Carbon dioxide combining power as a predictor of unplanned ICU transfer in hospital wards: a retrospective cohort...

Unplanned intensive care unit (ICU) transfers are associated with high mortality. Conventional early warning scores like modified Early Warning Score (MEWS) show limited predictive ability. Carbon dioxide combining power (CO2CP), a routine acid-base parameter, may enable earlier detection of clinical deterioration.

To evaluate CO2CP as a predictor of mortality in unplanned ICU transfers and assess its additive value to traditional scoring systems.

This single-center retrospective study analyzed 101 adults with unplanned ICU transfers (Feb 2024-Jun 2025). CO2CP levels within 8 h pre-transfer, MEWS, and APACHE II scores were collected. Primary outcome was ICU mortality. Analyses included ROC curves, survival analysis, multivariable logistic regression, decision curve analysis, and correlation assessment.

The area under the ROC curve (AUC) for CO2CP was 0.722 (95% CI: 0.615-0.829), higher than that of MEWS (AUC = 0.528, p = 0.040). Bootstrap validation yielded an AUC of 0.723 (95% CI: 0.581-0.849), indicating relative stability. Survival analysis showed poorer outcomes in patients with lower CO2CP (log-rank P = 0.003). In multivariable analysis, CO2CP remained independently associated with mortality (OR ≈ 0.87 per 1 mmol/L increase, p < 0.001), although the effect size was modest. The combined CO2CP-MEWS model demonstrated good discriminatory ability, although its clinical utility requires further confirmation.

Carbon dioxide combining power may serve as an independent predictor of mortality risk in unplanned ICU transfers. Its integration with MEWS may enhance early risk stratification in general wards, though these findings require validation in larger prospective studies.