac cycle, with a much higher accuracy for estimation of American Heart Association/Heart Failure Society of America and
PCWP than averaging velocities or time intervals from several cardiac European Society of Cardiology (ESC) each provide definitions of
cycles.84,85 Looking at variability of mitral inflow with varying cycle HFpEF and suggested diagnostic evaluation22,90-92 (Table 8).
length is a practical method to determine whether LV filling pressure The recognition of the clinical HF syndrome is the first step in diag-
is elevated. This necessitates recording several cardiac cycles with a nosing HFpEF. Several criteria have been proposed to diagnose HF
sweep speed of 50 mm/s. Patients with less beat-to-beat mitral inflow including the Framingham, Boston, Gothenburg, and ESC
variability usually have elevated LV filling pressure.40 criteria.93-95 The Framingham criteria are among the most widely
accepted criteria for diagnosis of HF in epidemiologic studies.
Although the Framingham criteria demonstrate excellent specificity
for the diagnosis of HF, they lack sensitivity, particularly in elderly
Key Points patients who may have not had acute HF decompensation.96,97
Therefore, a patient may have HFpEF even in the absence of satisfying
1. In patients with atrial fibrillation, several echocardiographic pa- specific clinical criteria for HF.
rameters are associated with LAP, but no single parameter has Early presenting symptoms of HFpEF may be relatively nonspe-
a strong association. cific, including exertional dyspnea, exercise intolerance, or fatigue.
2. A decision algorithm that combines multiple echocardio- Clinical suspicion should be further raised if a patient presenting
graphic parameters can differentiate between normal and with these symptoms has one or more risk factors commonly associ-
elevated LV filling pressure with moderate accuracy. ated with HFpEF. The most commonly associated comorbidities with
3. One should use average values from several cardiac cycles and HFpEF that may raise pretest probability of the diagnosis include the
the selected heart cycles should be reflective of the average history of hypertension, elderly age (>60 years), obesity (body mass
heart rate. index > 30 kg/m2), history of diabetes mellitus, or history of atrial
fibrillation.8,98

LAP Estimation in Atrial Fibrillation

1. Mitral E velocity ≥100 cm/s
2. Septal E/e’ ratio > 11
3. TR velocity > 2.8 m/s or PASP > 35 mm Hg
4. DT ≤160 ms

None or 1 of above 2 of above 3 or more of above

LARS <18%
Pulm vein S/D ratio <1
BMI > 30 kg/m2

None 2/3 or 3/3

1 only or not available or
not reliable

Normal LAP Indeterminate Elevated LAP

Figure 8 Algorithm for estimation of mean LAP with atrial fibrillation. BMI, Body mass index; Pulm, pulmonary.
564 Nagueh et al Journal of the American Society of Echocardiography
July 2025

History and Physical Examination, Exclude non-cardiac

Chest X-Ray, Labs, Natriuretic peptides causes
for dyspnea
Comprehensive Echocardiogram

• Exclude Significant MS, primary MR,

AS, AR, primary TR
* Exclude significant CAD, non-cardiac PH
* Exclude cardiac amyloidosis, HCM
* Exclude pericardial constriction
Algorithm for LAP estimation
(figure 3)

Dyspnea with LAP

LAP elevated normal by rest echo

Positive test Diastolic Exercise echo or

RHC with/without exercise
HFpEF confirmed Negative test
Non cardiac dyspnea

Figure 9 Algorithm for HFpEF diagnosis. *Multimodality imaging and cardiac catheterization should be used as needed to establish
the presence of alternative diagnoses to HFpEF. CAD, Coronary artery disease; MS, mitral stenosis; RHC, right heart catheterization.

Physical examination may demonstrate signs of congestion, fibrillation, which may be occult in HFpEF and is prevalent in 15%
including elevated jugular venous pressure, presence of an S3 gallop, to 41% of patients.99,100 Last, chest radiography should be performed
pulmonary crackles or rales, hepatomegaly, ascites, or lower extrem- in the evaluation of HFpEF, particularly in patients presenting with
ity edema. Electrocardiography should be performed in all patients dyspnea. Chest radiography may show signs of volume overload
with suspected HFpEF. Although there are no pathognomonic or (Kerley B-lines, pleural effusion, pulmonary congestion) or reveal
diagnostic signs to identify HFpEF on electrocardiography per se, pa- other noncardiac causes of dyspnea. Radiographic findings appear
tients may have features of LV hypertrophy or LA enlargement. to have low sensitivity in detecting patients with HFpEF compared
Furthermore, electrocardiographic evaluation can screen for atrial with PCWP measurements by cardiac catheterization.101

Table 8 Consensus definitions of HFpEF

ACC/AHA/HFSA ESC

Definition of EF $ 50% and: EF $ 50% and:

