Author + information
- aUCLA Cardiac Arrhythmia Center, UCLA Health System, David Geffen School of Medicine at UCLA, Los Angeles, California
- bDepartment of Cardiology, the Royal Melbourne Hospital, University of Melbourne, Melbourne, Australia
- ↵∗Address for correspondence:
Dr. Noel G. Boyle, UCLA Cardiac Arrhythmia Center, UCLA Health System, David Geffen School of Medicine at UCLA, 100 UCLA Medical Plaza, Suite 660, Los Angeles, California 90095-7392.
More than 100 years after its introduction into clinical practice, the electrocardiogram (ECG) remains the most easily accessible, inexpensive, and central diagnostic tool in cardiology. Well before the advent of echocardiography, the ECG was the only method to diagnose cardiac chamber hypertrophy.
An important aspect of electrocardiographic analysis is the assessment of left ventricular hypertrophy (LVH). The best known and widely used Sokolow-Lyon diagnostic criterion for LVH, based on QRS voltages, was published in 1949 (1). This measure was derived from a comparison of ECGs in 147 patients with hypertension with a mean blood pressure of 197/117 mm Hg and a mean increase in the transverse diameter of the heart of 15.8% on chest radiograph versus the ECGs of 151 healthy control subjects. The Romhilt Estes point scoring system for the diagnosis of LVH (≥5 of 13 points) was published in 1968 and took into account other features of the ECG, such as left atrial P waves, left-axis deviation, QRS duration, ST-T wave changes, and the intrinsicoid deflection. This scoring system was based on a correlation of pre-mortem ECGs with pathologic findings of hypertrophied heart at autopsy (2). In 1985, Casale et al. (3) published the “Cornell criteria,” also based on QRS voltages in 414 subjects, using echocardiogram-derived left ventricular (LV) mass index as the gold standard. These were extended to include sex-specific criteria in 1987 (4). All these criteria have a low sensitivity of approximately 20% to 30%, and a high specificity of about 90%, which can also vary with the type of heart disease (5). Although the Cornell criteria have higher specificity, the Sokolow-Lyon criteria are in wider use.
In formulating the Cornell criteria, Casale et al. (3)suggested that the sensitivity of electrocardiographic detection of LVH can be improved by incorporating changes due to age, obesity, and smoking, in addition to sex (3,4). This observation was also confirmed in the Framingham study of LVH criteria and echocardiogram correlation (6). This study of 4,684 subjects, published in 1990, determined the sensitivity and specificity of electrocardiographic criteria for LVH. The study found that the overall sensitivity for the electrocardiographic diagnosis of LVH was 6.9%, with a specificity of 98.8%. Previously mentioned factors such as sex, age, obesity, and smoking status affected sensitivity, but their incorporation did not affect specificity.
Electrocardiographic changes of LVH have been correlated with LV mass at autopsy, assessment by chest radiograph, LV angiography, echocardiography, and, in recent years, with cardiac magnetic resonance imaging (7). Because of its easy availability, echocardiography remains the main investigation for assessing LV mass and allows easy comparison with the electrocardiographic criteria for LVH. The current American Heart Association/American College of Cardiology/Heart Rhythm Society Recommendations for Standardization and Interpretation of the Electrocardiogram, Part V, statement lists a total of 36 criteria (single or combined) that can be used for LVH diagnosis and emphasizes that none is superior. The sensitivities of the various criteria are low (<50%) with specificities which approach 90% (8).
In this issue of the Journal, Peguero et al. (9) have developed a novel electrocardiographic criterion for the diagnosis of LVH. The authors used a sum of the amplitude of the deepest S wave in any lead (SD) and the S-wave in V4 as their measured characteristic (SD + SV4). LVH is diagnosed on the ECG when SD + SV4 is ≥2.3 mV in women and ≥2.8 mV in men. When the deepest S-wave was found in V4, the value was doubled. In addition, the authors found that the deepest S-wave in V3 or V4 correlated well with LVH. These electrocardiographic parameters were correlated with transthoracic echocardiograms, and LV mass was calculated by using the Devereux formula (10). LVH was defined as a LV mass index >115 g/m2 in men and >95 g/m2 in women.
