Article Text

Download PDFPDF
Use of S-100B, NSE, CRP and ESR to predict neurological outcomes in patients with return of spontaneous circulation and treated with hypothermia
  1. Seungwoon Choi1,
  2. Kyunam Park2,
  3. Seokyong Ryu1,
  4. Taekyung Kang1,
  5. Hyejin Kim1,
  6. Sukjin Cho1,
  7. Sungchan Oh1
  1. 1Department of Emergency Medicine, Inje University Sanggye Paik Hospital, Seoul, Korea
  2. 2Department of Emergency Medicine, College of Medicine, The Catholic University of Korea, Seoul, Korea
  1. Correspondence to Professor Sungchan Oh, Department of Emergency Medicine, Inje University Sanggye Paik Hospital, Sanggye 6, 7 dong, Nowon-gu, Seoul 139-707, Korea; scoh{at}paik.ac.kr

Abstract

Background With the introduction of therapeutic hypothermia (TH), the prediction of neurological outcomes in cardiac arrest (CA) survivors is challenging. Early, accurate determination of prognosis by emergency physicians is important to avoid unnecessarily prolonged critical care with a likely poor neurological outcome.

Methods This prospective observational study included patients with non-traumatic CA and return of spontaneous circulation (ROSC) between March 2009 and May 2012 at a tertiary academic hospital. Unconscious patients with ROSC were treated with mild TH (32°C–34°C) for 24 hours. Blood samples were collected for S-100B, neuron-specific enolase (NSE), C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) at 0, 24 and 48 hours post-ROSC. Neurological outcomes were evaluated at hospital discharge and dichotomised as good (cerebral performance category (CPC) 1 or 2) or poor (CPC 3, 4 or 5).

Results Of the 119 patients (68.1% male, 53±15.6 years old) who underwent TH, 46 patients had a good outcome (38.9%). Poor neurological outcomes were predicted using receiver operating characteristic analyses at cut-off values of 0.12 g/L for S-100B at 24 hours post-ROSC (sensitivity, 95.0%; specificity, 75.6%; area under the curve (AUC) 0.916; 95% CI of AUC: 0.846 to 0.961), 31.03 ng/mL for NSE at 48 hours post-ROSC (sensitivity, 83.9%; specificity, 96.9%; AUC 0.929; 95% CI of AUC: 0.836 to 0.979) and 11.2 mg/dL for CRP at 48 hours post-ROSC (sensitivity, 69.4%; specificity, 75.0%; AUC 0.731; 95% CI of AUC: 0.617 to 0.827). ESR was not significant.

Conclusions Among the biomarkers, S-100B at 24 hours and NSE at 48 hours post-ROSC were highly predictive of neurological outcomes in patients treated with TH after CA.

  • cardiac arrest
  • hypothermia
  • neurology

Statistics from Altmetric.com

Key messages

What is already known on this subject?

  • The prediction of neurological outcomes in cardiac arrest survivors following the introduction of therapeutic hypothermia is challenging.

  • Although the biomarkers S-100B protein and neuron-specific enolase (NSE) are considered useful for predicting neurological outcome in the early stages after return of spontaneous circulation (ROSC), few studies have measured S-100B and NSE during therapeutic hypothermia.

  • In addition, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels are correlated with more harmful brain injuries; however, few studies have investigated the relationships between these inflammatory markers and neurological outcome.

What might this study add?

  • In this prospective study of 119 cardiac arrest victims undergoing hypothermia treatment S-100B results 24 hours after ROSC and NSE results 48 hours after ROSC had the best prognostic value for neurological outcome.

  • Although the predictive value of CRP for neurological outcome was not as strong, it is still clinically relevant owing to its common use in most hospitals and greater availability than NSE or S-100B.

Introduction

The annual incidence of out-of-hospital cardiac arrest (OHCA) is approximately 382 000 in the USA and 24 442 in Korea.1 ,2 Patients experience ischaemic reperfusion brain injury. Because the prognosis with a return of spontaneous circulation (ROSC) is closely related with brain injury, strategies to protect the brain from injury have been developed. Therapeutic hypothermia (TH) has significantly contributed to improvements in neurological outcomes and survival rates.3

