How should left ventricular ejection fraction (LVEF) be assessed by ventriculography and correlated with echocardiographic measurements?

LVEF Assessment: Ventriculography vs Echocardiography Correlation

Echocardiography and contrast ventriculography measurements of LVEF show good correlation (r=0.73-0.90) but can vary widely in individual patients, with approximately one-third differing by >10 percentage points—making them not truly interchangeable despite common clinical practice. 1

Key Measurement Hierarchy

When multiple LVEF determinations are available, the ACC/AHA guidelines establish this hierarchy 2:

• Radionuclide ventriculography (highest priority)
• Magnetic resonance imaging (MRI)
• Echocardiography (2D or 3D)
• Contrast left ventriculography
• Technetium myocardial perfusion imaging

Correlation Data Between Modalities

Echocardiography vs Contrast Ventriculography

The most robust comparison study found 1:

• Correlation coefficient: r = 0.73
• No systematic bias between methods
• Critical limitation: ~33% of patients show differences >10 percentage points
• ~5% of patients show differences >20 percentage points
• These variations occur regardless of timing between tests, baseline LVEF level, or test order

Echocardiography vs Radionuclide Ventriculography

In post-MI patients at 6 months 3:

• Correlation coefficient: r = 0.90
• Mean difference: 1% (p=0.04)
• 95% limits of agreement: -13.0% to 10.3%
• Only 1.2% of patients misclassified between normal and significantly reduced function
• Overall accuracy: 86% (kappa = 0.78)

Clinical Implications

When Consistency Matters Most

The same imaging modality and method must be used for serial measurements during treatment and surveillance 4. This is critical because:

• LVEF values differ significantly across techniques
• Digital images should be compared with previous ones to minimize interobserver variability
• Treatment decisions (ACE inhibitors, device therapy) depend on accurate LVEF thresholds 5

Acceptable Measurement Methods by Echocardiography

Per ACC/AHA/ASE guidelines 6, 7:

Quantitative methods (preferred):

• Calculated LVEF: Using formula LVEF = ([EDV-ESV]/EDV) × 100
• Biplane method of disks (Simpson's method) - recommended standard
• 3D echocardiography when feasible (enhanced reproducibility)

Visual estimate:

• Correlates well with formal methods (r=0.86-0.92) 8, 9
• May be most accurate when performed by experienced observers
• Reported as percentage or midpoint of range

Qualitative Categories

When quantitative measurement unavailable 2:

• Normal: LVEF >50%
• Mildly diminished: LVEF 41-49%
• Moderately diminished: LVEF 26-40%
• Severely diminished: LVEF ≤25%

Common Pitfalls to Avoid

1. Assuming interchangeability: Do not treat LVEF from different modalities as equivalent for individual patient decisions, despite good population-level correlation

2. Switching modalities during follow-up: This introduces measurement variability that may be misinterpreted as clinical change

3. Over-reliance on single threshold: LVEF <50% is highly unlikely to be normal 10, but values 45-55% have poor reproducibility and limited prognostic discrimination 11

4. Ignoring image quality: Poor endocardial definition significantly impacts accuracy—consider contrast enhancement or alternative modality 12

5. Regional wall motion abnormalities: Linear measurements (fractional shortening) become unreliable; volumetric methods required 13

Practical Algorithm for Clinical Use

For initial diagnosis:

• Transthoracic echocardiography is the preferred first-line test 12, 14
• Use biplane Simpson's method when feasible
• Consider 3D echo in experienced labs for enhanced reproducibility

When echocardiography is inadequate 14:

• Radionuclide ventriculography or MRI can assess LVEF and volumes
• Contrast ventriculography during cardiac catheterization if already planned

For serial monitoring 4:

• Use identical modality and method throughout
• Compare actual images, not just numerical values
• Recognize that differences <10 points may reflect measurement variability rather than true change

When values are borderline or discordant 15:

• CMR as second method to corroborate or dismiss echocardiographic values
• Particularly important when LVEF is near treatment thresholds (35-40%)