Is Higher Cerebral Oxygen Saturation Really Predictive of Favorable Neurologic Outcome After In-Hospital Cardiac Arrest?

    loading  Checking for direct PDF access through Ovid


In a recent issue of Critical Care Medicine, we read with interest the article by Parnia et al (1) who investigated the utility of cerebral oximetry for predicting the return of spontaneous circulation (ROSC) and favorable neurologic outcome after in-hospital cardiac arrest (IHCA). The researchers demonstrated that increased regional cerebral oxygen saturation (rSO2) was predictive of ROSC and better cerebral performance category (CPC) after IHCA, concluding that rSO2 during cardiopulmonary resuscitation (CPR) was a useful noninvasive monitor for predicting survival and favorable neurologic outcomes. We appreciate the value of these research findings, which provide clinically useful information for the treatment of patients with IHCA.
However, we feel that several factors that may have potentially affected the results warrant further discussion. First, the statistical methodology seems to be inadequate. The authors reported that the increased rSO2 was predictive of better neurologic outcomes based on the results shown in Tables 4 and 10 (1). However, the CPC 3–5 group did not include patients with pure ROSC with CPC 3–5 but included patients with no ROSC. As shown in Figure 3 (1), there was no significant difference in rSO2 between patients with CPC 1–2 and those with CPC 5, whereas there was a significant difference in rSO2 between the CPC 5 and no ROSC groups. Therefore, the differences in several rSO2 variables between CPC 1–2 and CPC 3–5 in Table 4 were likely to have been overestimated. In addition, the authors described that rSO2 greater than or equal to 65% was predictive of ROSC. However, rSO2 greater than or equal to 45% was the best cut off value for predicting ROSC from the receiver operating characteristic curve analysis (Appendix 1: the arrow indicates the closest point to 100% sensitivity and 100% specificity).
Second, the baseline characteristics potentially affecting outcome of IHCA were not thoroughly evaluated. For instance, the presence of bystander CPR, witness, initial cardiac rhythm, presence of hypothermia, and underlying diseases could have affected the rate of ROSC and neurologic outcomes. Previous studies have demonstrated that bystander CPR is a predictor of increased likelihood of survival (2) and better neurologic outcome (3) in patients with out-of-hospital cardiac arrest. Terman et al (4) revealed that cardiac arrest with shockable cardiac rhythm was predictive of favorable CPC at hospital discharge compared with nonshockable rhythm. In the current study, initial cardiac rhythms on cardiac arrest were significantly different between patients with CPC 1–2 and those with CPC 3–5, which should be analyzed as covariates for predicting survival and neurologic outcomes.
In conclusion, the provision of additional data by the authors would be helpful for better understanding the significance of rSO2 in patients with IHCA.

Related Topics

    loading  Loading Related Articles