Assessment of changes in left ventricular systolic function with oesophageal Doppler.

We tested the ability of mean acceleration (Acc) and peak velocity (V peak) of the aortic velocity signal measured by oesophageal Doppler to reflect left ventricular (LV) systolic performance. We included critically ill patients in whom a fluid challenge (n=25) or the introduction of dobutamine, 5 µg kg(-1) min(-1) (n=25), was planned by the attending physician. Before and after therapeutic interventions, we measured Acc and V peak (CardioQ device) and LV ejection fraction (LVEF) using echocardiography. For all pairs of measurements, the absolute values of Acc and V peak correlated with LVEF (r=0.36 and 0.57, respectively). The correlation was significantly higher for V peak than for Acc. Volume expansion did not significantly change LVEF and Acc, but significantly increased V peak by 7 (8)%. Dobutamine increased LVEF by 30 (15)%, Acc by 33 (25)%, and V peak by 20 (10)%. Considering the pooled effects of volume expansion and dobutamine, changes in Acc and V peak and those of LVEF were correlated (r=0.53 and 0.67, respectively). When excluding changes <18% (i.e. the least significant change for LVEF), the concordance rate was 96% for Acc and 100% for V peak. V peak and, to a lesser extent, Acc measured by oesophageal Doppler behaved as markers of LV systolic performance as they were almost insensitive to fluid administration and changed to a much larger extent with dobutamine. These indices could be used to estimate LV systolic performance and to assess the effects of inotropic therapy.