HIGH CENTRAL VENOUS-TO-ARTERIAL CO2 DIFFERENCE IS ASSOCIATED WITH POOR OUTCOMES IN PATIENTS AFTER CARDIAC SURGERY: A PROPENSITY SCORE ANALYSIS

Purpose: In contrast to arterial lactate, previous studies have proposed central venous-to-arterial CO2 difference (P (v-a)CO2) as a more useful guide for categorizing patients at risk of developing septic shock. It is worthwhile studying P (v-a)CO2 in determining whether it could serve as a useful predictor for poor postoperative outcomes in patients undergoing cardiac surgery. We investigated the ability of P(v-a)CO2 to predict poor outcomes of patients with postoperative cardiogenic shock. Methods: In total, 1,672 patients were enrolled in this study from January 1, 2014 to June 1, 2017. Of these patients, 143 exhibited complicated and poor outcomes. To address any bias, we derived a propensity score predicting the functions of P(v-a)CO2 on poor outcomes and matched 114 cases to 114 controls with a similar risk profile. In this cohort study, poor outcomes were defined as the occurrence of any adverse complications, including sudden death, cardiac arrest, extracorporeal membrane oxygenation, oliguria, and the administration of a large amount of vasoactive-inotropic drugs. Results: In propensity-matched patients, significant differences in P(v-a)CO2 (6.11 +/- 2.94 mm Hg vs. 11.21 +/- 5.22 mm Hg, P<0.001) were noted between the control group and poor outcome group. The area under the receiver operating characteristic curve of P(v-a)CO2 (AUC: 0.837, 95% CI: 0.782-0.892) for the detection of poor outcomes was significantly better compared to that of the central venous oxygen saturation and arterial lactate. Additionally, there was a negative correlation between cardiac index and P(v-a)CO2 (R=-0.68, P<0.001). Conclusion: We have shown a correlation between P(v-a)CO2 to cardiac output which may be used as an alternative metric to predict the poor outcomes of patients with postoperative cardiogenic shock.