Chapter 8 — Left Ventricular Diastolic Function and Filling Pressures

📚 Guideline basis

Nagueh SF, Sanborn DY, Oh JK, et al. Recommendations for the Evaluation of LV Diastolic Function by Echocardiography and for HFpEF Diagnosis: An Update from the American Society of Echocardiography. J Am Soc Echocardiogr 2025;38:537–69. This chapter is written against the 2025 algorithm, which replaces the 2016 ASE/EACVI algorithm and eliminates the indeterminate category. Where the ICU literature was built on the 2016 thresholds, both are given.

Pathophysiology & Mechanisms

Diastole comprises four phases: isovolumic relaxation, rapid early filling, diastasis, and atrial contraction. Two independent properties determine LV diastolic pressure at any volume:

• Relaxation — an active, ATP-dependent process of calcium re-sequestration by SERCA2a and cross-bridge detachment. Quantified invasively as the time constant τ; impaired when τ > 48 ms. Relaxation is load-dependent, and its sensitivity to afterload is steepest in ventricles with systolic dysfunction.
• Compliance / chamber stiffness — a passive property of sarcomeric titin isoform composition, interstitial collagen, chamber geometry, and wall thickness. Abnormal when the chamber stiffness constant > 0.015.

Superimposed are restoring forces (elastic recoil producing diastolic suction) and pericardial constraint — critical in the ICU, where an acutely dilated RV shifts the septum and raises LV diastolic pressure at any given LV volume without any change in intrinsic LV compliance. This is diastolic dysfunction produced entirely outside the left ventricle.

The invasive reference points

LVEDP and mean LAP are not the same pressure. Elevated LAP is always accompanied by elevated LVEDP in diastolic dysfunction, but LVEDP may be elevated while mean LAP is normal — the earliest stage. Throughout, "LV filling pressure" means mean LAP or its correlates.

Which echo parameter reflects which pressure:

Diagnostic Synthesis — the 2025 algorithm

Step 1: is diastolic dysfunction present?

Assess e′ as the marker of impaired relaxation:

• Septal e′ ≤ 6 cm/s, or lateral e′ ≤ 7 cm/s, or average e′ ≤ 6.5 cm/s → abnormal relaxation, irrespective of age.

Age-specific alternatives:

Then assess markers of LA/LV remodelling and elevated LAP:

• Average E/e′ > 14
• LARS ≤ 18%
• E/A ≤ 0.8 or ≥ 2
• LAVi > 34 mL/m² (after excluding athletic LA enlargement, anaemia, AF/flutter, and mitral valve disease)
• Alternatively, LV mass index > 95 g/m² (women) or > 115 g/m² (men)

Diastolic dysfunction is present if: e′ is reduced and ≥1 Step-2 marker is present; or e′ is preserved and ≥2 Step-2 markers are present.

Step 2: is LAP elevated? (main algorithm, sinus rhythm)

Three primary variables:

1. Reduced e′: septal ≤ 6, or lateral ≤ 7, or average ≤ 6.5 cm/s
2. Increased E/e′: septal ≥ 15, or lateral ≥ 13, or average ≥ 14
3. Increased TR velocity ≥ 2.8 m/s, or PASP ≥ 35 mmHg

Step 3: secondary variables when the primary set is discordant

Any one of:

• Pulmonary vein S/D ≤ 0.67 (equivalent to a systolic filling fraction ≤ 40%)
• LARS ≤ 18%
• LAVi > 34 mL/m²
• Alternatively IVRT ≤ 70 ms

≥1 present → elevated LAP, then grade by E/A. None present → LAP normal; if the patient is symptomatic, proceed to diastolic exercise echocardiography.

If none of these are obtainable or reliable, supplemental methods: pulmonary regurgitation end-diastolic velocity ≥ 2 m/s (or PA diastolic pressure ≥ 16 mmHg); mitral inflow L-wave velocity ≥ 50 cm/s; Ar−A duration > 30 ms; decrease in mitral E/A ≥ 50% with Valsalva.

Exclusions — reproduced in full

The 2025 main algorithm must not be applied to patients with: atrial fibrillation; any degree of mitral stenosis; moderate or severe mitral annular calcification; severe primary mitral regurgitation; mitral valve repair, replacement, or transcatheter edge-to-edge repair; LV assist device; non-cardiac pulmonary hypertension; heart transplant recipients; pericardial constriction. It is also not applicable in children, normal pregnancy, or the intraoperative setting.

The ICU reader should note what this exclusion list means in practice: a large proportion of ICU patients fall into at least one exclusion category, most commonly atrial fibrillation.

Special populations relevant to the ICU

Atrial fibrillation

Indicators of elevated filling pressure:

• Mitral E deceleration time < 160 ms (in depressed LVEF)
• Peak acceleration rate of mitral E ≥ 1900 cm/s²
• IVRT ≤ 65 ms
• Deceleration time of pulmonary venous D velocity ≤ 220 ms
• E/Vp ≥ 1.4
• Septal E/e′ ≥ 11
• Peak TR velocity > 2.8 m/s

Average ≥ 5–10 beats, or match preceding R-R intervals.

