Diastolic blood pressure and end-tidal carbon dioxide during adult ICU cardiopulmonary resuscitation: association with return of spontaneous circulation.

Abstract

To evaluate the association of intra-arrest diastolic blood pressure and end-tidal carbon dioxide with return of spontaneous circulation in adult ICU cardiac arrest. This prospective observational study was conducted in an adult medical-surgical ICU over 9 months. Adult patients with an indwelling arterial catheter and mainstream capnography at the time of cardiac arrest were included. Diastolic blood pressure (DBP) and end-tidal carbon dioxide (EtCO₂) were recorded at the end of each 2-min CPR cycle. The primary analysis used mean DBP and mean EtCO₂ across the CPR episode. Last recorded pre-outcome values were analyzed separately as secondary peri-outcome measures. The primary outcome was ROSC. Associations were assessed using logistic regression, and discrimination was evaluated using receiver operating characteristic curve analysis. A total of 68 cardiac arrest events from 63 patients were analyzed; ROSC was achieved in 29 events (42.6%). Mean DBP across the CPR episode was higher in events achieving ROSC than in those without ROSC (39 ± 18 vs 24 ± 10 mmHg; mean difference 15 mmHg, 95% CI 8-23; p < 0.001). Mean EtCO₂ was also higher in events achieving ROSC (19 ± 5 vs 15 ± 6 mmHg; mean difference 4 mmHg, 95% CI 1-6; p = 0.010). For mean values, DBP showed numerically higher discrimination than EtCO₂, but the difference was not statistically significant (AUC 0.78 vs 0.69; DeLong p = 0.30). In secondary peri-outcome analysis, last recorded pre-outcome values had higher AUCs, particularly for DBP (0.96 vs 0.85; DeLong p = 0.04), although this time point was identifiable only after ROSC or termination of resuscitation. In adult ICU cardiac arrest events with simultaneous arterial pressure and capnographic monitoring, both DBP and EtCO₂ were associated with ROSC. DBP showed numerically higher discrimination than EtCO₂, although this was not statistically significant for mean values across the CPR episode. Last recorded pre-outcome values should be interpreted only as peri-outcome associations and not as real-time predictors or treatment targets. DBP and EtCO₂ should be viewed as complementary physiologic markers during CPR.