HFpEF  Clinical symptoms and/or signs of HF and  Clinical symptoms and/or signs of HF and
 Diastolic dysfunction  Elevated NP levels and at least one of the following:
◦ Relevant structural heart disease (LVH and/or LAE)
◦ Diastolic dysfunction
Clinical signs  Obtain detailed history and physical  Obtain detailed history and physical
examination  Assess symptoms and signs of HF for evidence of congestion
 Assess volume status and vital signs to
determine evidence of congestion
NPs  Can guide the diagnosis of HF, especially in  Can guide the diagnosis or exclusion of HF
the setting of clinical uncertainty  Normal levels exclude HF
Echocardiography  EF measurement and assessment of valvular  Evaluate for evidence of functional or structural abnormalities
or myocardial abnormalities  Key structural abnormalities:
 Evidence of LV diastolic dysfunction may be ◦ LAVi > 34 mL/m2
considered to define the syndrome of HFpEF ◦ LVMi $ 115 g/m2 for men and $95 g/m2 for women
 Key functional abnormalities:
◦ E/e0 $13
◦ Mean e0 (mean of septal and lateral e0 ) < 9 cm/s
Invasive testing  RHC if refractory to initial therapy or if a  RHC at rest followed by exercise hemodynamics if below the threshold of
specific clinical question needs to be PCWP 15 mm Hg, may be considered in cases of clinical uncertainty
addressed
 LHC is recommended if HF and angina
present
ACC, American College of Cardiology; AHA, American Heart Association; ESC, European Society of Cardiology; HFSA, Heart Failure Society of
America; LAE, LA enlargement; LHC, left heart catheterization; LVH, LV hypertrophy; LVMi, LV mass index; RHC, right heart catheterization.
Journal of the American Society of Echocardiography Nagueh et al 565
Volume 38 Number 7

Table 9 Differential diagnosis of HFpEF tions for estimation of LV filling pressures for these groups should be
followed.
Cardiac, myocardial Although echocardiography remains at the cornerstone of HF eval-
Restrictive cardiomyopathy uation, CMR is a robust imaging modality to characterize myocardial
Cardiac amyloidosis tissue abnormalities.15 CMR can be considered in the workup of pa-
tients with HFpEF, particularly if echocardiographic imaging quality is
Endomyocardial fibrosis
suboptimal or if there are concerns for an infiltrative myopathic pro-
Systemic sclerosis
cess such as amyloidosis.
Radiation fibrosis
Hemochromatosis
C. Natriuretic Peptides
Fabry disease
Levels of natriuretic peptides (NPs), including brain NP and N-termi-
Glycogen storage disease
nal pro–B-type NP, should be checked in patients presenting with sus-
Metastatic cancer pected HFpEF or undifferentiated dyspnea.102 NP levels are generally
HCM higher in patients with HFpEF compared with subjects without HF
Arrhythmogenic RV cardiomyopathy and have useful prognostic implications.102 Of note, NP levels are
Myocardial ischemia known to be lower in patients with HFpEF compared with those
HF with recovered EF with HF with reduced EF, and obesity, a common comorbidity asso-
Pulmonary arterial hypertension ciated with HFpEF, is well known to be associated with lower NP
levels.103 Prior studies have indicated that up to 30% of patients
Cardiac, nonmyocardial
with HFpEF can have normal NP levels despite signs and symptoms
Valvular heart disease (>mild stenosis or $moderate
of HF, echocardiographic abnormalities, and elevated LV filling pres-
regurgitation)
sures on invasive hemodynamic testing.104 Therefore, although
Pericardial disease important in the evaluation of HFpEF, a normal NP level does not
Constrictive or effusive constrictive pericarditis necessarily rule out a diagnosis of HFpEF, particularly in obese pa-
Cardiac tamponade tients with HFpEF.
High-output HF
Noncardiac D. Role of HFpEF Prediction Scores
Pulmonary disease For patients with an uncertain probability of HFpEF, a few scoring
Anemia systems have been developed to aid in predicting the likelihood of
Venous insufficiency the disease. Reddy et al.98 developed the H2FPEF score by retrospec-
tively comparing clinical findings in 414 patients referred for evalu-
ation of unexplained dyspnea who underwent invasive
B. Echocardiographic Imaging hemodynamic exercise testing. The H2FPEF score is a weighted
Once the clinical diagnosis of HFpEF is suspected in the appropriate score ranging from 0 to 9 and based on six clinical variables: heavy
clinical setting, physical examination findings, and/or the presence of (body mass index >30 kg/m2, 2 points), hypertensive (two or more
risk factors, the next step in the diagnostic evaluation of HFpEF should antihypertensive medications, 1 point), atrial fibrillation (paroxysmal
be imaging. Echocardiography is the most common initial imaging mo- or persistent, 3 points), PH (Doppler echocardiographic estimated
dality in the evaluation of HFpEF, providing information on structural PASP > 35 mm Hg, 1 point), elder (age > 60 years, 1 point), and
changes and hemodynamic parameters that may support a diagnosis filling pressure (septal E/e0 > 9, 1 point). A score of 0 to 2 is indicates
of HFpEF, while also useful for the evaluation of other cardiac causes a low probability of HFpEF. An intermediate score of 2 to 5 is corre-
of dyspnea, such as valvular disease, infiltrative cardiomyopathy, or lated with a 40% to 80% likelihood of HFpEF, whereas a score >5
pericardial disease. Though different EF cutoff points have been pro- indicates high HFpEF probability. Although the H2FPEF score was
posed for the diagnosis of HFpEF, an LVEF of $50% is accepted as formulated on the basis of a relatively small referral population, it
normal or preserved and is consistent with a diagnosis of HFpEF.8 has been validated in a test cohort and showed good discrimination
Patients with HFpEF often have morphologic and functional ab- of HFpEF from noncardiac dyspnea.98 Additionally, the clinical vari-
normalities on the echocardiogram (e.g., LV hypertrophy, concentric ables are commonly assessed for, allowing wide applicability in
remodeling, LA enlargement, diastolic dysfunction, reduced LV GLS, various clinical settings.
reduced LARS), though by themselves these findings are not diag- The Heart Failure Association of the ESC also recently published a
nostic of HFpEF. Diastolic function is assessed as recommended in consensus recommendation for the diagnosis of HFpEF, the HFA-
the previous sections (see algorithms in Figures 2 and 3). A compre- PEFF diagnostic algorithm.8 The HFA-PEFF algorithm is a stepwise
hensive echocardiographic examination with measurement of these approach to the diagnosis of HFpEF meant to be performed in the
parameters should be performed in the initial diagnostic evaluation ambulatory setting. It begins with a pretest probability assessment
of a patient with suspected HFpEF. If the resting echocardiogram (signs and symptoms of HF, presence of comorbidities or risk factors,
shows increased LV filling pressure (grade 2 or 3 diastolic dysfunc- electrocardiography, echocardiography, NP levels, and functional
tion), the diagnosis of HFpEF is confirmed in the appropriate clinical testing such as the 6-minute walk test or cardiopulmonary exercise
setting. However, if only diastolic function grade 1 is present at rest in testing). If pretest assessment is overall suggestive of a HFpEF diag-
a patient with exertional dyspnea, diastolic exercise echocardiogra- nosis, further testing, including comprehensive echocardiography
phy or cardiac catheterization should be performed (see the following and NP levels, are used to tabulate a likelihood score. If a diagnosis
discussion). For patients in atrial fibrillation, the specific recommenda- of HFpEF is intermediate, noninvasive or invasive exercise stress
566 Nagueh et al Journal of the American Society of Echocardiography
July 2025

