Limits of the spatial ventricular gradient and QRST angles in patients with normal electrocardiograms and no known cardiovascular disease stratified by age, sex, and race

Abstract

IntroductionMeasurement of the spatial ventricular gradient (SVG), spatial QRST angles, and other vectorcardiographic measures of myocardial electrical heterogeneity have emerged as novel risk stratification methods for sudden cardiac death and other adverse cardiovascular events. Prior studies of normal limits of these measurements included primarily young, healthy, White volunteers, but normal limits in older patients are unknown. The influence of race and body mass index (BMI) on these measurements is also unclear.MethodsNormal 12‐lead electrocardiograms (ECGs) from a single center were identified. Patients with abnormal cardiovascular, pulmonary, or renal history (assessed by International Classification of Disease [ICD‐9/ICD‐10] codes) or abnormal cardiovascular imaging were excluded. The SVG and QRST angles were measured and stratified by age, sex, and race. Multivariable linear regression was used to assess the influence of age, BMI, and heart rate (HR) on these measurements.ResultsAmong 3292 patients, observed ranges of SVG and QRST angles (peak and mean) differed significantly based on sex, age, and race. Sex differences attenuated with increasing age. Men tended to have larger SVG magnitude (60.4 [46.1–77.8] vs. 52.5 [41.3–65.8] mv*ms, p < .0001) and elevation, and more anterior/negative SVG azimuth (−14.8 [−25.1 to −4.3] vs. 1.3 [−9.8 to 10.5] deg, p < .0001) compared to women. Men also had wider QRST angles. Observed ranges varied significantly with BMI and HR. SVG and QRST angle measurements were robust to different filtering bandwidths and moderate fiducial point annotation errors, but were heavily affected by changes in baseline correction.ConclusionsAge, sex, race, BMI, and HR significantly affect the range of SVG and QRST angles in patients with normal ECGs and no known cardiovascular disease, and should be accounted for in future studies. An online calculator for prediction of these “normal limits” given demographics is provided at https://bivectors.github.io/gehcalc/.