Strain Imaging

Strain imaging is referenced on nearly every disease page across this entire site — as the parameter that catches dysfunction a normal ejection fraction misses, as the bull’s-eye pattern that narrows a differential, as the technique whose layer-specific physiology explains why certain diseases affect certain segments first. This page covers the technique itself: how it works, why it replaced tissue Doppler imaging for most deformation measurement, and what the resulting patterns actually mean.

What Strain Measures

Strain, in everyday language, means “stretching” — in echocardiography, it quantifies tissue deformation, and strain rate is simply the rate at which that deformation occurs. Formally, strain (ε) describes the normalized change in length of a segment of tissue between an initial length and its length at a given time.

• Lagrangian strain uses a known, fixed initial length.
• Natural strain accounts for small temporal variations during the deformation process itself, and is considered more appropriate for the large deformations the heart actually undergoes through systole and diastole.

Speckle-Tracking Echocardiography: How It Actually Works

Speckle-tracking echocardiography (STE) tracks natural acoustic markers — “speckles” — that appear as bright and dark spots within standard B-mode myocardial images, arising from the interaction of the ultrasound beam with tissue (reflection, scattering, and interference from structures smaller than the ultrasound wavelength). A cluster of speckles, called a kernel, is tracked frame to frame, and the spatiotemporal displacement between speckles provides deformation data — without relying on Doppler velocity at all.

This is the central reason STE is largely angle-independent: unlike TDI-derived strain, STE isn’t measuring a Doppler shift along a beam direction — it’s tracking physical markers across the 2D image. This makes STE genuinely unaffected by the tethering and translational motion (including respiration-driven motion) that fundamentally limit TDI-based measurement, and substantially less sensitive to signal noise. STE’s reliability has been validated directly against sonomicrometry and cardiac MRI, with excellent correlation to both gold-standard techniques.

Technical requirements worth knowing explicitly: image quality must be high enough to track speckles accurately, foreshortening should be minimized, and frame rate matters in both directions — too low, and speckles move out of the imaging plane between frames; too high, and there’s insufficient displacement for the tracking algorithm to work reliably. A frame rate of 40 to 80 frames per second is considered optimal.

Why the Heart’s Architecture Produces Three Distinct Strain Components

The LV wall is classically described as three layers — endocardium, myocardium, epicardium — though more recent work describes a continuum of two helical fiber geometries instead: a right-handed helix in the subendocardium, transitioning gradually to a left-handed helix in the subepicardium.

This architecture isn’t incidental — it’s mechanically essential. A classic demonstration found that if LV fibers were arranged purely longitudinally, shortening would reach only about 15%; with purely circumferential fibers, about 30%; with the heart’s actual oblique spiral arrangement, 60% or more. This structure is exactly why myocardial deformation is described through three distinct components:

• Longitudinal strain — negative (shortening).
• Circumferential strain — negative (shortening).
• Radial strain — positive (lengthening, representing wall thickening).

The helical architecture also produces twisting: counterclockwise rotation of the apex and clockwise rotation of the base around the LV long axis (viewed from the apex), with the subsequent recoil — untwisting — releasing restoring forces that directly contribute to diastolic suction and facilitate early LV filling.

Why Longitudinal Strain Is the Most Sensitive Marker of Early Disease

This is worth understanding mechanistically, not just as a memorized fact. Longitudinal fibers predominate in the subendocardium — the layer most vulnerable to ischemic and non-ischemic insult, since it sits furthest from the epicardial coronary supply (the classic watershed territory). When endocardial function is impaired in isolation, while midmyocardial and epicardial function remain unaffected, the result is reduced longitudinal strain with relatively preserved circumferential strain and twist mechanics — keeping EF normal or near-normal, since EF depends heavily on the less-vulnerable deeper layers. Only with a transmural insult, or disease progression, do midmyocardial and subepicardial layers become involved simultaneously, finally reducing circumferential strain, twist, and EF together.

This single mechanistic fact explains a pattern referenced throughout this entire site: strain detecting dysfunction before EF does isn’t a statistical curiosity — it’s a direct consequence of which myocardial layer each measurement actually reflects.

A Genuinely Important Shared Limitation: Load Dependence

Both EF and longitudinal strain are load-dependent, and this is worth stating as a real limitation rather than glossing over it: increased afterload reduces longitudinal strain even in a ventricle with genuinely preserved contractility, which can be misread as true myocardial dysfunction if load isn’t considered. Myocardial work — a newer technique combining 2D STE strain data with a non-invasively estimated LV pressure-strain loop (using brachial blood pressure as a surrogate for LV pressure) — was developed specifically to address this afterload-dependence, and has been validated against invasive pressure-volume loops and PET-based myocardial metabolic imaging.

Global Longitudinal Strain: The Primary Clinical Parameter

Global longitudinal strain (GLS) — the average relative length change of the LV myocardium between end-diastole and end-systole across all segments — is the most widely used and best-validated strain parameter, genuinely more reproducible than circumferential or radial strain, which both suffer from higher measurement variability in routine practice.