Owing to limited medical resources, patients likely to have a higher survival rate and good neurological outcome should receive more focus; unnecessary treatment for patients with minimal likelihood of neurological recovery should be avoided, reducing both suffering and medical costs. As a result, early neurological assessment is important for directing future treatment of OHCA in the clinical setting. A number of studies have been conducted to improve the quality of prognostic assessment tools including physical examinations of the brain stem, motor function tests, electroencephalography and imaging studies such as CT and MRI. These tools tend to be more useful 72 hours after ROSC rather than during the early stages after ROSC.4 ,5 Because TH sedates the patient for the first 24 hours and rewarms the patient in the next 8–12 hours, it is sometimes not possible to use these tests soon after ROSC, making prediction of the neurological outcome during the first 72 hours more difficult.6

The biomarkers S-100B protein and neuron-specific enolase (NSE) are considered useful for predicting neurological outcome in the early stages post-ROSC. They are not influenced by sedatives or muscle relaxants, and they can be measured using blood samples, which are easily obtained.7 S-100B protein is an acidic protein with a molecular weight of 21 kDa and a calcium-binding motif and is expressed in brain astrocytes and peripheral adipocytes, with an approximate half-life of 30 min.8 NSE is a glycolytic protein expressed in neurons and neuroendocrine cells with a molecular weight of 39 kDa and half-life of 30 hours.9 However, few studies have measured S-100B and NSE during TH, with varying results depending on blood sample timing and patient groups.10–12

In addition, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) tests are commonly used in clinical settings to assess inflammatory reactions, and higher ESR and CRP levels are correlated with more harmful brain injuries.13 ,14 However, few studies have investigated the relationships between these inflammatory markers and neurological outcome.

This study aimed to evaluate S-100B, NSE, CRP and ESR levels and neurological outcome in patients treated with TH after ROSC following cardiac arrest.

Methods

Patient groups

This prospective observational study included patients who experienced ROSC and underwent TH after OHCA between March 2009 and May 2012 at a tertiary academic hospital in Seoul, Korea. All cardiac arrest patients with sustained unconsciousness, regardless of location of arrest or initial rhythm, who showed ROSC, were considered for TH. The exclusion criteria for TH were age <18 years old, arrest after trauma and marked signs of infection.

The study was approved by the Institutional Review Board of the hospital. Informed consent for participation was obtained from family members or proxies of the patients.

Hypothermia treatment

Ice bags and intravenous normal saline at 4°C were used to induce rapid hypothermia with the aid of the cooling apparatus Cool Guard 3000 (Zoll, Chelmsford, Massachusetts, USA). Low body temperature was maintained with a cold blanket that had an applied thermostat and central line catheter, and a rectal probe was used to continuously monitor body temperature. After the body temperature reached 33°C, temperature was maintained at 32°C–34°C in the first 24 hours, and slow rewarming to 36.5°C was performed over the next 12 hours.

Blood samples and analysis

Blood samples were drawn from the patient three times, at ROSC, 24 and 48 hours after ROSC. S-100B was analysed using a Modular E170 (Roche Diagnostics, Mannheim, Germany), and NSE was analysed using the immunoradiometric assay Cispack NSE (Cisbio International, Codolet, France) and a GAMMA 10 (Shinjin Medics, Seoul, Korea) instrument to check for gamma rays.

Evaluation of neurological outcome

The neurological outcome was evaluated at hospital discharge. The outcome was assessed independently from the attending physicians and categorised using the cerebral performance category (CPC).15 The CPC scale ranges from 1 to 5: (1) good cerebral performance (conscious, alert, able to work, might have mild neurological or psychological deficit); (2) moderate cerebral disability (conscious, sufficient cerebral function for independent activities of daily life); (3) severe cerebral disability (conscious, dependent on others for daily support because of impaired brain function); (4) coma or vegetative state (any degree of coma without the presence of complete brain death criteria; unaware, even if appears awake (vegetative state), without interaction with the environment; may have spontaneous eye opening and sleep/awake cycles) and (5) death. A CPC of 1 or 2 is considered a good outcome, and a CPC score of 3–5 is considered a poor outcome.10 ,11 For patients who died after discharge from the intensive care unit, the better of the two scores was used, as recommended by the Utstein template.16

Analysis

Categorical variables are reported as frequencies (%), and χ2 tests were used to analyse independent variables. Continuous variables were analysed using t-tests, and variables that were not normally distributed were analysed using Mann-Whitney tests. Receiver operating characteristic curves were used to identify the cut-off S-100B, NSE, CRP and ESR values that separated the poor and good outcome groups. The area under the curve (AUC) was used to validate the usefulness of the borderline value. We also attempted a multivariate model. All clinical and laboratory predictors were included.