The patient population for the study (9) consisted of all patients admitted to the authors’ institution over a 1-month period who had both ECGs and echocardiograms at the same hospitalization. The test cohort consisted of 100 patients with hypertension, defined as systolic blood pressure >180 mm Hg and diastolic blood pressure >120 mm Hg. Although originally divided into hypertensive emergency (patients with end-organ damage) and hypertensive urgency (without end-organ damage) groups, both groups were analyzed together because no differences were found. Six of the 100 patients were excluded because their echocardiograms were unsatisfactory.
These 94 test group patients were compared with a validation group of 122 patients (9). These patients were referred to the institution for echocardiograms and also had ECGs. The validation cohort was older, had a higher incidence of hypertension, and had more co-morbidities. The incidence of LVH was 32% in the test cohort and 42% in the validation cohort. The sensitivity, specificity, and accuracy of different criteria were estimated by using area under the curve (AUC) analysis. Figure 1 of the paper shows AUC analyses of various electrocardiographic measurements, which included R in aVL, R in L1, Sokolow-Lyon and Cornell voltage criteria, and the suggested new SD +SV4–Peguero-Lo Presti criterion. The new proposed criterion was superior to all others.
What is different about this proposed criterion compared with the many others in the literature? Peguero et al. (9) suggest that other criteria emphasize the amplitude of the R-wave and initial 30 ms vectors of ventricular depolarization, while their emphasis on the S-wave parallels the later vectors of depolarization, which makes it more sensitive in detecting milder forms of LVH. This theory is illustrated in Figure 5 of their paper.
The attractive feature of the SD + SV4 formula as suggested by the authors (9) is its simplicity. Even assessment of the largest S-wave in any lead (SD) performed well, with an AUC of 0.80 and a p value of <0.001. However, relying on electrocardiographic voltage criteria for estimation of LV mass has inherent limitations. As the authors point out, LVH on the ECG is dependent on electrical voltage changes and correlation with the LV mass does not hold well for increases in LV mass due to infiltrative disease such as amyloid or where there is a significant amount of LV fibrosis. Also, the ECG and echocardiogram correlation holds for concentric but not for eccentric LV hypertrophy (11).
Can the assessment of LV mass on echocardiogram be taken as the final arbitrator of its presence or absence? Although the echocardiogram is an important and easily available tool, it is not as sensitive as cardiac magnetic resonance imaging. Also, the presence of increased LV voltages may be an adverse factor for increases in the incidence of sudden cardiac death even if the echocardiogram does not show an increase in the LV mass. The Oregon Sudden Unexpected Death Study has shown that electrical changes of LVH without echocardiographic changes may be an independent risk factor for sudden cardiac death (12,13).
Despite some limitations of this study, which are outlined in the paper, the authors have produced an efficient and straightforward algorithm for the assessment of LVH (9). In this era of high-tech medicine it is refreshing to emphasize the value of a simple diagnostic tool such as the ECG.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation
- Casale P.N.,
- Devereux R.B.,
- Kligfield P.,
- et al.
- Casale P.N.,
- Devereux R.B.,
- Alonso D.R.,
- Campo E.,
- Kligfield P.
- Levy D.,
- Labib S.B.,
- Anderson K.M.,
- Christiansen J.C.,
- Kannel W.B.,
- Castelli W.P.
- Bacharova L.,
- Ugander M.
- Hancock E.W.,
- Deal B.J.,
- Mirvis D.M.,
- et al.
- Peguero J.G.,
- Lo Presti S.,
- Perez J.,
- Issa O.,
- Brenes J.C.,
- Tolentino A.