testing is recommended for further evaluation. The HFA-PEFF diag- scores have a high yield of an indeterminate category where stress
nostic algorithm highlights the importance of a combination of testing or right heart catheterization is needed. The utility of compre-
abnormal structural and functional echocardiographic parameters in hensive echocardiography (at rest) in the indeterminate category us-
the diagnosis of HFpEF. Although NP levels are incorporated in this al- ing LA strain, LA stiffness, pulmonary vein flow, IVRTand PA diastolic
gorithm, a diagnosis of HFpEF does not rely only on elevated NP pressure should be evaluated. If successful, it can decrease the need
levels. for additional testing.
These scores may be used in the initial evaluation of patients with
possible HFpEF but confirmation of elevated LV filling pressures Key Points
(noninvasive estimation or invasive measurement) is needed. In
1. Clinical data, radiographic findings, and NP levels should be
borderline cases with either approach, exercise stress testing should
be pursued to determine if mean PCWP is abnormally increased considered in trying to determine whether there is a cardiac
with exercise ($25 mm Hg). cause for dyspnea.
2. The next step is a comprehensive echocardiographic examina-
tion that includes the acquisition and measurement of LV GLS
E. Alternative Diagnoses
and LARS.
Certain cardiac and noncardiac conditions that mimic HFpEF should
3. Before reaching a diagnosis of HFpEF, valvular heart disease,
be ruled out (Table 9). The prevalence of TTR cardiac amyloidosis in
noncardiac PH, significant coronary artery disease, infiltrative
HFpEF is estimated to be 5% to 13%; though this is based on limited
data from autopsy studies and nuclear scan–based screening of pa- and HCM, and pericardial constriction should be excluded.
tients with HFpEF.105,106 In the only prospective study using endo- 4. Apply the algorithm for the estimation mean LAP shown in
myocardial biopsy in the evaluation of HFpEF, 14% of patients Figure 3. If LAP is elevated at rest in symptomatic patients,
were found to have cardiac amyloidosis, the majority being TTR car- HFpEF diagnosis is reached.
diac amyloidosis.107 5. If LAP at rest is normal in a symptomatic patient, the next step
is diastolic exercise stress echocardiography. If positive, HFpEF
F. Exercise and Invasive Hemodynamic Testing diagnosis is reached. If negative, then noncardiac cause of dys-
Although many patients with HFpEF present with clear HF syndrome pnea is present. If the test is inconclusive, then the next step
and echocardiographic findings consistent with HFpEF, some present should be right heart catheterization.
with little more than dyspnea on exertion and fatigue. These patients
often require additional testing to further evaluate for a diagnosis of
HFpEF. Exercise stress echocardiography, typically performed with NOTICE AND DISCLAIMER
supine bicycle or treadmill exercise, is the primary modality of nonin-
vasive stress testing to evaluate for occult HFpEF. Compared with This report is made available by the ASE as a courtesy reference
invasive hemodynamic testing, exercise echocardiography is less source for its members. It contains recommendations only and should
expensive, with less risk, and may be more accessible. Additionally, not be used as the sole basis for making medical practice decisions or
exercise stress testing may be useful to assess for concomitant obstruc- for taking disciplinary action against any employee. The statements
tive coronary artery disease. and recommendations contained in this report are based primarily
Right heart catheterization is considered the gold-standard test to on the opinions of experts rather than on scientifically verified data.
establish the diagnosis of HFpEF, during which exercise hemody- The ASE makes no express or implied warranties regarding the
namics can often be obtained. PCWP > 15 mm Hg at rest, or completeness or accuracy of the information in this report, including
$25 mm Hg with exercise, measured at end-expiration, is consistent the warranty of merchantability or fitness for a particular purpose. In
with the diagnosis of HFpEF.108,109 The use of exercise invasive he- no event shall the ASE be liable to you, your patients, or any other
modynamics has been shown to significantly increase the diagnostic third parties for any decision made or action taken by you or such
yield of HFpEF, particularly for patients presenting at an earlier stage other parties in reliance on this information. Nor does your use of
of the disease.108 In patients who are unable to exercise, saline loading this information constitutes offering of medical advice by the ASE
during right heart catheterization can be considered as an alternative or create any physician-patient relationship between the ASE and
diagnostic modality but has been shown to be less sensitive compared your patients or anyone else.
with exercise in the diagnosis of HFpEF.110

