To investigate the feasibility of using a commercially available cerebral oximeter during in-hospital cardiac arrest, and determine whether this parameter predicts return of spontaneous circulation (ROSC).
Cerebral oximetry was incorporated in cardiac arrest management in 19 in-hospital cardiac arrest cases, five of whom had ROSC. The primary outcome measure was the relationship between rSO<sub>2sub> and ROSC.
The use of cerebral oximetry was found to be feasible during in hospital cardiac arrest and did not interfere with management. Patients with ROSC had a significantly higher overall mean ¡À SE rSO<sub>2sub> (35 ¡À 5 vs. 18 ¡À 0.4, p < 0.001). The difference in mean rSO<sub>2sub> between survivors and non-survivors was most pronounced in the final 5 min of cardiac arrest (48 ¡À 1 vs. 15 ¡À 0.2, p < 0.0001) and appeared to herald imminent ROSC. Although spending a significantly higher portion of time with an rSO<sub>2sub> > 40 % was found in survivors (p < 0.0001), patients with ROSC had an rSO<sub>2sub> above 30 % for >50 % of the duration of cardiac arrest, whereas non-survivors had an rSO<sub>2sub> that was below 30 % > 50 % of their cardiac arrest. Patients with ROSC also had a significantly higher change in rSO<sub>2sub> from baseline compared to non-survivors (310 % ¡À 60 % vs. 150 % ¡À 27 % , p < 0.05).
Cerebral oximetry may have a role in predicting ROSC and the optimization of cerebral perfusion during cardiac arrest.