Quality controlled manual chest compressions and cerebral oxygenation during in-hospital cardiac arrest

The quality of cardiopulmonary resuscitation (CPR) is associated with the rate of return of spontaneous circulation (ROSC) during human cardiac arrest. Current advances in defibrillator technology enable measurement of CPR quality during resuscitation, but it is not known whether this is directly reflected in cerebral oxygenation. In this descriptive study we aimed to evaluate whether the quality of feedback-monitored CPR during in-hospital cardiac arrest is reflected in near infrared frontal cerebral spectroscopy (NIRS).

Nine patients suffering an in-hospital cardiac arrest in a university hospital were included. All patients underwent quality-controlled CPR performed by a dedicated medical emergency team using a Philips HeartStart MRx defibrillator (Philips, Eindhoven, Netherlands) with a CPR quality (Q-CPR, Laerdal Medical, Stavanger, Norway) analysis feature. Simultaneously, bilateral frontal cerebral oximetry was measured using INVOS 5100c (Somanetics, Troy, MI, USA) NIRS.

During quality controlled resuscitation, regional cerebral oxygenation (rSO(2)) as measured with NIRS was low but it improved during CPR (p=0.043) and 8 min after ROSC (p=0.022). After the onset of NIRS recording, there were four episodes exceeding 30s, during which the quality of CPR was substandard. When CPR technique was corrected and maintained for 2 min, a minor non-significant increase in rSO(2) was observed in two cases.

High quality CPR was not significantly reflected in cerebral oxygenation as quantified using NIRS. Even after ROSC and subsequent significant increase in cerebral oxygenation, rSO(2) readings were below previously suggested threshold of cerebral ischaemia. Improving CPR technique after an episode of low quality CPR did not significantly increase rSO(2).