G. Research Needs REVIEWERS

A novel method for prediction of LV diastolic pressures using ultra-
This document was reviewed by members of the 2024–2025 ASE
sound enhancing agent microbubbles and subharmonic-aided pres-
Guidelines and Standards Committee, ASE Board of Directors, and
sure estimation has been reported in a single-center study.111 High–
designated reviewer Dr. Andrew Pellett.
frame rate echocardiography was used to measure shear wave elas-
tography to gain insight into myocardial stiffness.112 The clinical feasi-
bility, accuracy, and incremental value of these new techniques await
evaluation in multicenter studies. There are several AI models for the ACKNOWLEDGMENTS
assessment of LV diastolic function. Validation of these and future
models using invasive hemodynamics and clinical outcomes in multi- The writing committee would like to thank Arshama Dehghan, MS
center studies is needed to establish their accuracy and clinical and Brian Claggett, PhD for their assistance in analyses of reference
relevance. For HFpEF diagnosis, the existing approaches relying on limits for diastolic indices by age and sex.
Journal of the American Society of Echocardiography Nagueh et al 567
Volume 38 Number 7

SUPPLEMENTARY DATA 19. Hasegawa H, Little WC, Ohno M, et al. Diastolic mitral annular velocity
during the development of heart failure. J Am Coll Cardiol 2003;41:
Supplementary data related to this article can be found at https://doi. 1590-7.
org/10.1016/j.echo.2025.03.011. 20. Rivas-Gotz C, Khoury DS, Manolios M, et al. Time interval between
onset of mitral inflow and onset of early diastolic velocity by tissue
Doppler: a novel index of left ventricular relaxation: experimental
studies and clinical application. J Am Coll Cardiol 2003;42:1463-70.
REFERENCES 21. Lababidi H, Rahi W, Smiseth OA, et al. New Algorithm for Estimating
Left Ventricular Filling Pressure by Echocardiography. Circulation 2025
1. Nagueh SF, Smiseth OA, Appleton CP, et al. Recommendations for the (in press).
evaluation of left ventricular diastolic function by echocardiography: an up- 22. Smiseth OA, Morris DA, Cardim N, et al. Multimodality imaging in pa-
date from the American Society of Echocardiography and the European tients with heart failure and preserved ejection fraction: an expert
Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2016;29: consensus document of the European Association of Cardiovascular Im-
277-314. aging. Eur Heart J Cardiovasc Imaging 2022;23:e34-61.
2. Andersen OS, Smiseth OA, Dokainish H, et al. Estimating left ventricular 23. Obokata M, Kane GC, Reddy YN, et al. Role of diastolic stress testing in
filling pressure by echocardiography. J Am Coll Cardiol 2017;69: the evaluation for heart failure with preserved ejection fraction: a
1937-48. simultaneous invasive-echocardiographic study. Circulation 2017;
3. Inoue K, Khan FH, Remme EW, et al. Determinants of left atrial reservoir 135:825-8.
and pump strain and use of atrial strain for evaluation of left ventricular 24. Ha JW, Lulic F, Bailey KR, et al. Effects of treadmill exercise on mitral
filling pressure. Eur Heart J Cardiovasc Imaging 2021;23:61-70. inflow and annular velocities in healthy adults. Am J Cardiol 2003;91:
4. Shah AM, Claggett B, Kitzman D, et al. Contemporary assessment of left 114-5.
ventricular diastolic function in older adults: the Atherosclerosis risk in 25. Burgess MI, Jenkins C, Sharman JE, et al. Diastolic stress echocardiogra-
Communities Study. Circulation 2017;135:426-39. phy: hemodynamic validation and clinical significance of estimation of
5. Paulus WJ, Tsch€ ope C, Sanderson JE, et al. How to diagnose diastolic ventricular filling pressure with exercise. J Am Coll Cardiol 2006;47:
heart failure: a consensus statement on the diagnosis of heart failure 1891-900.
with normal left ventricular ejection fraction by the heart failure and 26. Ritzema JL, Richards AM, Crozier IG, et al. Serial Doppler echocardiog-
Echocardiography Associations of the European Society of Cardiology. raphy and tissue Doppler imaging in the detection of elevated directly
Eur Heart J 2007;28:2539-50. measured left atrial pressure in ambulant subjects with chronic heart fail-
6. Nagueh SF. Cardiac resynchronization therapy and dynamic changes in ure. JACC Cardiovasc Imaging 2011;4:927-34.
right ventricular function. Circ Cardiovasc Imaging 2018;11:e008195. 27. Talreja DR, Nishimura RA, Oh JK. Estimation of left ventricular filling
