Protein S100B as a reliable tool for early prognostication after cardiac arrest

doi:10.1016/j.resuscitation.2020.08.010

Resuscitation

Volume 156, November 2020, Pages 251-259

https://doi.org/10.1016/j.resuscitation.2020.08.010 Get rights and content

Abstract

Purpose

Early and reliable prognostication after cardiac arrest (CA) remains crucial. We hypothesized that protein-S100B (PS100B) could predict more accurately outcome in the early phase of CA compared with other current biomarkers.

Methods

This prospective single-center study included 330 adult comatose non-traumatic successfully resuscitated CA patients, treated with targeted temperature management but not extra-corporeal life support. Lactate, pH, creatinine, NSE, and PS100B were sampled in ICU early after return of spontaneous circulation (ROSC) corresponding to admission (Adm). Serial measurements were also performed at H24 and H48. PS100B was the sole biomarker blinded to physicians.

Measurements and main results

The median delay between ROSC and first PS100B sampling was 220 min. At admission, all biomarkers were significantly associated with good outcome (CPC1–2; 109 patients) at 3-month follow-up (P ≤ 0.001, except for NSE: P = 0.03). PS100B-Adm showed the best AUC of ROC curves for outcome prediction at 3-month (AUC 0.83 [95%-CI: 0.78−0.88]), compared with other biomarkers (P < 0.0001), while AUC for lactate-Adm was higher than for NSE-Adm. AUC for PS100B-H24 was significantly higher than for other biomarkers except NSE-H24 (P ≤ 0.0001), while AUC for NSE-H24 was higher than for lactate-H24 and pH-H24. AUCs for PS100-H48 and NSE-H48 were significantly higher than for all other biomarkers (P < 0.001). Compared to patients with decreased PS100B values over time, an increasing PS100B value between admission and H24 was significantly associated with poor outcome at 3 months (P = 0.001). No-flow, initial non-shockable rhythm, PS100B-Adm, lactate-Adm, pH-Adm, clinical seizures, and absence of therapeutic hypothermia were independent predictors associated with poor outcome at 3-month in multivariate analysis. Net-Reclassification-Index was 70%, 64%, and 81% when PS100B-Adm was added to the clinical model, to clinical model with NSE-Adm, and to clinical model with standard biological parameters, respectively.

Conclusions

Early PS100B compared with other biomarkers was independently correlated with outcome after CA, with an interesting added value.

Introduction

Favourable outcome after cardiac arrest (CA) remains poor.1, 2 Early and reliable prognostication in CA patients seems of major importance. It could avoid futile treatment in patients with low chance of good outcome and could help clinicians to maximize treatment in patients who have a high likelihood of good outcome. However, guidelines emphasize that prognostication needs to be delayed, especially when targeted temperature management (TTM) is applied, and that decisions to limit care should be supported by a multimodal approach including clinical, biological, electrophysiological, and/or imaging parameters.2, 3

The usefulness of biomarkers to help clinicians in optimizing outcome prediction after CA is described mainly for Neuron Specific Enolase (NSE).2, 3, 4, 5, 6, 7, 8 The initial NSE threshold validated in 2006 did not remain 100%-predictive after a broader evaluation in TTM-treated patients,2, 4, 5 while NSE accuracy seemed better at a delayed phase after return of spontaneous circulation (ROSC).3, 4, 7, 8, 9 Other biomarkers, such as lactate, pH, and creatinine levels on admission, have also been proposed to early predict outcome after CA.10, 11, 12, 13 However, specificity of these biomarkers is not 100%, leading to difficulties in assessing their precise cut-off values and predictive accuracy.4, 14 S100B protein (PS100B) is also presumed to be an interesting prognostication tool besides NSE.3, 4, 5, 6, 7, 8, 14, 15, 16, 17, 18, 19, 20, 21, 22 S100 protein is an intracellular dimeric protein with at least 4 sub-types, with S100A1B and S100BB being presently measured by usual tests.¹⁸ Besides S100A1 protein, S100B protein (PS100B) is normally found in astroglial and Schwann cells, and in neuroctodermal tumoral cells in pathological context, while small amount of PS100 was also found in adipocytes, muscles and chondrocytes. However, the large sub-TTM report by Stammet et al. showed that PS100B could be better than NSE at H24 to predict poor outcome, but the added information was limited in all prognosticating models with or without NSE.22, 23 However, biomarkers were here only sampled 24 h after ROSC and thereafter. Therefore, NSE contrary to PS100B is preferred in international guidelines as a validated and useful biomarker since it is superior at H48 and H72.2, 3

Considering the relative short half-life of PS100B compared to NSE, we hypothesized that PS100B could more accurately predict the outcome of CA patients in the early phase after ROSC compared to biomarkers such as lactate, pH, creatinine, and NSE. The aim of the present study was to evaluate the usefulness of early PS100B sampling for prognostication in a large cohort of successfully resuscitated CA patients.

