Programme Hospitalier Recherche Clinique, Institut Pasteur, Inserm, French Public Health Agency.

Measurement of cardiac output using thermodilution technique in mechanically ventilated patients is associated with significant morbidity. The goal of the present study was to assess the validity of cardiac output measurement using transesophageal Doppler in critically ill patients. Forty-six patients from three different intensive care units underwent 136 paired cardiac output measurements using thermodilution (COTH) and transesophageal Doppler (COTED). In addition, simultaneous suprasternal Doppler and indirect calorimetry (Fick principle) were used to measure cardiac output in 26 patients from one center. A good correlation was found between COTH and COTED (r = 0.95), with a small systematic underestimation (bias = 0.24 L/min) using transesophageal Doppler. The limits of agreement between COTH and COTED were +2 L/min and -1.5 L/min. Variations in cardiac output between two consecutive measures using either transesophageal Doppler or thermodilution techniques were similar in direction and magnitude (bias = 0 L/min; limits of agreement = +/-1.7 L/min). Suprasternal Doppler and indirect calorimetry yielded similar correlations and agreements in the subset of patients in whom they were used. These results confirm that transesophageal Doppler can provide a noninvasive, clinically useful estimate of cardiac output and detect hemodynamic changes in mechanically ventilated, critically ill patients.

, surgical anatomic correction was performed in 86 newborn infants, 2-23 days old (6.8 +±3.6 days, mean+± SD) with simple transposition of the great arteries. In all patients, the pulmonary artery was reconstructed by end-to-end anastomosis according to the Lecompte maneuver, including eight patients with side-by-side position of the great arteries. Three different approaches were used. In the first 10 patients (group 1, six survivors), two separate patches of preserved tanned pericardium were used to reconstruct the pulmonary artery, whereas in the next 15 patients (group 2, 13 survivors), a single patch of the same material was used, and in the last 61 patients (group 3, 56

Left ventricular (LV) subendocardial segmental behavior was analyzed during the whole cardiac cycle for different loading and inotropic conditions in six conscious dogs that were instrumented with ultrasonic crystals in the basal (B) and apical (A) LV regions, a LV micromanometer, and an aortic cuff occluder. There were large variations of A and B segmental behavior during isovolumic contraction and relaxation. In contrast, a linear relationship between A and B was observed during ejection but segmental shortening was larger during control in A than in B, whether it was expressed as a percentage of systolic shortening (34.6 +/- 1.1 and 25.0 +/- 1.2%, respectively; P less than 0.005), or whether segments were normalized for passive resting length. This linear relationship during ejection with a slope of 1.49 was not significantly modified by alterations of loading conditions or inotropic state. The larger A than B shortening, independent of the normalization procedure, is attributed to the regional stress distribution in the LV. The absence of regional and cycle invariance particularly during isovolumic phases prevents the inference of ventricular volume from a limited number of dimensions.