If a blood sample data was missing then multiple imputation was used to address missing non-outcome data.

Statistical analysis was conducted using SPSS V.16.0 (SPSS, Chicago, Illinois, USA) and MedCalc (free trial, MedCalc software, Mariakerke, Belgium), and a p<0.05 was considered statistically significant.

Results

Patient characteristics

Of the 429 OHCA patients who visited our emergency department, 211 patients experienced ROSC.

The families of 125 of these patients consented to blood sampling and TH. A patient with a traumatic injury, three patients with signs of infection and two patients without blood samples were excluded, resulting in 119 patients in the analyses (figure 1).

Figure 1

Flow diagram of included patients. *CPC, cerebral performance categories.

The mean age was 53.8±15.6, and 81 patients (68.1%) were male (table 1). A previous medical history of hypertension was present for 30 patients (25.2%) and of diabetes mellitus for 18 patients (15.1%). The initial ECG rhythm showed asystole in 51 patients (42.9%), ventricular fibrillation or pulseless ventricular tachycardia in 48 patients (40.3%) and pulseless electrical activity in 20 patients (16.8%). On-the-scene cardiopulmonary resuscitation was conducted for 55 patients (46.2%). The cause of arrest was cardiac for 90 patients (75.6%) and non-cardiac for 29 patients (24.4%). A good neurological outcome was present for 46 patients (38.7%), and a poor neurological outcome was present for 73 patients (61.3%).

Table 1

The characteristics of the patients who experienced return of spontaneous circulation and received therapeutic hypothermia after out-of-hospital cardiac arrests

The time elapsed to ROSC from the initial arrest was substantially shorter in the good neurological outcome group (22.3±12.2 min) than in the poor neurological outcome group (36.9±17.9 min) (p<0.001). Ventricular fibrillation was the initial ECG rhythm was significantly more likely in the good neurological outcome group (n=33, 71.7%) than in the poor neurological outcome group (n=15, 20.5%) (p<0.001). Significantly less epinephrine was used for the good neurological outcome group (2.0±2.7 mg) than the poor neurological outcome group (4.0±3.3 mg) (p=0.001).

Comparison of serum S-100B, NSE, CRP and ESR between good and poor neurological outcome groups

The S-100B level decreased until 48 hours post-ROSC in the good neurological outcome group, and it increased until 24 hours post-ROSC in the poor neurological outcome group, after which it decreased until 48 hours post-ROSC. The NSE values decreased after 24 hours in the good neurological outcome group and increased until 48 hours in the poor neurological outcome group. The CRP and ESR values were not statistically different between the two groups (table 2, figure 2).

Table 2

Comparison of the good and poor neurological outcome groups in relation to the S-100B, NSE and CRP levels and the ESRs

Figure 2

Serial test results of S-100B, neuron-specific enolase (NSE), C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR; 0, 24 and 48 hours after return of spontaneous circulation) in different neurological outcomes.**significantly different from the good at p<0.01; ***significantly different from the good at p<0.001.

Receiver operating characteristic curves for S-100B, NSE, CRP and ESR cut-off values

The NSE values at 48 hours and S-100B values at 24 hours had the highest AUCs (0.929 and 0.916, respectively; table 3). The 48 hours NSE cut-off value was 31.03 ng/mL, with a sensitivity of 83.9% (95% CI 66.3 to 94.5) and specificity of 96.9% (95% CI 83.8 to 99.9) for predicting a poor neurological outcome. The 24 hours S-100B cut-off value was 0.12 µg/L, with a sensitivity of 95.0% (95% CI 86.1 to 99.0) and specificity of 75.6% (95% CI 60.5 to 87.1) (figure 3).

Table 3

Receiver operating characteristic analysis of S-100B, CRP, ESR and NSE

Figure 3

Receiver operator characteristic curve for predicting poor neurological outcome using, S-100B, neuron-specific enolase (NSE), C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR).

The multivariate model did not substantially change results.

Discussion

In the present study, S-100B and NSE values were significantly different between the group of patients with good neurological outcome compared with poor neurological outcome at 0, 24 and 48 hours, S-100B was more useful than NSE at 24 hours to predict poor neurological outcomes, whereas at 48 hours, NSE was a better predictor. ESR and CRP were not useful.

