Strain as Predictor of Outcomes: ECG LVH Criteria beyond Voltage

The 12-lead ECG was the first noninvasive tool available to cardiologists to make the premortem diagnosis of left ventricular hypertrophy (LVH). ECG diagnosis of hypertrophy is frequently reported and remains an important part of evaluating cardiac patients to this day.

Most of the dozens of ECG criteria developed in the last fifty years for diagnosing LVH have relied on the magnitude of the voltage in limb and precordial leads. The Cornell and Sokolow-Lyon criteria, which exclusively utilize ECG voltage, track echocardiographic left ventricular mass better, but their sensitivity is still unacceptably low—to the point where some authors have declared the assessment of ECG LVH clinically useless.

However, recent studies suggest that information contained within the standard ECG beyond voltage can provide important insights into myocardial structure and function.

Incorporating ECG Information Beyond Voltage

Romhilt and Estes published a different approach to utilizing ECG findings to diagnose hypertrophy. Their system, called the R-E score, incorporated multiple ECG components, including ST changes, P wave changes, and QRS axis. ECGs with five points from this six-point scoring system are considered to have LVH.

As Estes pointed out in the American Heart Journal in 2015, the sensitivity was low, and "attempts to improve the sensitivity of R-E score proved fruitless, as each such modification led to an unacceptable increase in false positives. The advent of and widely increased availability of imaging technology has made optimizing current ECG LVH criteria less relevant."

If current LVH criteria are decreasingly relevant, what should cardiologists be paying attention to in the ECG?

Defining ECG Strain

An important component of the R-E score is ST strain, which Estes defined as "ST segment and T wave in opposite direction to QRS in V5 or V6, without digitalis." It garnered three points, equivalent in value to the R-E voltage component (R or S wave in any limb lead ≥2 mv, or S wave in V1 or V2 ≥3 mv, or R wave in V5 or V6 ≥3 mv).

ECG strain has also been more recently defined in the Journal of the American Heart Association (JAHA) as "coexistence, in leads I, II, aVL, or V3 to V6 of ST-segment horizontal or downward sloping depression ≥0.05 mV plus negative T wave" (Minnesota code 4–1 or 4–2 and 5–1 or 5–2).

The precise cause of underlying ECG strain is unknown. It has been hypothesized in a Circulation study that it is a marker of "LV decompensation," and recent studies confirm a high association with markers of LV fibrosis or scarring.

This ECG* demonstrates strain pattern in leads I, aVL, V5, and V6. The patient had severe concentric LVH by echo, but no ECG voltage criteria for LVH.


This ECG* demonstrates a strain pattern isolated to V5 and V6. In addition, classic voltage criteria for LVH are present—Cornell criteria >28 mm in RaVL (19 mm) and S V3 12 mm—along with left axis deviation and left atrial enlargement, fulfilling Romhilt-Estes criteria for ECG LVH. The patient had moderate concentric LVH by echo, normal coronaries, and recent decompensated heart failure.


To learn more about the power of the ECG in today's clinical landscape, browse our Diagnostic ECG Clinical Insights Center.

ECG Strain and Prognosis

Multiple studies have demonstrated that ECG strain is an important independent prognostic finding beyond its contribution to diagnosing LVH.

Prior to good antihypertensive therapy, the prevalence of ECG strain was noted in a 2008 Cardiovascular Journal of Africa review to be as high as 36 percent, but other studies indicate a much lower prevalence (as low as 2.1 percent) depending on the population studied.

The 2008 review of ECG strain found that the sensitivity of strain for LVH ranged from 3.8 to 50 percent, whereas specificity was high, ranging from 89.8 to 100 percent. More importantly, the strain pattern "identifies cardiac patients at higher risk of cardiovascular-related as well as all-cause morbidity and mortality. ECG strain pattern is associated with a higher cardiovascular risk, abnormal LV structure and function, incident heart failure, stroke, and coronary artery disease."

ECG Strain in Aortic Stenosis

More recent studies utilizing cardiac MRI have examined ECG strain in relation to myocardial tissue changes in patients with aortic stenosis (AS). As AS patients transition from hypertrophy to heart failure, there is progressive myocyte cell death and myocardial fibrosis. Cardiovascular magnetic resonance (CMR) can quantify and visualize this mid-wall fibrosis, which has been associated with LV decompensation and adverse prognosis in patients with AS.

The Circulation study followed 140 Scottish patients with AS for 10.6 years. Compared to patients without ECG strain, patients with ECG strain had more severe AS, higher LV mass index, high troponin I levels, and more diffuse fibrosis by CMR.

All patients with ECG strain had mid-wall late gadolinium enhancement, which was independently associated with ECG strain. ECG strain was an independent predictor of aortic valve replacement or cardiovascular death.

These same investigators went on to develop and validate a novel clinical score using variables associated with mid-wall myocardial fibrosis on CMR, which was published in 2016 in the European Heart Journal. The primary outcome predicted by the score was aortic-stenosis-related events (CV death, heart failure, angina, dyspnea, or syncope).

The factors that were found to be independent predictors in patients with aortic stenosis were age, sex, maximum aortic velocity from Doppler, high-sensitivity troponin I concentration, and ECG strain (presence of ≥1 mm concave downsloping ST depression with asymmetrical T-wave inversion in the lateral leads).

AS related events were more than tenfold higher in patients with a high score versus those with a low score.

Impact of ECG Strain on Long-Term Mortality after Surgical AVR for AS

Findings in the American Journal of Cardiology explore how ECG strain affects long-term mortality after surgical aortic valve replacement (AVR) for AS.

The researchers followed 390 patients for an average of 4.8 years, and 28 percent had ECG strain. Compared to those without ECG strain, the ECG strain group had significantly lower body mass index, higher mean transaortic pressure gradient, and Cornell-product ECG LVH.

They also had significantly lower 8-year survival rates than those without ECG strain. ECG strain was associated with a fourfold increase in the risk of long-term mortality after isolated AVR, regardless of preoperative LVH.

ECG Strain in Apparently Healthy Individuals

The JAHA study examined the significance of ECG strain in a cohort of 6,441 participants without cardiovascular disease in the Multi-Ethnic Study of Atherosclerosis trial.

ECG strain was found to be an independent predictor of all‐cause death, incident heart failure, myocardial infarction, and incident cardiovascular disease (CVD) over 10 years in multiethnic participants without past CVD. In addition, ECG strain was associated with the development of LV concentric remodeling, a decline in LV systolic function, and LV myocardial scar (demonstrated on CMR) after 10 years of followup, independent of voltage criteria for LVH.

The authors concluded that ECG strain is an "early marker of LV structural remodeling that contributes to the development of adverse cardiovascular events." This study extends the prognostic value of ECG strain in predicting adverse cardiovascular events to apparently healthy individuals in multiethnic populations.

ECG strain provides powerful insights into the status of the LV myocardium in populations at risk of LV decompensation: it appears to be a specific marker of mid-wall left ventricular myocardial fibrosis and predictive of adverse clinical outcomes in seemingly normal patients and patients with AS before and after AVR.

Since strain identifies early myocardial decompensation, clinicians should consider treating individuals with ECG strain aggressively for modifiable risk factors of LVH, regardless of the presence of ECG voltage criteria for LVH.

*Anonymized images provided by and published with permission from Dr. Anthony Pearson.