The Correlation Study And Clinical Evaluation Between Pratial CO₂ Rebreathing And Thermodilution For Cardiac Output Monitoring

Objective: To study the correlation between the methods of partial CO2 rebreathing(RB) technique and thermodilution(TD) technique for cardiac output; to investigate the effect of ventilatory settings on accuracy of CORB,; to evaluate the clinical application of RB technique.Methods: Six anesthetized ventilated dog were enrolled in this study. Each animal was inserted with Swan-Ganz catheter, and instrumented a partial CO2 rebreathing ventilator circuit. Simultaneous measurement of CORB and COTD were performed in following situations: (1)cardiac output was varied by manipulating inotropic drugs(adrenine, dobutamine, dopamine, verapamil); (2)under six different ventilatory settings: ①volume controlled ventilation with tidal volume of 12ml/kg(VCV, VT 12ml/kg);②volume controlled ventilation with tidal volume of 6ml/kg(VCV, VT 6ml/kg); ③pressure controlled ventilation with tidal volume of 12ml/kg(PCV, VT 12ml/kg); ④pressure controlled ventilation with tidal volume of 6ml/kg(PCV, VT 6ml/kg);⑤volume controlled ventilation with high positive end-expiratory pressure(high PEEP). The correlation between the two methods was evaluated with Paired t-test, linear regression and Bland-Altman analysis. Results: (1) Paired t-test analysis revealed no significant difference between CORB and COTD measurement at base line or after injection of inotropic drugs;(2) Regression analysis revealed an R value of 0.89 at base line, and the regression equation was as follows: CORB=0.78COTD+0.25; Bland-Altman analysis resulted in the limit of agreement was -1.67~1.78 L/min; (3) There was a good agreement between CO measurement using RB technique and TD technique after injection of inotropic drugs, the correlation coefficient was 0.81, with the regression equation CORB=0.91COTD +0.61, while the limit of agreement was -2.13~2.03 L/min; (4) A typical tracking curve illustrated the close relationship between the variations in CO as measured by RB technique and TD techniques;(5)Regardless of ventilation miode, when Vt was set at 6ml/kg, PetCO2,PaCO2 was higher than that of Vt was set at 12ml/kg; (6) When Vt was set at 12ml/kg, CORB correlated well with that COTD (R=0.70; bias: 0.29L/min; limits of agreement: -1.76~2.34L/min). However, when Vt was set at 6ml/kg, the CORB technique underestimated cardiac output compared with COTD technique (R=0.64; bias: -1.65L/min: limits of agreement: -3.78~1.58L/min); (7) Linear multiple regression analysis revealed that the ventilatory setting most affecting the discrepancy between CORB and COTD was minute ventilation(VE). Conclusion: (1) There was a good correlation between CORB and COTD measurement no matter before or after injection inotropic drugs; (2) Ventilatory settings may affect the accuracy of CORB, linear multiple regression analysis revealed that the ventilatory settings most affecting the discrepancy between CORB and COTD was VE; (3) RB technique is a simple, noninvasive, economical and reliable method for measurement of CO.