TY - JOUR
T1 - Prediction of left ventricular ejection fraction using a unique method of chest x-ray and ECG analysis
T2 - A noninvasive index of cardiac performance based on the concept of heart volume and mass interrelationship
AU - Ostojic, Miodrag C.
AU - Young, James B.
AU - Hess, Kenneth R.
N1 - Funding Information:
From Belgrade University and Baylor College of Medicine. Supported in part by a visiting professorship grant from Houston ogy Associates, The Research Foundation of Serbia, Yugoslavia, Sakowitz Computer Laboratory of Baylor College of Medicine, Texas. Reprint requests: James B. Young, MD, 6535 Fannin SM 491, Houston, TX 77030, or Miodrag C. Ostojic, MD, PhD, Kardioloska Klinika, Dr. Subotica 13, 11000 Belgrade, Yugoslavia. *See “Appendix” at the end of this article.
PY - 1989/3
Y1 - 1989/3
N2 - A reasonably accurate, simple, inexpensive, noninvasive method of determining ejection fraction (EF) is necessary to evaluate left ventricular function in epidemiologic studies and individual patients. Using the concepts of left ventricular mass reflected by precordial R wave summation (M̂) and left ventricular volume (V̂) estimated by chest roentgenography in 114 consecutive patients with myocardial disease undergoing left ventriculography, EF was predicted with the formula: EF = 63.74 - (2.16 · V̂) + (0.34 · M̂); R2 = 0.69; standard error of the estimate (SEE) = 11. Because angiographic inferior wall motion (IWM) abnormalities significantly affected the results, but inferior Q waves were usually only present in patients having one infarct, a noninvasive technique to predict the presence of inferior wall motion abnormality (IM̂A) in patients having multiple infarcts was developed and based on the relationship of precordial R wave summation (M̂) and roentgenographic heart volumenometry (V̂). By combining V̂, M̂, and IM̂A (which predicted IWM) to determine EF, multiple linear regression analysis showed that EF = 67.30 - (1.56 · V̂) + (0.23 · M̂) - (14.18 IM̂A) (R2 = 0.77; SEE = 9). Prospective validation of the formula was then done in 139 consecutive individuals, with R2 = 0.49 and SEE = 9. This study demonstrates that routinely performed, simple, inexpensive clinical tests provide data that can be combined by multiple regression analysis to predict resting EF in patients with heart disease affecting the left ventricle. This unique method may allow inexpensive ventricular function screening in large population studies and in addition might provide an independent index of myocardial performance for clinical use, since it reflects the amount of contractile mass per unit of left ventricular volume. The formula can be adapted to a nomogram or hand-held calculator.
AB - A reasonably accurate, simple, inexpensive, noninvasive method of determining ejection fraction (EF) is necessary to evaluate left ventricular function in epidemiologic studies and individual patients. Using the concepts of left ventricular mass reflected by precordial R wave summation (M̂) and left ventricular volume (V̂) estimated by chest roentgenography in 114 consecutive patients with myocardial disease undergoing left ventriculography, EF was predicted with the formula: EF = 63.74 - (2.16 · V̂) + (0.34 · M̂); R2 = 0.69; standard error of the estimate (SEE) = 11. Because angiographic inferior wall motion (IWM) abnormalities significantly affected the results, but inferior Q waves were usually only present in patients having one infarct, a noninvasive technique to predict the presence of inferior wall motion abnormality (IM̂A) in patients having multiple infarcts was developed and based on the relationship of precordial R wave summation (M̂) and roentgenographic heart volumenometry (V̂). By combining V̂, M̂, and IM̂A (which predicted IWM) to determine EF, multiple linear regression analysis showed that EF = 67.30 - (1.56 · V̂) + (0.23 · M̂) - (14.18 IM̂A) (R2 = 0.77; SEE = 9). Prospective validation of the formula was then done in 139 consecutive individuals, with R2 = 0.49 and SEE = 9. This study demonstrates that routinely performed, simple, inexpensive clinical tests provide data that can be combined by multiple regression analysis to predict resting EF in patients with heart disease affecting the left ventricle. This unique method may allow inexpensive ventricular function screening in large population studies and in addition might provide an independent index of myocardial performance for clinical use, since it reflects the amount of contractile mass per unit of left ventricular volume. The formula can be adapted to a nomogram or hand-held calculator.
UR - http://www.scopus.com/inward/record.url?scp=0024589766&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0024589766&partnerID=8YFLogxK
U2 - 10.1016/0002-8703(89)90733-3
DO - 10.1016/0002-8703(89)90733-3
M3 - Article
C2 - 2919538
AN - SCOPUS:0024589766
SN - 0002-8703
VL - 117
SP - 590
EP - 598
JO - American Heart Journal
JF - American Heart Journal
IS - 3
ER -