• Normal cutoffs are reasonably consistent but not perfectly uniform across sources: current guidelines cite roughly −20% ± 2%; other large studies place the lower limit of normal around −18%, with severe reduction below −12%. Treat these as a converging reference range rather than a single universal number, and note that GLS is conventionally discussed by its absolute value, despite the underlying measurement being negative.
• Sex-specific reference data (from the EACVI NORRE study) give lower limits of normal longitudinal strain around −16.7% in men and −17.8% in women — with the genuinely notable finding that longitudinal strain declines with age, while circumferential and radial strain both increase with age — opposite directional trends worth knowing explicitly rather than assuming all strain components behave the same way over time.
• Layer-specific GLS can be obtained by adjusting region-of-interest width — endocardial, midwall, epicardial, or multilayer — with a natural gradient from highest strain at the endocardium to lowest at the epicardium. Endocardial GLS is the only layer-specific value provided by all vendors, a practical limitation worth knowing when comparing studies across different ultrasound platforms.
• Technical workflow: region-of-interest landmarks are placed in the apical three-, four-, and two-chamber views, with care to avoid placement on the atrial side of the mitral annulus or into the LV outflow tract — manual adjustment of segmental contours is often genuinely necessary for accurate results.

Reading the Bull’s-Eye Map: Disease-Specific Patterns Worth Recognizing

The bull’s-eye polar map — displaying regional strain values across all LV segments in a single circular projection — reveals patterns specific enough to carry real diagnostic weight, not just confirm that “strain is reduced”:

• Hypertrophic cardiomyopathy: a pronounced apex-to-base gradient of strain impairment, despite normal or supernormal EF — regional strain alterations correspond closely with fibrosis on CMR, and GLS impairment is an independent predictor of cardiac death or appropriate ICD discharge in this population.
• Cardiac amyloidosis: the relatively distinctive “apical-sparing” pattern — impairment concentrated at the base, with relative preservation of strain at the apex — genuinely useful in the differential diagnosis of LV hypertrophy, though not entirely exclusive to amyloidosis.
• Aortic stenosis: a similar base-worse, relative apical-sparing pattern to amyloidosis — directly explaining why distinguishing severe AS-related hypertrophy from infiltrative disease by strain pattern alone can be genuinely difficult, and why GLS has been incorporated into current multimodality risk stratification for considering early surgery in asymptomatic severe AS.
• Takotsubo syndrome: the inverse of the HCM/amyloid pattern — apical strain impairment with basal hypercontractility, a clean, recognizable distinguishing feature from either infiltrative or hypertrophic disease.
• Myocarditis: impaired contractility without the coronary-territory distribution seen in ischemic disease — consistent with the broader “non-coronary wall motion pattern” principle that distinguishes inflammatory and infiltrative disease from true ischemia throughout this site.
• Coronary artery disease: regional strain impairment that recapitulates the actual coronary territory distribution — see Coronary Territories and Infarct Localization for the underlying map this pattern follows. Longitudinal strain is particularly accurate for detecting subtle ischemia in acute coronary occlusion and non-ST-elevation ACS that a standard wall motion score index can miss, precisely because it reflects the endocardial fibers most sensitive to ischemic insult.
• Physiologic versus pathologic hypertrophy: GLS is preserved in athletic, physiologically hypertrophied hearts, but reduced in hypertensive heart disease even before overt LV hypertrophy develops — a genuinely useful discriminator when wall thickness alone is ambiguous.

How to Approach Strain Imaging in Practice: A Sequence

1. Ensure adequate image quality and an appropriate frame rate (40–80 fps) before acquiring data — poor tracking produces unreliable numbers regardless of software sophistication.
2. Obtain GLS from the standard three apical views, placing region-of-interest landmarks carefully and adjusting segmental contours manually when needed.
3. Interpret the absolute GLS value against a converging reference range (roughly −18% to −20% as the lower limit of normal) rather than a single hard cutoff, and note sex and age as genuine modifiers of normal range.
4. Read the bull’s-eye map for pattern, not just the global number — apex-to-base gradients, apical sparing, and coronary-territory distributions each carry distinct diagnostic meaning.
5. Consider afterload before attributing reduced strain to intrinsic dysfunction, particularly in conditions with elevated afterload (severe hypertension, severe AS) — myocardial work can help resolve genuine ambiguity.
6. Use strain to detect what EF alone would miss — subclinical dysfunction in early cardiomyopathy, chemotherapy-related cardiotoxicity surveillance, and subtle ischemia in ACS are all scenarios where strain’s layer-specific sensitivity provides real, additive diagnostic value.

Clinical Importance

Strain imaging’s real value isn’t that it’s “more sensitive” in some abstract sense — it’s that the layer-specific physiology explained on this page gives every disease-specific pattern referenced throughout this site a genuine mechanistic basis. The apex-to-base gradient in HCM, apical sparing in amyloidosis, the coronary-territory recapitulation in ischemic disease — none of these are arbitrary associations; each reflects exactly which myocardial layer and which region a given disease process affects first, made visible by a technique built specifically to measure deformation rather than just overall chamber-level motion.