Abstract 282: Cerebral Oximetry to Identify Return of Spontaneous Circulation in Patients Treated With Manual or Mechanical Chest Compressions

Abstract

Background: Near infrared spectroscopy can be used to monitor cerebral oxygen saturation in out-of-hospital cardiac arrest (OHCA). Brain oxygen saturation increase has been associated to cardiopulmonary resuscitation therapy (CPR), and higher saturation values are observed during return of spontaneous circulation (ROSC) and in survivors. In this study we evaluate cerebral oxygen saturation to identify ROSC, and compare its value for patients treated with manual or mechanical CPR. Materials and Methods: A total of 87 OHCA patients treated by the Emergency medical system (EMS) of the Basque Country were included, 32 treated with manual CPR and 55 with the LUCAS-3 device (Stryker). Brain oxygen saturation was measured in the left and right lobes using the NIRO NX-200 (Hamamatsu) in the manual group, and the SenSmart X-100 (NONIN) in the LUCAS-3 group. For each patient the mean value of the left and right lobe saturation in the first minute after the end of chest compressions was computed. Saturation values for the ROSC/no-ROSC groups were compared using the Wilcoxon rank sum test, p<0.01 was considered significant. The area under the curve (AUC) was computed for each CPR group to measure how well brain saturation discriminates ROSC from no-ROSC. Results: In the manual CPR group median (IQR) saturation values were 39.2% (33.6-49.7) for the no-ROSC (n=23) and 64.3% (55.2-67.0) for ROSC (n=9) (p<0.01), and the AUC was 0.92. In the LUCAS-3 CPR group saturation values were 41.5% (29.8-50.5) for the no-ROSC (n=37) and 67.0% (61.8-75.0) for ROSC (n=18) (p<0.01), and the AUC was 0.91. No significant differences in saturation were observed between manual and mechanical CPR for both ROSC (p=0.46) and no-ROSC patients (p=0.28). Conclusions: Differences in cerebral oximetry were significant between ROSC and no-ROSC patients. No differences were observed in saturation between the manual and mechanical CPR groups, despite using different oximeter models. Cerebral oximetry could be used to predict outcome regardless of the type of CPR.