Elsevier

Resuscitation

Volume 80, Issue 7, July 2009, Pages 784-789
Resuscitation

Clinical paper
Neuron specific enolase and S-100B as predictors of outcome after cardiac arrest and induced hypothermia

https://doi.org/10.1016/j.resuscitation.2009.03.025Get rights and content

Abstract

Aim

To assess the prognostic value of repetitive serum samples of neuron specific enolase (NSE) and S-100B in cardiac arrest patients treated with hypothermia.

Methods

In a three-centre study, comatose patients after cardiac arrest were treated with hypothermia at 33 °C for 24 h, regardless of cause or the initial rhythm. Serum samples were collected at 2, 24, 48 and 72 h after the arrest and analysed for NSE and S-100B in a non-blinded way. The cerebral performance categories scale (CPC) was used as the outcome measure; a best CPC of 1–2 during 6 months was regarded as a good outcome, a best CPC of 3–5 a poor outcome.

Results

One centre was omitted in the NSE analysis due to missing 24 and 48 h samples. Two partially overlapping groups were studied, the NSE group (n = 102) and the S-100B group (n = 107). NSE at 48 h >28 μg/l (specificity 100%, sensitivity 67%) and S-100B >0.51 μg/l at 24 h (specificity 96%, sensitivity 62%) correlated with a poor outcome, and so did a rise in NSE of >2 μg/l between 24 and 48 h (odds ratio 9.8, CI 3.5–27.7). A majority of missing samples (n = 123) were from the 2 h sampling time (n = 56) due to referral from other hospitals or inter-hospital transfer.

Conclusion

NSE was a better marker than S-100B for predicting outcome after cardiac arrest and induced hypothermia. NSE above 28 μg/l at 48 h and a rise in NSE of more than 2 μg/l between 24 and 48 h were markers for a poor outcome.

Introduction

Neurological sequels are common among survivors after cardiac arrest (CA) and early prognostication is difficult.1 The gold standard for prediction of outcome is a clinical neurological examination at 3 days after the arrest. A lack of motor response to pain at this time,2 or at best a flexor response,2, 3 are clinical signs consistent with a poor prognosis in unsedated CA patients.

Therapeutic hypothermia (TH) demands sedation and intermittent paralysis, limiting the value of a neurological examination. Complementary means of prognostication, unaffected by sedatives, are therefore warranted. Ideally, diagnostic tests must have a specificity of 100% for a poor outcome in order to support a decision to withdraw active treatment. Various methods have been evaluated; somatosensory evoked potentials (SSEPs) is an established method, where a bilateral lack of cortical activity to median nerve stimulation correlates with a poor outcome,4, 5, 6 but the sensitivity is low. In two recent studies, SSEP7 as well as amplitude-integrated EEG (aEEG)8 have shown promising results in predicting outcome in hypothermia treated CA patients. Among the biochemical markers, neuron-specific enolase (NSE) and S-100B are the best characterized, both have been proposed as markers of brain injury after CA,9, 10, 11, 12, 13 and have been evaluated in hypothermia treated patients as well.14, 15, 16

Neuron specific enolase is a 78 kDa dimeric intracellular glycolytic enzyme, its half-life is estimated to be 30 h.17 It is present in neurons and in other cells of neuroectodermal origin, but can also be found in erythrocytes18 and in platelets.19 S-100 is a Ca2+-binding dimeric family of proteins comprising several different subunits.20 The S-100B protein is built from either two B subunits or one B and one A1 subunit and is predominantly found in astroglia and Schwann cells, but has also been found in oligodendrocytes, in the ependymal cells of the plexus choroideus21 and in extra-cerebral tissues.22 The half-life of S-100B in peripheral blood is approximately 30 min23 and is dependant on renal function.24 NSE and S-100B have also been assessed as indicators of brain injury in stroke,25, 26 subarachnoid haemorrhage,27 trauma,28, 29and after cardiac surgery.17, 30

In a prospective study, we assessed serial serum S-100B and NSE samples in hypothermia treated patients the first 72 h after CA and correlated our findings to neurological outcome during the first 6 months. Our aim was to define cut-off values with a high specificity and a high sensitivity for a poor outcome.