Sinus tachycardia

• Predominant early filling pattern with depressed LVEF
• IVRT ≤ 70 ms — specificity 79%
• Pulmonary vein systolic filling fraction ≤ 40% — specificity 88%
• Average E/e′ > 14 — high specificity, low sensitivity
• When E and A are fused, the compensatory pause after an extrasystole often separates them; use that beat.

Pulmonary hypertension (groups I, III–V)

Septal flattening invalidates septal and average E/e′. Use lateral E/e′: > 13 indicates elevated LV filling pressure, < 8 normal, 8–13 indeterminate. E/A ≤ 0.8 with E ≤ 50 cm/s favours pre-capillary PH; E/A ≥ 2 with reduced e′ favours post-capillary; between those, lateral E/e′ > 13, LAVi > 34 mL/m², and LARS < 16–18% favour group II.

Mitral regurgitation and mitral stenosis

Restrictive physiology and constriction

Restrictive cardiomyopathy: E/A > 2.5, deceleration time < 150 ms, IVRT < 50 ms, septal and lateral e′ 3–4 cm/s. In advanced cardiac amyloidosis, the "5-5-5" sign — mitral annular s′, e′ and a′ all < 5 cm/s — and apical sparing on strain polar plot.

Constriction: respirophasic septal shift, septal bounce, mitral inflow variation > 25% and tricuspid > 40% with respiration, expiratory hepatic vein end-diastolic flow reversal (reversal/forward ratio ≥ 0.8), septal e′ usually > 7 cm/s, and annulus reversus (septal e′ exceeding lateral e′). A normal septal e′ in a patient carrying a heart failure diagnosis should raise suspicion of constriction.

ICU-Specific Limitations

⚠️ Evidence quality

The 2025 ASE algorithm was validated in ambulatory and acute hospital-care adults in sinus rhythm. It was not validated in mechanically ventilated ICU patients. ICU-specific diastolic assessment rests on a smaller literature and the practical position is: use E/e′ and TR velocity as the workhorses, treat grading as approximate, and interpret against the clinical trajectory rather than as a fixed diagnosis.

Therapeutic Logic

Diastolic dysfunction changes ICU management in four specific ways:

1. Fluid tolerance. A stiff ventricle converts small volume increments into large increases in LAP and pulmonary oedema. Elevated E/e′ with a dilated IVC is a signal to stop fluid regardless of fluid responsiveness (Chapter 12).
2. Rate and rhythm. Filling depends on diastolic time and on atrial contraction. Restoring sinus rhythm and controlling rate to 60–90/min is a haemodynamic intervention in the stiff ventricle, not merely an arrhythmia intervention. Loss of the atrial kick can reduce cardiac output by 20–30% in restrictive physiology.
3. Weaning failure. Weaning-induced pulmonary oedema is a diastolic phenomenon: negative intrathoracic pressure raises venous return and LV afterload simultaneously in a ventricle that cannot accommodate it. E/e′ rising above 14 during a spontaneous breathing trial identifies it (Chapter 34).
4. Prognosis. Grade 3 diastolic dysfunction is associated with poor outcome across populations.

🛑 Critical pitfall: Applying the algorithm to a patient in atrial fibrillation. It is explicitly excluded, and doing so produces a confidently wrong filling pressure. AF has its own parameter set.

🛑 Critical pitfall: Using septal E/e′ in a patient with acute cor pulmonale. Septal flattening depresses septal e′ for purely mechanical reasons, inflating septal E/e′ and generating a false diagnosis of elevated LAP in a patient whose real problem is right-sided.

🛑 Critical pitfall: Treating "elevated LAP" as "hypervolaemia". Elevated LAP in a patient with a stiff, small, hypertensive ventricle can coexist with intravascular volume depletion. The measurement describes pressure, not volume.

• 💡 Clinical pearl: In a ventilated tachycardic patient where E and A are fused and pulmonary vein flow is unobtainable, TR velocity ≥ 2.8 m/s plus lateral E/e′ ≥ 13 is the most robust obtainable pair.
• 💡 Clinical pearl: LARS ≤ 18% is specific but insensitive for elevated LAP in preserved-EF patients; a normal LARS does not exclude high filling pressure.
• 💡 Clinical pearl: A normal septal e′ in a patient labelled "heart failure" should prompt evaluation for constrictive pericarditis, particularly post-cardiac-surgery and post-radiation.

References

1. Nagueh SF, Sanborn DY, Oh JK, et al. 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. J Am Soc Echocardiogr 2025;38:537–69. doi:10.1016/j.echo.2025.03.011
2. Nagueh SF, Smiseth OA, Appleton CP, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography: an update from the ASE and the EACVI. J Am Soc Echocardiogr 2016;29:277–314. PMID 27037691.
3. Pessoa et al. Diastolic function assessment in the intensive care unit. J Am Soc Echocardiogr 2020. PMID 32007409.
4. Kasner M, Westermann D, Steendijk P, et al. Utility of Doppler echocardiography and tissue Doppler imaging in the estimation of diastolic function in heart failure with normal ejection fraction. Circulation 2007;116:637–47.