Background

To evaluate the association between intraoperative end-tidal carbon dioxide (ETCO₂) levels and clinical outcomes in patients undergoing cardiac surgery.

Methods

This retrospective analysis included patients who underwent cardiac surgery at Seoul National University Hospital in South Korea from 2011 to 2020. The primary exposure variable was the mean ETCO₂ level, and the primary outcomes were in-hospital mortality and 30-day postoperative mortality. Secondary outcomes included the duration of postoperative mechanical ventilation, intensive care unit (ICU) length of stay (LOS), and total hospital LOS. Multivariable Cox proportional hazards models were employed to assess the association between ETCO₂ and the primary outcomes. These models were adjusted for demographic characteristics, preoperative variables, and intraoperative factors using a stepwise approach. Linear regression analyses were performed for the secondary outcomes. Restricted cubic splines (RCS) were utilized to explore potential non-linear associations. The optimal ETCO₂ cutoff value was identified using receiver operating characteristic (ROC) analysis and the maximal Youden index. Survival rates between groups were compared using the Kaplan–Meier method.

Results

A total of 2,161 patients were included in the study, of these, 2,060 survived and 101 died (4.7%). Adjusted Cox regression analysis demonstrated a negative association between ETCO₂ and in-hospital mortality (HR 0.89, 95% CI 0.81 to 0.97, P = 0.012), with RCS indicating a linear relationship. No significant association was observed between ETCO₂ and 30-day postoperative mortality (HR 0.89, 95% CI 0.78 to 1.02, P = 0.086). Secondary outcome analyses demonstrated a negative correlation between ETCO₂ and ICU LOS (β -0.27, 95% CI -0.38 to -0.17, P < 0.001) and mechanical ventilation duration (β -0.23, 95% CI -0.32 to -0.15, P < 0.001), but no significant association with total hospital LOS (β 0.56, 95% CI -0.27 to 1.38, P = 0.187). ROC curve analysis demonstrated that ETCO₂ predicted in-hospital mortality with an AUROC of 0.66.

Conclusions

Lower intraoperative ETCO₂ levels were independently associated with an increased risk of in-hospital mortality, as well as prolonged ICU stay and duration of mechanical ventilation. Given the single-center, retrospective design of this study, these findings are exploratory and require validation in prospective, multicenter cohorts.