Abstract Body: Abstract
Background:
Survival after in-hospital cardiac arrest (IHCA) remains at ~20%. The American Heart Association (AHA) recommends physiological feedback-guided strategies to potentially improve outcomes. While end-tidal CO2 (ETCO2) reflects CPR quality and regional cerebral oxygen saturation (rSO2) is associated with improved survival and neurological outcomes, the optimal thresholds for the use of these physiological markers during CPR remain unknown. In a recent multi-site study, we applied a machine learning (CART) model on 365 IHCA and determined that ETCO2 ≥35 mmHg and rSO2 ≥60% are optimal targets for ROSC and survival. We now sought to conduct a pilot randomized trial of CPR guided by these physiologic targets versus standard CPR in preparation for a future clinical trial.
Methods:
In this single-center, single-blinded pilot RCT (Feb 2022–May 2024), using waiver of informed consent, 93 adults with IHCA aged 18–85 were randomized 1:1 to receive standard CPR (2020 AHA guidelines with mechanical compressions) or physiologic-feedback CPR (standard CPR with mechanical compressions plus other therapies (a) high dose epinephrine at 3, 5 mg, b) 20 IU of vasopressin, c) hydrocortisone 200mg) to target ETCO2 ≥35 mmHg and/or rSO2 ≥60%. Resuscitation (Rhythm, duration, vasopressor dose) together with post-resuscitation variables (TTM, PaO2, PCO2, MAP) that are associated with survival were collected. The primary outcome was sustained ROSC; secondary outcomes were survival with neurological recovery. .
Results:
Overall, 47 received standard CPR and 46 physiologic-feedback CPR. Sustained ROSC occurred in 32/47 (68.1%) vs. 27/46 (58.7%) (P=0.34); survival to discharge was 19.1% vs. 17.4% (P=0.79). High-dose epinephrine (≥3 mg) was more frequent in the physiologic group (27% vs. 7%; P=0.01). At 24 hours, PaO2 was higher in the physiologic group (137 vs. 98 mmHg; P=0.04). Post-resuscitation analysis showed targeted temperature management was applied in 54% vs. 46% (P=0.60), but standard CPR reached lower hypothermic temperatures (32.6°C vs. 34.8°C; P=0.034). MAP was similar (70 vs. 76 mmHg; P=0.4). No differences in adverse events were observed.
Conclusion:
This pilot trial supports the feasibility of integrating rSO2/ETCO2 monitoring into IHCA resuscitation. Larger trials are needed to assess the efficacy of these parameters.