7. Nagueh SF. Heart failure with preserved ejection fraction: insights into pressure with exercise by Doppler echocardiography in patients with
diagnosis and pathophysiology. Cardiovasc Res 2021;117:999-1014. normal systolic function: a simultaneous echocardiographic-cardiac cath-
8. Pieske B, Tsch€ ope C, de Boer RA, et al. How to diagnose heart failure with eterization study. J Am Soc Echocardiogr 2007;20:477-9.
preserved ejection fraction: the HFA-PEFF diagnostic algorithm: a 28. Ye Z, Miranda WR, Yeung DF, et al. Left atrial strain in evaluation of heart
consensus recommendation from the Heart Failure Association (HFA) of failure with preserved ejection fraction. J Am Soc Echocardiogr 2020;33:
the European Society of Cardiology (ESC). Eur Heart J 2019;40:3297-317. 1490-9.
9. Nagueh SF. Left ventricular diastolic function: understanding pathophys- 29. Merli E, Ciampi Q, Scali MC, et al. Pulmonary congestion during exercise
iology, diagnosis, and prognosis with echocardiography. JACC Cardio- stress echocardiography in ischemic and heart failure patients. Circ Car-
vasc Imaging 2020;13:228-44. diovasc Imaging 2022;15:e013558.
10. Nagueh SF, Khan SU. Left atrial strain for assessment of left ventricular 30. Choi EY, Shim CY, Kim SA, et al. Passive leg-raise is helpful to identify
diastolic function: Focus on populations with normal LVEF. JACC Cardi- impaired diastolic functional reserve during exercise in patients with
ovasc Imaging 2023;16:691-707. abnormal myocardial relaxation. J Am Soc Echocardiogr 2010;23:
11. Kurt M, Wang J, Torre-Amione G, et al. Left atrial function in diastolic 523-30.
heart failure. Circ Cardiovasc Imaging 2009;2:10-5. 31. Luong CL, Anand V, Padang R, et al. Prognostic significance of elevated
12. Reddy YNV, Obokata M, Egbe A, et al. Left atrial strain and compliance left ventricular filling pressures with exercise: insights from a cohort of
in the diagnostic evaluation of heart failure with preserved ejection frac- 14,338 patients. J Am Soc Echocardiogr 2024;37:382-93.
tion. Eur J Heart Fail 2019;21:891-900. 32. Jiang R, Yeung DF, Behnami D, et al. A novel continuous left ventricular
13. Kim D, Seo JH, Choi KH, et al. Prognostic implications of left atrial stiff- diastolic function score using machine learning. J Am Soc Echocardiogr
ness index in heart failure patients with preserved ejection fraction. JACC 2022;35:1247-55.
Cardiovasc Imaging 2023;16:435-45. 33. Lee E, Ito S, Miranda WR, et al. Artificial intelligence-enabled ECG for left
14. Nagueh SF, Chang SM, Nabi F, et al. Cardiac imaging in patients with ventricular diastolic function and filling pressure. NPJ Digit Med 2024;7:4.
heart failure and preserved ejection fraction. Circ Cardiovasc Imaging 34. Tromp J, Seekings PJ, Hung CL, et al. Automated interpretation of systolic
2017;10:e006547. and diastolic function on the echocardiogram: a multicohort study. Lan-
15. Chamsi-Pasha MA, Zhan Y, Debs D, et al. CMR in the evaluation of dia- cet Digit Health 2022;4:e46-54.
stolic dysfunction and phenotyping of HFpEF: current role and future 35. Akerman AP, Porumb M, Scott CG, et al. Automated echocardiographic
perspectives. JACC Cardiovasc Imaging 2020;13:283-96. detection of heart failure with preserved ejection fraction using artificial
16. Lewandowski D, Yang EY, Nguyen DT, et al. Relation of left ventricular intelligence. JACC Adv 2023;2:100452.
diastolic function to global fibrosis burden: implications for heart failure 36. Chiou YA, Hung CL, Lin SF. AI-assisted echocardiographic prescreening
risk stratification. JACC Cardiovasc Imaging 2023;16:783-96. of heart failure with preserved ejection fraction on the basis of Intrabeat
17. Nagueh SF, Sun H, Kopelen HA, et al. Hemodynamic determinants of the dynamics. JACC Cardiovasc Imaging 2021;14:2091-104.
mitral annulus diastolic velocities by tissue Doppler. J Am Coll Cardiol 37. Carluccio E, Cameli M, Rossi A, et al. Left atrial strain in the assessment of
2001;37:278-85. diastolic function in heart failure: a machine learning approach. Circ Car-
18. Opdahl A, Remme EW, Helle-Valle T, et al. Determinants of left ventric- diovasc Imaging 2023;16:e014605.
ular early-diastolic lengthening velocity: independent contributions from 38. Pandey A, Kagiyama N, Yanamala N, et al. Deep-learning models for the
left ventricular relaxation, restoring forces, and lengthening load. Circula- echocardiographic assessment of diastolic dysfunction. JACC Cardiovasc
tion 2009;119:2578-86. Imaging 2021;14:1887-900.
568 Nagueh et al Journal of the American Society of Echocardiography
July 2025

