Search

Hyperoxemia during cardiac arrest (CA) may increase chances of successful resuscitation. However, episodes of severe hyperoxemia after intensive care unit admission occurs frequently (up to 60%), and these have been associated with higher mortality in CA patients. The impact of severe hyperoxemia on neurological outcome is more unclear.We conducted a systematic review and meta-analysis on Pubmed and EMBASE to evaluate the effects of severe hyperoxemia according to arterial blood gas analysis on neurological outcome and mortality in patients resuscitated from CA and admitted to intensive care unit.Thirteen observational studies were included, eight of them reporting data on neurological outcome and ten on mortality. Most studies reported odds ratio adjusted for confounders. Severe hyperoxemia was associated with worse neurological outcome (OR 1.37 [95%CI 1.01,1.86], P=0.04) and higher mortality at longest follow-up (OR 1.32 [95%CI 1.11,1.57], P=0.002). Subgroup analyses according to timing of hyperoxemia showed that any hyperoxemia during the first 36 hours was associated with worse neurological outcome (OR 1.52 [95%CI 1.12,2.08], P=0.008) and higher mortality (OR 1.40 [95%CI 1.18,1.66], P=0.0001), whilst early hyperoxemia was not (neurological: P=0.29; mortality: P=0.19). Sensitivity analyses mostly confirmed the results of the primary analyses.Severe hyperoxemia is associated with worse neurological outcome and lower survival in CA survivors admitted to intensive care unit. Clinical efforts should be made to avoid severe hyperoxemia during at least the first 36 hours after cardiac arrest.

Target temperature management (TTM) in cardiac arrest (CA) survivors is recommended after hospital admission for its possible beneficial effects on survival and neurological outcome. Whether a lower target temperature (i.e., 32–34 °C) improves outcomes is unclear. We conducted a systematic review and meta-analysis on Pubmed and EMBASE to evaluate the effects on mortality and neurologic outcome of TTM at 32–34 °C as compared to controls (patients cared with “actively controlled” or “uncontrolled” normothermia). Results were analyzed via risk ratios (RR) and 95% confidence intervals (CI). Eight randomized controlled trials (RCTs) were included. TTM at 32–34 °C was compared to “actively controlled” normothermia in three RCTs and to “uncontrolled” normothermia in five RCTs. TTM at 32–34 °C does not improve survival as compared to normothermia (RR:1.06 (95%CI 0.94, 1.20), p = 0.36; I2 = 39%). In the subgroup analyses, TTM at 32–34 °C is associated with better survival when compared to “uncontrolled” normothermia (RR: 1.31 (95%CI 1.07, 1.59), p = 0.008) but shows no beneficial effects when compared to “actively controlled” normothermia (RR: 0.97 (95%CI 0.90, 1.04), p = 0.41). TTM at 32–34 °C does not improve neurological outcome as compared to normothermia (RR: 1.17 (95%CI 0.97, 1.41), p = 0.10; I2 = 60%). TTM at 32–34 °C increases the risk of arrhythmias (RR: 1.35 (95%CI 1.16, 1.57), p = 0.0001, I2 = 0%). TTM at 32–34 °C does not improve survival nor neurological outcome after CA and increases the risk of arrhythmias.

Aim Predicting the return of spontaneous circulation (ROSC) during cardiopulmonary resuscitation in victims of cardiac arrest (CA) remains challenging. Cerebral regional oxygen saturation (rSO2) measured during resuscitation is feasible, and higher initial and overall values seem associated with ROSC. However, these observations were limited to the analysis of few small single-centre studies. There is a growing number of studies evaluating the role of cerebral rSO2 in the prediction of ROSC. Methods We conducted an updated meta-analysis aimed at investigating the association of initial and overall values of cerebral rSO2 with ROSC after CA. We performed subgroups analyses according to the location of CA and conducted a secondary analysis according to the country where the study was conducted (resuscitation practice varies greatly for out-of-hospital CA). Results We included 17 studies. Higher initial rSO2 values (11 studies, n = 2870, 16.6% achieved ROSC) were associated with ROSC: Mean Difference (MD) -11.54 [95%Confidence Interval (CI)-20.96, -2.12]; p = 0.02 (I2 = 97%). The secondary analysis confirmed this finding when pooling together European and USA studies, but did not for Japanese studies (p = 0.06). One multi-centre Japanese study was an outlier with large influence on 95%CI. Higher overall rSO2 values during resuscitation (9 studies, n = 894, 33.7% achieving ROSC) were associated with ROSC: MD-10.38; [-13.73, -7.03]; p Conclusions This updated meta-analysis confirmed the association between higher initial and overall values of cerebral rSO2 and ROSC after CA. However, we found geographical differences, since this association was not present when Japanese studies were analysed separately. Keywords near infrared spectrometry; advanced cardiac life support; resuscitation order; cardiopulmonary resuscitation.

Objective We performed a systematic review to summarize the knowledge regarding the prevalence of burnout among intensive care unit physicians. Methods We conducted a systematic review of the MEDLINE and PubMed® databases (last update 04.02.2019) with the goal of summarizing the evidence on burnout among intensive care unit physicians. We included all studies reporting burnout in intensive care unit personnel according to the Maslach Burnout Inventory questionnaire and then screened studies for data on burnout among intensive care unit physician specifically. Results We found 31 studies describing burnout in intensive care unit staff and including different healthcare profiles. Among these, 5 studies focused on physicians only, and 12 others investigated burnout in mixed intensive care unit personnel but provided separate data on physicians. The prevalence of burnout varied greatly across studies (range 18% - 49%), but several methodological discrepancies, among them cut-off criteria for defining burnout and variability in the Likert scale, precluded a meaningful pooled analysis. Conclusion The prevalence of burnout syndrome among intensive care unit physicians is relatively high, but significant methodological heterogeneities warrant caution being used in interpreting our results. The lower reported levels of burnout seem higher than those found in studies investigating mixed intensive care unit personnel. There is an urgent need for consensus recommending a consistent use of the Maslach Burnout Inventory test to screen burnout, in order to provide precise figures on burnout in intensive care unit physicians.