Central Venous to Arterial Co 2 Difference After Cardiac Surgery in Infants and Neonates*

Objectives: Venous to arterial C o 2 difference correlates with cardiac output in critically ill adults, but its utility in pediatric patients is unclear. We sought to correlate venous to arterial C o 2 difference with other cardiac output surrogates (arteriovenous oxygen saturation difference, central venous oxygen saturation, and lactate) and investigate its capacity to predict poor outcomes associated with low cardiac output (low cardiac output syndrome) in infants after cardiac surgery with cardiopulmonary bypass. Design: Retrospective chart review. Poor outcome was defined as any inotrope score greater than 15; death, cardiac arrest, extracorporeal membrane oxygenation; and unplanned surgical reintervention. Setting: Pediatric cardiovascular ICU. Patients: One hundred thirty-nine infants less than 90 days who underwent cardiopulmonary bypass, from October 2012 to May 2015. Intervention: None. Measurements and Main Results: Two hundred ninety-six arterial and venous blood gas pairs from admission ( n = 139), 6 ( n = 62), 12 ( n = 73), and 24 hours ( n = 22) were included in analysis. For all pairs, venous to arterial C o 2 difference was moderately correlated with arteriovenous oxygen saturation difference ( R 2 = 0.53; p < 0.01) and central venous oxygen saturation ( R 2 = –0.43; p < 0.01), but not lactate. At admission, venous to arterial C o 2 difference was also moderately correlated with central venous oxygen saturation ( R 2 = –0.40; p < 0.01) and arteriovenous oxygen saturation difference ( R 2 = 0.55; p < 0.01), but not lactate. Thirty-four of 139 neonates (24.5%) had poor outcome. Median admission venous to arterial C o 2 difference was 5.9 mm Hg (3.8–9.2 mm Hg). Patients with poor outcome had median admission venous to arterial C o 2 difference 8.3 (5.6–14.9) versus 5.4 mm Hg (3.0–8.4 mm Hg) in those without poor outcome. Venous to arterial C o 2 difference (area under the curve = 0.69; p < 0.01), serum lactate (area under the curve = 0.64; p = 0.02), and central venous oxygen saturation (area under the curve = 0.74; p < 0.01) were predictive of poor outcome. After controlling for covariates, admission venous to arterial C o 2 difference remained significantly associated with poor outcome (odds ratio, 1.3; 95% CI, 1.1–1.45), including independent association with mortality (odds ratio, 1.2; 95% CI, 1.07–1.31). Conclusions: Venous to arterial C o 2 difference is correlated with important surrogates of cardiac output, and is associated with poor outcome and mortality related to low cardiac output syndrome after cardiac surgery in infants. Prospective validation of these findings, including confirmation that venous to arterial C o 2 difference can identify low cardiac output syndrome in real time, is warranted.