[Evaluation of heart function in chronic coronary disease non-invasive methods]

In 58 patients with a clinical diagnosis of ischemic heart disease the value of Ecg, phonocardiogram, systolic time intervals and derived indices, apex-cardiogram (Acg), and kinetocardiogram (Kcg) in assessment of cardiac function was examined. In 52 patients coronary artery lesions were demonstrated on selective coronary angiograms, in 40 patients dyssynergy was seen on ventriculograms, and in 14 patients a left vencribular aneurysm was diagnosed. Ecg criteria indicated a normal ejection fraction with a sensitivity of 80% and a specificity of 84%, an ejection fraction less than or equal to 30% with a sensitivity of 72% and a specificity of 77%, a localized dyssynergy with a sensitivity and a specificity of 83% respectively, and a ventricular aneurysm with a sensitivity of 86% and a specifity of 92%. In respect to the latter diagnosis Q-wave criteria proved superior to ST- and T-alterations. An atrial gallop bore nor relationship to the parameters of cardiac function. A protodiastolic gallop proved to be a relative specific but insensitive sign of poor ventricular function. In the majority of cases the corrected injection time was diminished in respect to normal values; a moderate correlation was to be calculated between its shortening and the ejection fraction (r = -0.50). The quotient ejection time divided by preejection period allowed the estimation of an ejection fraction below 30% in 64% of cases without false positive results. There was a close correlation between this quotient and the ejection in the pathologic range below 50% (r = -0.69), whereas no correlation was to be found in the normal range of EF. An Acg could be obtained only in 72% of the patients. A moderate correlation was established to the left ventricular enddiastolic pressure (r = 0.42). The diagnostic score was poor in respect to ventricular aneurysms, which could be recognized in 43%, but were falsely assumed in 40%. Kcg records yielded relatively specific but insensitive indications of a normal fraction (specificity 84%, sensitivity 32%) and of an elevated left ventricular enddiastolic pressure (sensitivity 54%, specificity 80%). Large non-contractile ventricular segments and ventricular aneurysms were recognized with a sensitivity of 72% each and a specificity of 62% and 52% respectively. Multiple regression analysis between various non-invasive methods and hemodynamic data resulted in a closer correlation, but nontheless no exact prediction of the "true" values was possible in the single case. It is concluded that the accuracy of the parameters of the non-invasive methods under study in assessing left ventricular performance in coronary heart disease is worse than generally accepted giving reliable information only in patients with very poor and with excellent cardiac function.