Section snippets

Methods

This study was performed in the intensive care units of three hospitals in southern Sweden and consists of two patient cohorts, one for S-100B and one for NSE. The duration of the study was not pre-determined. S-100B was collected from all three hospitals between August 2002 and January 2005 and ended due to a change in analysis methodology at the local laboratory. NSE was originally collected from two hospitals, but NSE data from one hospital was omitted in the final analysis due to missing

Results

The NSE cohort consisted of 107 consecutive patients; one patient was excluded due to intracranial haemorrhage, one due to visible hemolysis in all samples and in three patients all values were missing due to early death. The S-100B cohort consisted of 116 consecutive patients; one patient was excluded due to intracranial haemorrhage, one due to renal failure and seven patients because of sampling errors. Thus, data from 102 patients in the NSE cohort and 107 patients in the S-100B cohort was

Discussion

In this prospective study, our aim was to identify cut-off values for NSE and S-100B with a high specificity and a high sensitivity for a poor outcome in CA patients treated with hypothermia. The best predictor was NSE; a value above 28 μg/l at 48 h after CA and/or a rise in NSE of more than 2 μg/l between 24 and 48 h were associated with a poor neurological outcome at 6 months.

Conclusions

NSE above 28 μg/l at 48 h and an increase in NSE of more than 2 μg/l between 24 and 48 h were markers of a poor outcome. However, a clinical decision to limit or withdraw care in comatose survivors after cardiac arrest should not rely on one single marker. Instead, we recommend the use of serial measures of NSE in combination with repetitive clinical neurological examinations, with a final decision on level of care no earlier than 72 h after normothermia.

Conflict of interest

None.

Acknowledgements

We would like to thank the statisticians at the competence centre for clinical research at Lund University and Region Skane for their assistance.

Funding sources: This study was supported by Skane county council's research and development foundation and by the governmental funding of clinical research within the Swedish National Health Service. The funding sources had no influence in study design, data collection or interpretation, writing or decision to submit the manuscript.

References (40)

  • B. Jennett et al.

    Assessment of outcome after severe brain damage

    Lancet

    (1975)
  • R. Pfeifer et al.

    Outcome after cardiac arrest: predictive values and limitations of the neuroproteins neuron-specific enolase and protein S-100 and the Glasgow Coma Scale

    Resuscitation

    (2005)
  • C.M. Booth et al.

    Is this patient dead, vegetative, or severely neurologically impaired? Assessing outcome for comatose survivors of cardiac arrest

    JAMA

    (2004)
  • D.E. Levy et al.

    Predicting outcome from hypoxic-ischemic coma

    JAMA

    (1985)
  • C. Madl et al.

    Improved outcome prediction in unconscious cardiac arrest survivors with sensory evoked potentials compared with clinical assessment

    Crit Care Med

    (2000)
  • E.G.J. Zandbergen et al.

    SSEPs and prognosis in postanoxic coma: only short or also long latency responses?

    Neurology

    (2006)
  • L.R. Robinson et al.

    Predictive value of somatosensory evoked potentials for awakening from coma

    Crit Care Med

    (2003)
  • M. Tiainen et al.

    Somatosensory and brainstem auditory evoked potentials in cardiac arrest patients treated with hypothermia

    Crit Care Med

    (2005)
  • M. Rundgren et al.

    Amplitude-integrated EEG (aEEG) predicts outcome after cardiac arrest and induced hypothermia

    Intensive Care Med

    (2006)
  • P. Martens et al.

    Serum S-100 and neuron-specific enolase for prediction of regaining consciousness after global cerebral ischemia

    Stroke

    (1998)
  • Cited by (167)

    • Biomarkers of Cerebral Injury for Prediction of Postoperative Cognitive Dysfunction in Patients Undergoing Cardiac Surgery

      2022, Journal of Cardiothoracic and Vascular Anesthesia
      Citation Excerpt :

      The lack of difference in NSE levels among patients with POCD versus patients without POCD could be a result of hemolysis caused by the CPB circuit in both groups. Supporting this is the fact that NSE levels have been associated with poor outcome in numerous other settings of acute neuronal damage.8,9,23 Tau is an axonal cytoskeleton protein, and elevated levels have been shown to be highly predictive of poor neurologic outcome after cardiac arrest.24

    View all citing articles on Scopus

    A Spanish translated version of the summary of this article appears as Appendix in the final online version at doi:10.1016/j.resuscitation.2009.03.025.

    View full text