39. Chao CJ, Kato N, Scott CG, et al. Unsupervised machine learning for 59. Olson JJ, Costa SP, Young CE, et al. Early mitral filling/diastolic mitral
assessment of left ventricular diastolic function and risk stratification. J annular velocity ratio is not a reliable predictor of left ventricular filling
Am Soc Echocardiogr 2022;35:1214-25. pressure in the setting of severe mitral regurgitation. J Am Soc Echocar-
40. Nagueh SF, Kopelen HA, Quinones MA. Assessment of left ventricular diogr 2006;19:83-7.
filling pressures by Doppler in the presence of atrial fibrillation. Circula- 60. Avenatti E, Little SH, Barker CM, et al. Changes in left atrial function after
tion 1996;94:2138-45. Transcutaneous mitral valve repair. Am J Cardiol 2018;122:1204-9.
41. Sohn DW, Song JM, Zo JH, et al. Mitral annulus velocity in the evaluation 61. Zaid RR, Barker CM, Little SH, et al. Pre- and post-operative diastolic
of left ventricular diastolic function in atrial fibrillation. J Am Soc Echocar- dysfunction in patients with valvular heart disease: diagnosis and thera-
diogr 1999;12:927-31. peutic implications. J Am Coll Cardiol 2013;62:1922-30.
42. Khan FH, Zhao D, Ha JW, et al. Evaluation of left ventricular filling pres- 62. Ito S, Miranda WR, Nkomo VT, et al. Prognostic risk stratification of pa-
sure by echocardiography in patients with atrial fibrillation. Echo Res tients with moderate aortic stenosis. J Am Soc Echocardiogr 2021;34:
Pract 2024;11:1. 248-56.
43. Nagueh SF, Mikati I, Kopelen HA, et al. Doppler estimation of left ven- 63. Klein AL, Ramchand J, Nagueh SF. Aortic stenosis and diastolic dysfunc-
tricular filling pressure in sinus tachycardia. A new application of tissue tion: Partners in crime. J Am Coll Cardiol 2020;76:2952-5.
Doppler imaging. Circulation 1998;98:1644-50. 64. Oh JK, Hatle LK, Sinak LJ, et al. Characteristic Doppler echocardio-
44. Sohn DW, Kim YJ, Kim HC, et al. Evaluation of left ventricular diastolic graphic pattern of mitral inflow velocity in severe aortic regurgitation. J
function when mitral E and A waves are completely fused: role of assess- Am Coll Cardiol 1989;14:1712-7.
ing mitral annulus velocity. J Am Soc Echocardiogr 1999;12:203-8. 65. Vejpongsa P, Torre-Amione G, Marcos-Abdala HG, et al. Long term
45. Nagueh SF, Phelan D, Abraham T, et al. Recommendations for multimo- development of diastolic dysfunction and heart failure with preserved
dality cardiovascular imaging of patients with hypertrophic cardiomyop- left ventricular ejection fraction in heart transplant recipients. Sci Rep
athy: an update from the American Society of Echocardiography, in 2022;12:34.
collaboration with the American Society of Nuclear Cardiology, the So- 66. Appleton CP. Influence of incremental changes in heart rate on mitral
ciety for Cardiovascular Magnetic Resonance, and the Society of Cardio- flow velocity: assessment in lightly sedated, conscious dogs. J Am Coll
vascular Computed Tomography. J Am Soc Echocardiogr 2022;35: Cardiol 1991;17:227-36.
533-69. 67. Appleton CP, Basnight MA, Gonzalez MS. Diastolic mitral regurgitation
46. Klein AL, Hatle LK, Burstow DJ, et al. Doppler characterization of left with atrioventricular conduction abnormalities: relation of mitral flow ve-
ventricular diastolic function in cardiac amyloidosis. J Am Coll Cardiol locity to transmitral pressure gradient in conscious dogs. J Am Coll Car-
1989;13:1017-26. diol 1991;18:843-91.
47. Dorbala S, Cuddy S, Falk RH. How to image cardiac amyloidosis: A prac- 68. Xiao HB, Lee CH, Gibson DG. Effect of left bundle branch block on dia-
tical approach. JACC Cardiovasc Imaging 2020;13:1368-83. stolic function in dilated cardiomyopathy. Br Heart J 1991;66:443-7.
48. Appleton CP, Hatle LK, Popp RL. Demonstration of restrictive ventricu- 69. Grines CL, Bashore TM, Boudoulas H, et al. Functional abnormalities in
lar physiology by Doppler echocardiography. J Am Coll Cardiol 1988;11: isolated left bundle branch block. The effect of interventricular asyn-
757-68. chrony. Circulation 1989;79:845-53.
