Associations of intraoperative end–tidal CO<inf>2</inf> levels with postoperative outcome–secondary analysis of a worldwide observational study

Abstract

Background: Patients receiving intraoperative ventilation during general anesthesia often have low end–tidal CO2 (etCO2). We examined the association of intraoperative etCO2 levels with the occurrence of postoperative pulmonary complications (PPCs) in a conveniently–sized international, prospective study named ‘Local ASsessment of Ventilatory management during General Anesthesia for Surgery’ (LAS VEGAS). Methods: Patients at high risk of PPCs were categorized as ‘low etCO2’ or ‘normal to high etCO2’ patients, using a cut–off of 35 mmHg. The primary endpoint was a composite of previously defined PPCs; the individual PPCs served as secondary endpoints. The need for unplanned oxygen was defined as mild PPCs and severe PPCs included pneumonia, respiratory failure, acute respiratory distress syndrome, barotrauma, and new invasive ventilation. We performed propensity score matching and LOESS regression to evaluate the relationship between the lowest etCO2 and PPCs. Results: The analysis included 1843 (74 %) ‘low etCO2’ patients and 648 (26 %) ‘normal to high etCO2’ patients. There was no difference in the occurrence of PPCs between ‘low etCO2’ and ‘normal to high etCO2’ patients (20 % vs. 19 %; RR 1.00 [95 %–confidence interval 0.94 to 1.06]; P = 0.84). The proportion of severe PPCs among total occurring PPCs, were higher in ‘low etCO2’ patients compared to ‘normal to high etCO2’ patients (35 % vs. 18 %; RR 1.16 [1.08 to 1.25]; P < 0.001). Propensity score matching did not change these findings. LOESS plot showed an inverse relationship of intraoperative etCO2 levels with the occurrence of PPCs. Conclusions: In this cohort of patients at high risk of PPCs, the overall occurrence of PPCs was not different between ‘low etCO2’ patients and ‘normal to high etCO2’ patients, but severe PPCs occurred more often in ‘low etCO2’, with an inverse dose–dependent relationship between intraoperative etCO2 levels and PPCs. Funding: This analysis was performed without additional funding. LAS VEGAS was partially funded and endorsed by the European Society of Anesthesiology and Intensive Care (ESAIC) and the Amsterdam University Medical Centers, location ‘AMC’. Registration: LAS VEGAS was registered at Clinicaltrials.gov (NCT01601223), first posted on May 17, 2012.