NSE is a biomarker that predicts poor neurological outcomes in patients regardless of hypothermia treatment.5 ,10 In one study, decreasing NSE values at 24 and 48 hours correlated with good neurological outcome, while S-100B values were not correlated with neurological outcomes.10 However, S-100B was a more potent marker for neurological outcome than NSE in another study.17 In both studies, the entire samples were not administered hypothermia treatment.

S-100B values at 0 hour were higher than at 24 and 48 hours in both groups. We believe that, because the half-life of S-100B is approximately 30 min, S-100B levels might have quickly increased and also rapidly dissolved after the cardiac arrest-related ischaemic brain injury. NSE has a relatively longer half-life of 30 hours, S-100B might have dissolved from fat cells or chondrocytes into the blood stream during chest compression, in addition to astrocytes.18 ,19 The longer duration until ROSC experienced by the poor neurological outcome group (36.9 min), compared with the good neurological outcome group (22.3 min), could also have contributed to the higher S-100B values at 0 hour. Between 0 and 24 hours after ROSC, the S-100B value rapidly decreased in the group with a good outcome and increased in the group with a poor outcome. Because S-100B at high level stimulates the inflammatory cytokine, the increased S-100B could have led to aggravation of post-ROSC brain injury.19

Based on these results, we postulate that S-100B could be used as a biomarker for early prediction of neurological outcomes at 24 hours post-ROSC, supporting the results of Shinozaki et al.17

NSE, which indicates nerve cell damage, increases with hypoxic brain damage.11 The NSE value decreased between 0 and 24 hours in the good neurological outcome group and increased until 48 hours in the poor outcome group. In a previous study by the American Academy of Neurology, an NSE value of 33 ng/mL in the first 3 days after cardiac arrest was correlated with poor neurological outcomes.4

Rundgren et al reported that NSE values >28 ng/mL at 48 hours were indicative of a poor outcome,12 while in the present study, the NSE value of 31.03 ng/mL at 48 hours predicted a poor neurological outcome. Owing to the absence of an established gold standard method, direct comparison of NSE values is difficult, and the ability of NSE values to predict a poor neurological outcome differs by study.

CRP increases during the acute phase of inflammation; because cardiac arrest syndrome can also be considered an inflammatory response, it is expected to increase in patients experiencing cardiac arrest.20 However, additional studies are needed. Traditional CRP tests detect 3–10 ng/mL and have been replaced with high-sensitivity CRP (hsCRP) tests.21 hsCRP was evaluated in the present study instead of CRP to compare the good neurological outcome and poor neurological outcome groups; hsCRP levels were continuously elevated in both groups, from the time of ROSC until 48 hours after ROSC. Moreover, the values were significantly different between the groups at 48 hours. However, its discriminatory ability was relatively weak compared with NSE and S-100B. Because the test for CRP is used in most hospitals and more readily available than NSE or S-100B, the CRP test has its clinical relevance. Measurement of ESR has long been used as a simple, indirect test to assess inflammation; however, this test takes time (1 hour) and is influenced by sex, age or chronic disease such as anaemia, obesity and chronic kidney disease, resulting relatively low sensitivity and specificity.22 ,23 Therefore, the ESR values were not significantly different between the groups at ROSC or 24 or 48 hours after ROSC, limiting its use as a prognostic marker.

This study had limitations that should be acknowledged. First, although we attempted to obtain S-100B, NSE, CRP and ESR values at 0, 24 and 48 hours post-ROSC in all enrolled patients, this was not possible, resulting in some missing data, which might have led to biased results. Second, neurological outcomes were evaluated at hospital discharge, while in other studies, they were evaluated 6 months after hospital discharge. Hence, direct comparisons between our study and other studies are difficult because a CPC score of 1, 4 or 5 at hospital discharge would not change much after 6 months, while a CPC score of 2 or 3 would be more likely to change.24 However, since the mean CPC scores were 1.09±0.29 and 4.39±0.76 in the good neurological outcome and poor neurological outcome groups, respectively, we expect only a small difference.

In summary, for patients who underwent TH, S-100B results 24 hours after ROSC and NSE results 48 hours after ROSC had the best prognostic value for neurological outcome. However, because the specificity was not 100%, it would be imprudent to make clinical decisions solely on the results of S-100B and NSE values.

References

Footnotes

  • SC and SR contributed equally to this study.

  • Contributors SO contributed to conception and design, SR contributed to acquisition of data, TK and HK contributed to analysis to data, SeC contributed to drafting of the manuscript and SuC contributed to critical revision of the manuscript.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval IRB, Saggye Paik Hospital.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.