49. Cohen GI, Pietrolungo JF, Thomas JD, et al. A practical guide to assess- 70. D’Souza KA, Mooney DJ, Russell AE, et al. Abnormal septal motion af-
ment of ventricular diastolic function using Doppler echocardiography. fects early diastolic velocities at the septal and lateral mitral annulus,
J Am Coll Cardiol 1996;27:1753-60. and impacts on estimation of the pulmonary capillary wedge pressure.
50. Ruan Q, Nagueh SF. Clinical application of tissue Doppler imaging in pa- J Am Soc Echocardiogr 2005;18:445-53.
tients with idiopathic pulmonary hypertension. Chest 2007;131: 71. Rapezzi C, Aimo A, Barison A, et al. Restrictive cardiomyopathy: defini-
395-401. tion and diagnosis. Eur Heart J 2022;43:4679-93.
51. Inuoe K, Andersen OS, Remme EW, et al. Echocardiographic evaluation 72. Koyama J, Ray-Sequin PA, Falk RH. Longitudinal myocardial function as-
of left ventricular filling pressure in patients with pulmonary hyperten- sessed by tissue velocity, strain, and strain rate tissue Doppler echocardi-
sion. JACC Cardiovasc Imaging 2024;17:566-7. ography in patients with AL (primary) cardiac amyloidosis. Circulation
52. Diwan A, McCulloch M, Lawrie GM, et al. Doppler estimation of left 2003;107:2446-52.
ventricular filling pressures in patients with mitral valve disease. Circula- 73. Klein AL, Hatle LK, Taliercio CP, et al. Prognostic significance of Doppler
tion 2005;111:3281-9. measures of diastolic function in cardiac amyloidosis. A Doppler echo-
53. Rossi A, Cicoira M, Golia G, et al. Mitral regurgitation and left ventricular cardiography study. Circulation 1991;83:808-16.
diastolic dysfunction similarly affect mitral and pulmonary vein flow 74. Phelan D, Collier P, Thavendiranathan P, et al. Relative apical sparing of
Doppler parameters: the advantage of end-diastolic markers. J Am Soc longitudinal strain using two-dimensional speckle-tracking echocardiog-
Echocardiogr 2001;14:562-8. raphy is both sensitive and specific for the diagnosis of cardiac amyloid-
54. Bruch C, Klem I, Breithardt G, et al. Diagnostic usefulness and prog- osis. Heart 2012;98:1442-8.
nostic implications of the mitral E/E’ ratio in patients with heart fail- 75. Liu D, Hu K, Niemann M, et al. Effect of combined systolic and diastolic
ure and severe secondary mitral regurgitation. Am J Cardiol 2007; functional parameter assessment for differentiation of cardiac amyloid-
100:860-5. osis from other causes of concentric left ventricular hypertrophy. Circ
55. Abudiab MM, Chebrolu LH, Schutt RC, et al. Doppler echocardiography Cardiovasc Imaging 2013;6:1066-72.
for the estimation of LV filling pressure in patients with mitral annular 76. Pagourelias ED, Mirea O, Duchenne J, et al. Echo parameters for differ-
calcification. JACC Cardiovasc Imaging 2017;10:1411-20. ential diagnosis in cardiac amyloidosis: A head-to-head comparison of
56. Estep JD, Vivo RP, Krim SR, et al. Echocardiographic evaluation of hemo- Deformation and Nondeformation parameters. Circ Cardiovasc Imaging
dynamics in patients with systolic heart failure supported by a 2017;10:e005588.
continuous-flow LVAD. J Am Coll Cardiol 2014;64:1231-41. 77. Klein AL, Abbara S, Agler DA, et al. American society of echocardiogra-
57. Frea S, Centofanti P, Pidello S, et al. Noninvasive assessment of phy clinical recommendations for multimodality cardiovascular imaging
hemodynamic status in HeartWare left ventricular assist device patients: of patients with pericardial disease: Endorsed by the society for cardiovas-
validation of an echocardiographic approach. JACC Cardiovasc Imaging cular magnetic resonance and society of cardiovascular computed to-
2019;12:1121-31. mography. J Am Soc Echocardiogr 2013;26:965-1012.
58. Vejpongsa P, Marcos-Abdala HG, Bhimaraj A, et al. Echocardiographic 78. Klein AL, Wang TKM, Cremer PC, et al. Pericardial diseases: Interna-
evaluation of hemodynamics in heart transplant recipients. JACC Cardi- tional position statement on new concepts and advances in multimodal-
ovasc Imaging 2021;14:313-5. ity cardiac imaging. JACC Cardiovasc Imaging 2024;17:937-88.
Journal of the American Society of Echocardiography Nagueh et al 569
Volume 38 Number 7

