Electrocardiographic Criteria for the Diagnosis of Left Ventricular Hypertrophy

AbstractBackground

Current electrocardiographic (ECG) criteria for the diagnosis of left ventricular hypertrophy (LVH) have low sensitivity.

Objectives

The goal of this study was to test a new method to improve the diagnostic performance of the electrocardiogram.

Methods

The study was divided into 2 groups, a test and a validation cohort. In the test cohort, 94 patients were analyzed, including 47 with the diagnosis of hypertensive crisis and 47 with normal blood pressure at admission. Echocardiography was used to estimate the left ventricular mass index. Area under the curve (AUC) analysis was used for comparison of single and combined leads. The McNemar test was used to assess agreement among the ECG criteria against the left ventricular mass index. The proposed ECG criteria involved measuring the amplitude of the deepest S wave (S_D) in any single lead and adding it to the S wave amplitude of lead V₄ (SV₄). Currently accepted LVH ECG criteria such as Cornell voltage and Sokolow-Lyon were used for comparison. The validation cohort consisted of 122 consecutive patients referred for an echocardiogram regardless of the admitting diagnosis.

Results

The S_D was the most accurate single lead measurement for the diagnosis of LVH (AUC: 0.80; p < 0.001). When both cohorts were analyzed, the S_D + SV₄ criteria outperformed Cornell voltage with a significantly higher sensitivity (62% [95% confidence interval [CI]: 50% to 72%] vs. 35% [95% CI: 24% to 46%]). The specificities of all the criteria were ≥90%, with no significant difference among them.

Conclusions

The proposed criteria for the ECG diagnosis of LVH improved the sensitivity and overall accuracy of the test.

http://ift.tt/2oab6DK