Résumé
Left ventricle [LV] systolic function is a corner stone in diagnostic and therapeutic decision makings and is a potent predictor of patient's outcome. A lot of investigational techniques help to assess the systolic performance of LV but in practice echocardiography and LV angiography are still the most used methods. Because, in the last decade echocardiographic techniques have tremendously evolved, accuracy and limitations of LV systolic function assessment by echo has to be revisited. 100 successive patients scheduled for LV angiography, got estimations of LV stroke volume [SV] and ejection fraction [LVEF] echographically by Simpsom's methods and by pulsed Doppler at the level of aortic annulus in the second harmonic mode and invasively by a 30° RAO LV angiography. Echocardiography has been performed by two operators prior to the angio study. Statistics were done using linear regression methods with determination of the correlation coefficient [r] and the standard error estimation [SEE] when the variables are normally distributed or the spearmn's coefficient [rs] when not. Echocardiographic assessment was technically feasible in all patients by both operators. Echocardiographic methods had excellent correlation to angiography and correlation coefficients of biplane Simpson and pulsed Doppler at the level of the aortic annulus were respectively r = 0.9 and r = 0.85. These coefficients were r = 0.85 and r = 0.69 in the subset of patients with LV wall motion dyssynergie. The intra and inter observer variabilities were below 7% for all methods. Echocardiographic LV systolic function assessment is nowadays technically possible in almost all patients with an excellent correlation to LV angiography and a good inter observer reproducibility whether there is or not LV remodeling or wail motion abnormities