79. Choi JH, Choi JO, Ryu DR, et al. Mitral and tricuspid annular velocities in 95. McKee PA, Castelli WP, McNamara PM, et al. The natural history of
constrictive pericarditis and restrictive cardiomyopathy: Correlation with congestive heart failure: The Framingham study. N Engl J Med 1971;
pericardial thickness on computed tomography. JACC Cardiovasc Imag- 285:1441-6.
ing 2011;4:567-75. 96. Pfeffer MA, Shah AM, Borlaug BA. Heart failure with preserved ejection
80. Kusunose K, Dahiya A, Popovic ZB, et al. Biventricular mechanics in fraction in perspective. Circ Res 2019;124:1598-617.
constrictive pericarditis comparison with restrictive cardiomyopathy 97. L€ ofstr€
om U, Hage C, Savarese G, et al. Prognostic impact of Framingham
and impact of pericardiectomy. Circ Cardiovasc Imaging 2013;6: heart failure criteria in heart failure with preserved ejection fraction. ESC
399-406. Heart Failure 2019;6:830-9.
81. Welch TD, Ling LH, Espinosa RE, et al. Echocardiographic diagnosis of 98. Reddy YNV, Carter RE, Obokata M, et al. A simple, evidence-based
constrictive pericarditis: Mayo Clinic criteria. Circ Cardiovasc Imaging approach to help guide diagnosis of heart failure with preserved ejection
2014;7:526-34. fraction. Circulation 2018;138:861-70.
82. Qamruddin S, Alkharabsheh SK, Sato K, et al. Differentiating constriction 99. Kotecha D, Lam CSP, Van Veldhuisen DJ, et al. Heart failure with pre-
from restriction (from the Mayo clinic echocardiographic criteria). Am served ejection fraction and atrial fibrillation. J Am Coll Cardiol 2016;
Journal Cardiol 2019;124:932-8. 68:2217-28.
83. Temporelli PL, Scapellato F, Corra U, et al. Estimation of pulmonary 100. Reddy YNV, Obokata M, Gersh BJ, et al. High prevalence of occult heart
wedge pressure by transmitral Doppler in patients with chronic heart fail- failure with preserved ejection fraction among patients with atrial fibrilla-
ure and atrial fibrillation. Am J Cardiol 1999;83:724-7. tion and dyspnea. Circulation 2018;137:534-5.
84. Kusunose K, Yamada H, Nishio S, et al. Clinical utility of single-beat E/e’ 101. Rahi W, Hussain I, Nguyen DT, et al. Accuracy of clinical evaluation and
obtained by simultaneous recording of flow and tissue Doppler velocities chest X-ray for HFpEF diagnosis: comparison against right heart catheter-
in atrial fibrillation with preserved systolic function. J Am Coll Cardiol ization. Am J Cardiol 2023;206:219-20.
Img 2009;2:1147-56. 102. Chow SL, Maisel AS, Anand I, et al. Role of Biomarkers for the Preven-
85. Wada Y, Murata K, Tanaka T, et al. Simultaneous Doppler tracing of tion, assessment, and management of heart failure: a scientific statement
transmitral inflow and mitral annular velocity as an estimate of elevated from the American Heart Association. Circulation 2017;135:e1054-91.
left ventricular filling pressure in patients with atrial fibrillation. Circ J 103. Vaishnav J, Chasler JE, Lee YJ, et al. Highest obesity category associated
2012;76:675-81. with largest decrease in N-terminal pro-B-type natriuretic peptide in pa-
86. Goyal P, Almarzooq ZI, Horn EM, et al. Characteristics of hospitaliza- tients hospitalized with heart failure with preserved ejection fraction. J
tions for heart failure with preserved ejection fraction. Am J Med Am Heart Assoc 2020;9:e015738.
2016;129:635.e15-e26. 104. Anjan VY, Loftus TM, Burke MA, et al. Prevalence, clinical phenotype,
87. Tsao CW, Lyass A, Enserro D, et al. Temporal trends in the incidence of and outcomes associated with normal B-type natriuretic peptide levels
and mortality associated with heart failure with preserved and reduced in heart failure with preserved ejection fraction. Am J Cardiol 2012;
ejection fraction. JACC Heart Failure 2018;6:678-85. 110:870-6.
88. Owan TE, Hodge DO, Herges RM, et al. Trends in prevalence and 105. Gonz
alez-L
opez E, Gallego-Delgado M, Guzzo-Merello G, et al. Wild-
outcome of heart failure with preserved ejection fraction. N Engl J type transthyretin amyloidosis as a cause of heart failure with preserved
Med 2006;355:251-9. ejection fraction. Eur Heart J 2015;36:2585-94.
89. Sharma K, Kass DA. Heart failure with preserved ejection fraction: Mech- 106. Mohammed SF, Mirzoyev SA, Edwards WD, et al. Left ventricular amy-
anisms, clinical features, and therapies. Circ Res 2014;115:79-96. loid deposition in patients with heart failure and preserved ejection frac-
90. Yancy CW, Jessup M, Bozkurt B, et al. 2017 ACC/AHA/HFSA focused tion. JACC Heart Fail 2014;2:113-22.
update of the 2013 ACCF/AHA guideline for the management of heart 107. Hahn VS, Yanek LR, Vaishnav J, et al. Endomyocardial biopsy character-
failure: a report of the American College of Cardiology/American Heart ization of heart failure with preserved ejection fraction and prevalence of
Association Task force on clinical practice guidelines and the Heart Fail- cardiac amyloidosis. JACC Heart Fail 2020;8:712-24.
ure Society of America. Circulation 2017;136:e137-61. 108. Borlaug BA, Nishimura RA, Sorajja P, et al. Exercise hemodynamics
91. Yancy CW, Jessup M, Bozkurt B, et al. 2013 ACCF/AHA guideline for the enhance diagnosis of early heart failure with preserved ejection fraction.
management of heart failure: Executive summary. J Am Coll Cardiol Circ Heart Fail 2010;3:588-95.
2013;62:1495-539. 109. Simonneau G, Montani D, Celermajer DS, et al. Haemodynamic defini-
92. Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC Guidelines for the tions and updated clinical classification of pulmonary hypertension. Eur
diagnosis and treatment of acute and chronic heart failure: the task force Respir J 2019;53:1801913.
for the diagnosis and treatment of acute and chronic heart failure of the 110. Andersen MJ, Olson TP, Melenovsky V, et al. Differential hemodynamic
European Society of Cardiology (ESC) developed with the special contri- effects of exercise and volume expansion in people with and without
bution of the Heart Failure Association (HFA) of the ESC. Eur Heart J heart failure. Circ Heart Fail 2015;8:41-8.
2016;37:2129-200. 111. Esposito C, Machado P, McDonald ME, et al. Noninvasive evaluation of
93. Eriksson H, Caidaul K, Larsson B, et al. Cardiac and pulmonary causes of cardiac chamber pressures using subharmonic-aided pressure estimation
dyspnoea—validation of a scoring test for clinical-epidemiological use: with definity microbubbles. JACC Cardiovasc Imaging 2023;16:224-35.
The Study of Men Born in 1913. Eur Heart J 1987;8:1007-14. 112. Duchenne J, Pedrosa J, Van Keer JM, et al. Shear wave elastography using
94. Di Bari M, Pozzi C, Cavallini MC, et al. The diagnosis of heart failure in high-frame-rate imaging in the follow-up of heart transplantation recipi-
the community. J Am Coll Cardiol 2004;44:1601-8. ents. JACC Cardiovasc Imaging 2020;13:2304-13.