November 9, 2020

An article review from the ESICM Journal Review Club

Prediction of poor neurological outcome in comatose survivors of cardiac arrest: a systematic review

The evidence concerning predictors of poor neurological outcome in comatose survivors of Cardiac Arrest (CA) was systematically reviewed by the European Resuscitation Council (ERC) and the European Society of Intensive Care Medicine (ESICM) in 2013. Therefore, an update on more recent studies was recommended.

Sandroni and co-authors attempted to assess the ability of clinical examination, blood biomarkers, electrophysiology, or neuroimaging recorded within 7 days from return of spontaneous circulation (ROSC) to predict poor neurological outcome, defined as death, vegetative state, or severe disability (CPC 3–5) at hospital discharge/1 month or later, in comatose adult survivors from CA.

PubMed, EMBASE, Web of Science, and the Cochrane Database of Systematic Reviews (January 2013–April 2020) were searched.

94 studies (30,200 patients) were included. Sensitivity and false-positive rate (FPR) for each predictor were calculated.

Reliable predictors of poor neuro-prognosis post-ROSC with 0% FPR for poor outcome in most studies resulted in:

  • Absent pupillary/corneal reflexes after day 4 from ROSC. Consider the use of automated infrared pupillometry for an earlier and more accurate prediction.
  • High values of neuron-specific enolase (NSE) from 24 hours after ROSC, with a maximal accuracy achieved at 48-72 hours. NSE can predict poor outcome up to 7 days after ROSC.
  • Absent N20 waves of short latency Somatosensory Evoked Potentials (SSEPs), from the day of ROSC.
  • EEG background of suppression and burst-suppression, especially after 24 hours from ROSC (EEG patterns were defined using the ACNS terminology).
  • Unequivocal seizures (ACNS-defined) from 0-8 hours to 77 hours after ROSC.
  • Quantitative estimates of cerebral oedema on a brain CT (2 hours after ROSC) and reduced diffusion on MRI (2–5 days after ROSC).


  • An impressive number of studies and patients were considered.
  • Conclusions based on a solid review of current knowledge and recent advances in prognostication.
  • The results of the review could be part of the upcoming guidelines on post-resuscitation care by the ESICM and ERC.
  • Indices of neurological recovery were not considered and analysed.
  • Only the reliability of single predictors was assessed, not their combination.
  • A prognostication based on a neurological outcome at 6 months or later, instead of at hospital discharge/1 month, may be more appropriate.
  • Most of the studies included were biased by lack of blinding and by potential confounding factors from sedation and the use of TTM.
  • None of the indices evaluated enabled prediction of poor neurological outcome with absolute certainty.

In comatose resuscitated patients, clinical, biochemical, neurophysiological, and radiological tests have a potential to predict poor neurological outcome, with no false-positive predictions within the first week after CA.  However, most of the indices have a low sensitivity and a risk of bias and confounders.  A multimodal approach still appears to be the most prudent prognostication strategy.

This article review was prepared and submitted by Dr. Serena Ranieri, Department of Anaesthesia, Intensive Care and Emergency Medical Services, Maggiore Hospital, Bologna (Italy), on behalf of the ESICM Journal Review Club.

Sandroni, Claudio, et al. “Prediction of poor neurological outcome in comatose survivors of cardiac arrest: a systematic review.” Intensive Care Medicine (2020): 1-49.

Comment on this news