Section snippets

Methods

This prospective single-center study was carried out between March 2010 and May 2016 in the medical ICU of a university hospital (Lariboisiere Hospital, Assistance Publique des Hôpitaux de Paris, France). The Ethics Committee of our institution approved the study (Institutional Review Board of Paris North Hospital: CERB GHU Nord, N°00006477). All surviving patients hospitalized for CA -or their next of kin if necessary- gave their written informed consent. The study was declared at National

Results

Among 370 CA without ECLS implementation, 351 patients without exclusion criteria were included, of which 21 were lost to 3-month follow-up (Supplemental Fig. S1). Patients’ general characteristics are described in Table 1 and Supplemental Table S1. The delay between CA and the first blood sample measuring NSE and PS100B (Admission) was 245 min [180–338], the delay between ROSC and the first sample being 220 min [155–316]. The delay between ROSC and the second sample (H24) was 23.5 h

Discussion

Results can be summarized as follows: 1/ Initial PS100B after admission was significantly associated with good outcome at hospital discharge and 3-month after CA; 2/ PS100B-Adm. was the most accurate biomarker to correctly predict good outcome as evaluated by ROCs analyses, compared with lactate, creatinine, pH, and NSE; 3/ An increasing PS100B value from admission to H24 was significantly associated with poor outcome (1 false negative); 4/ A PS100B-Adm. threshold of 3.78 μg/L correctly

Conclusions

In our cohort of comatose patients resuscitated from CA without ECLS implementation, PS100B after hospital admission was the biomarker with the best accuracy for outcome’s prognostication, and the sole biomarker with a high accuracy persisting during the first 3 days after CA.

Funding

No specific grant was received for this study. However, this study was performed under the coordination of both “Inserm UMR S-942 et le Centres des Ressources Biologiques” of Lariboisiere University Hospital, Paris, and the doctoral school “Frontieres du Vivant, Centre de Recherches Interdisciplinaires”, Paris, France.

Conflict of interest statement and disclosures

All the authors declare no competing interest in relation with the paper.

Outside the present work, Nicolas Deye (corresponding author) declared past lecture and travel fees from Bard and ongoing lecture and travel fees from Zoll Company.

CRediT authorship contribution statement

Nicolas Deye: Conceptualization, Methodology, Supervision, Data curation, Formal analysis, Writing - original draft, Validation. Philippe Nguyen: Data curation, Formal analysis, Writing - original draft, Validation. Nicolas Vodovar: Conceptualization, Methodology, Supervision, Formal analysis, Writing - original draft, Validation. Malha Sadoune: Data curation, Validation. Corinne Collet: Data curation, Formal analysis, Writing - original draft, Validation. Sebastian Voicu: Data curation, Formal

References (34)

J.P. Nolan et al.
European Resuscitation Council and European Society of Intensive Care medicine guidelines for post-resuscitation care 2015. Section 5 of the European Resuscitation Council guidelines for resuscitation
Resuscitation
(2015)
C. Sandroni et al.
Predictors of poor neurological outcome in adult comatose survivors of cardiac arrest: A systematic review and meta-analysis. Part 2: Patients treated with therapeutic hypothermia
Resuscitation
(2013)
E. Mörtberg et al.
S-100B is superior to NSE, BDNF and GFAP in predicting outcome of resuscitation from cardiac arrest with hypothermia treatment
Resuscitation
(2011)
P. Stammet et al.
Neuron Specific Enolase as a predictor of death or poor neurological outcome after out-of-hospital cardiac arrest and targeted temperature management
J Am Coll Cardiol
(2015)
K. Shinozaki et al.
Blood ammonia and lactate levels on hospital arrival as a predictive biomarker in patients with out-of-hospital cardiac arrest
Resuscitation
(2011)
K. Shinozaki et al.
Serum S100B is superior to neuron-specific enolase as an early prognostic biomarker for neurological outcome following cardiopulmonary resuscitation
Resuscitation
(2009)
S. Einav et al.
Modeling serum biomarkers S100 beta and neuron-specific enolase as predictors of outcome after out-of-hospital cardiac arrest. An aid to clinical decision making
J Am Coll Cardiol
(2012)
S. Einav et al.
Brain biomarkers and management of uncertainty in predicting outcome of cardiopulmonary resuscitation: a nomogram paints a thousand words
Resuscitation
(2013)
M. Rundgren et al.
Neuron specific enolase and S-100B as predictors of outcome after cardiac arrest and induced hypothermia
Resuscitation
(2009)
C. Sandroni et al.
Predictors of poor neurological outcome in adult comatose survivors of cardiac arrest: A systematic review and meta-analysis. Part 1: patients not treated with therapeutic hypothermia
Resuscitation
(2013)

