The correlation between cardiac output (CO) evaluated by echocardiography and CO measured by thermodilution (COth) varies according to different studies. A new transthoracic echocardiography (TTE) tool allows automatic calculation of the subaortic velocity time index (VTIauto) and CO (COauto). The main objective was to evaluate the correlation between COth and COauto in an anesthetized, ventilated piglet hemorrhagic shock (HS) model. The secondary objectives were to evaluate the correlation between COth and CO evaluated by manual measurements of VTI, and the preload-dependency of VTIvaresp.
Eighteen piglets were bled until mean arterial pressure reached 40 mm Hg. Controlled hemorrhage was maintained for 30 min before a resuscitation phase. CO was measured by Pulse index Contour Cardiac Output thermodilution methods. At each time of the experiment, three VTI values were measured (min, med, max) and the average value was calculated. COs were calculated by TTE (COmax, COmed, COmin, COave).
For the 204 measures attempted, the success rate was 197 (97%) manually and 122 (60%) automatically (P < 0.01). The correlation coefficients (r) between COth and, respectively, COauto, COave, COmax, COmed, and COmin were: 0.83 (95% CI [0.76; 0.88]; P < 0.01), 0.54 (95% CI [0.43; 0.63]; P < 0.01), 0.43 (95% CI [0.31; 0.54]; P < 0.01), 0.58 (95% CI [0.48; 0.67]; P < 0.01), and 0.52 (95% CI [0.41; 0.62]; P < 0.01).
In an experimental model of HS, a new ultrasound tool, COauto, seems better correlated with COth than manual echocardiographic measurements.