While the use of rSO2, to inform on the quality of cardiopulmonary resuscitation (CPR) or to predict ROSC [
13], supports the feasibility of NIRS in cardiac arrest and its potential to guide acute resuscitation, the main focus of this text is on post-resuscitation care. Multiple studies have been conducted, giving variable results regarding the possible differences in rSO2 values between patients with a good versus a poor functional outcome (Table
1). It appears biologically plausible that rSO2 values indicating that brain oxygen homeostasis has been maintained would be associated with survival and favorable neurological outcome. In an observational study of 28 cardiac arrest patients, rSO2 was lower following the initiation of hypothermia in non-survivors (n = 10) compared to survivors (n = 28) censored at hospital discharge [
14]. Similar results were reported in 60 cardiac arrest patients, in which rSO2 during the first 40 h of intensive care unit (ICU) monitoring, including hypothermia and rewarming, was higher in patients with good outcomes (cerebral performance category [CPC] 1–2) compared to poor outcomes (CPC 3–5), both at ICU discharge and at 6 months, albeit with a large overlap in rSO2 values [
15]. A larger prospective study (n = 107) of rSO2 during the first 48 h of ICU admission, including hypothermia and rewarming, and its association with outcome at 3 months reported statistically higher rSO2 in patients with good outcomes (CPC 1–2) compared to those with poor outcomes (CPC 3–5). Yet the study authors noted that the numerical differences were small and not conducive to a clinically useful discrimination of outcomes [
16]. Based on data from the Japanese J-POP registry, an rSO2 > 40%—measured immediately upon arrival in the emergency department following cardiac arrest—was associated with favorable neurologic outcome at day 90 [
17,
18]. A review of 22 observational studies, encompassing 2436 patients, corroborated the associations between increasing and higher rSO2 in the post-cardiac arrest period and favorable outcomes [
19]. Meanwhile, several studies since the review—including 258 out-of-hospital cardiac arrest patients—have failed to demonstrate either a correlation or sufficient discriminative power for rSO2 and good versus poor outcomes [
20‐
24], or have found it only in a specific range of initial rSO2 (between 41 and 60%) during TTM [
25]. A recent review concluded that the clinical utility of monitoring rSO2 to prognosticate a favorable neurological outcome remains unclear [
26]. Further clinical research is needed to establish the role of static versus dynamic rSO2 values; the cut-off values for correlations to patient-centered outcomes, including during different interventions for hypoxic ischemic brain injury, notably TTM; and the minimal duration of monitoring. It is also important to address the variability in reported rSO2 signals across different NIRS monitors [
27] and overall cerebral tissue oxygenation [
28].
Table 1
A selection of studies evaluating associations between near infrared spectroscopy (NIRS) measured and derived variables with outcome, in intensive care unit (ICU)-treated out-of-hospital cardiac arrest
| 2013 | Observational study | 28 | CA patients treated with TTM | Functional outcome by CPC at hospital discharge | Decrease in rSO2 during induction of TTM. Lower rSO2 levels in patients with poor outcome |
| 2014 | Observational study | 60 | OHCA and IHCA | Functional outcome at discharge by CPC | Higher NIRS values in patients with good outcome. An rSO2 below 50% appeared associated with poor outcome |
| 2015 | Observational study | 51 | All types of CA | Functional outcome at 180 days by CPC | Disturbed autoregulation more common in patients with chronic hypertension. Time below an autoregulation-derived optimal MAP was negatively associated with outcome |
| 2015 | Observational study | 23 | OHCA | Functional outcome at 90 days by CPC | No difference in rSO2 in patients, by outcome. Suggestion of disturbed autoregulation in poor outcome patients |
| 2016 | Observational study | 43 | OHCA treated with TTM | Functional outcome by CPC on hospital discharge | Mean rSO2 was not different, when indexed by outcome, but the lowest measured was lower in poor outcome patients |
| 2016 | Observational study | 107 | OHCA | Functional outcome at 180 days by CPC | Slightly higher rSO2 in patients with good outcome. No reliable threshold value was identified |
| 2018 | Observational study | 25 | OHCA patients | Functional outcome by CPC on hospital discharge | No difference in rSO2, in patients with good and poor outcome |
| 2019 | Post-hoc analysis of interventional data | 120 | VF arrests with a cardiac cause | Six-month functional outcome by CPC and brain injury assessed with NSE | No association between the mean, median, lowest or highest NIRS value during the first 36 h of ICU care with outcome or the level of NSE at 48 h |