Abstract 16128: Electrocardiographic Frontal T-Axis and QRS-T Angle Significantly Improve Risk Prediction of Cardiovascular Death

Abstract

Introduction: Frontal T-axis and QRS-T angle are simple and easily obtained electrocardiographic measures. Abnormalities may represent subclinical ischemia and/or structural heart disease. No previous studies have examined their ability to improve CVD prediction and age/sex differences.

Methods: We studied 7865 adults aged 40-90 years without CHD from the NHANES III (1988-1994) study. Frontal T-Axis and QRS-T angle were measured from standard 12-lead ECG that was automatically processed at a central core ECG laboratory. Frontal T-axis deviation was measured as the absolute difference from 45 degrees, and frontal QRS-T angle was calculated as the absolute difference of the frontal QRS and T vector axes. Mortality and cause of death through 2006 were assessed.

Results: The mean age (SD) was 60 (13.5) years, and 1156 CVD deaths occurred in a median follow-up time of 14 years. T-axis deviations and QRS-T angles of greater than 30 degrees occurred in 22% and 40% of the sample, respectively. In multivariate analysis adjusting for age, sex, race-ethnicity, current/past smoking, systolic BP, total cholesterol, and diabetes, an incremental 30 degree deviation in frontal T-axis was associated with a 39% increase in the risk of CVD mortality (HR 1.39, 95% CI 1.31 -1.47). Similarly, a 30 degree increase in frontal QRS-T angle associated with a 38% increased risk (HR of 1.38, 95% CI, 1.33 - 1.44). Each measure was more pronounced in adults ≤ 60 years of age (age interaction p<0.05), with an adjusted HR of T-axis and QRS-T angle of 1.49 (95% CI, 1.30 [[Unable to Display Character: &#8211;]] 1.71) and 1.43 (95% CI, 1.28 [[Unable to Display Character: &#8211;]] 1.59), respectively. With both variables together in an adjusted model, significant sex interactions were found in the younger group only: T-axis was only predictive in men (n=1926, HR 1.34, 95% CI 1.09 -1.66) and QRS-T angle was only predictive in women (n=2180, HR 1.71, 95% CI 1.31-2.22). Together, they improved the C-statistic from 0.780 to 0.807 (p=0.002). Also, the net reclassification index improved by 6.5% when using traditional risk categories (<6%, 6-20%, and >20% risk), and by 37% when using a continuous reclassification scheme.

Conclusion: Frontal T-axis and QRS-T angle are simple ECG markers that could be used to improve prediction of CVD death, particularly in people ≤ 60 years of age.