P. Stammet et al.

Modeling Serum level of S100B and Bispectral Index to predict outcome after cardiac arrest

J Am Coll Cardiol

(2013)

L.M. Calderon et al.

Combining NSE and S100B with clinical examination findings to predict survival after resuscitation from cardiac arrest

Resuscitation

(2014)

T. Zellner et al.

NSE and S-100B are not sufficiently predictive of neurologic outcome after therapeutic hypothermia for cardiac arrest

Resuscitation

(2013)

C.H.V. Duez et al.

Neuron-specific enolase and S-100b in prolonged targeted temperature management after cardiac arrest: a randomised study

Resuscitation

(2018)

S.S. Gul et al.

Prognostic utility of neuroinjury biomarkers in post out-of-hospital cardiac arrest (OHCA) patient management

Med Hypotheses

(2017)

W. Bougouin et al.

Characteristics and prognosis of sudden cardiac death in Greater Paris: population-based approach from the Paris Sudden Death Expertise Center (Paris-SDEC)

Intensive Care Med

(2014)

K.G. Monsieurs et al.

European Resuscitation Council Guidelines for Resuscitation 2015 Section 1. Executive summary

Resuscitation

(2015)

Cited by (11)

S100B serum level: A relevant biomarker for the management of non-traumatic headaches in emergency care?
2023, American Journal of Emergency Medicine
The diagnostic of primary or secondary headaches in emergency units is mostly based on brain imaging, which is expensive and sometimes hardly accessible. An increase in serum S100B protein has already been found in several neurological conditions inducing brain damage. The objective of this study was to assess the diagnostic performance of S100B serum assay to distinguish primary and secondary headaches among patients with non-traumatic headaches in the emergency department.
This was a phase 2, prospective, monocentric diagnostic study. Eighty-one adult patients with non-traumatic headaches in the emergency department were included. In addition to the usual management, a blood assay of the S100B protein was performed in the emergency department, as well as a brain MRI between 48 and 96 h if not performed during the initial management. The primary or secondary headache diagnosis was made at one month by an expert committee, blindly of the results of the S100B assay. The primary outcome was the blood assay of the S100B protein.
There was 63 patients for analysis in the primary headache group and 17 in the secondary headache group. The S100B protein assay was significantly higher in secondary headaches than primary headaches, with an AUC of the ROC curve of 0.67. The optimal threshold of 0.06 μg.L⁻¹ allowed to obtain those diagnostic characteristics: sensitivity 75% [48; 93], specificity 62% [48; 74], PPV 35% [20; 54] and NPV 90% [76; 97]. The association between the S100B protein level and the onset of pain was significantly higher for patients with headaches <3 h.
The assay of the S100B protein could be useful in the management of this pathology in emergencies. Future studies taking into account dosing time and etiologies could be conducted in order to refine its use in practice.
Differential Effect of Targeted Temperature Management Between 32 °C and 36 °C Following Cardiac Arrest According to Initial Severity of Illness: Insights From Two International Data Sets
2023, Chest
Recent guidelines have emphasized actively avoiding fever to improve outcomes in patients who are comatose following resuscitation from cardiac arrest (ie, out-of-hospital cardiac arrest). However, whether targeted temperature management between 32 °C and 36 °C (TTM32-36) can improve neurologic outcome in some patients remains debated.
Is there an association between the use of TTM32-36 and outcome according to severity assessed at ICU admission using a previously derived risk score?
Data prospectively collected in the Sudden Death Expertise Center (SDEC) registry (France) between May 2011 and December 2017 and in the Resuscitation Outcomes Consortium Continuous Chest Compressions (ROC-CCC) trial (United States and Canada) between June 2011 and May 2015 were used for this study. Severity at ICU admission was assessed through a modified version of the Cardiac Arrest Hospital Prognosis (mCAHP) score, divided into tertiles of severity. The study explored associations between TTM32-36 and favorable neurologic status at hospital discharge by using multiple logistic regression as well as in tertiles of severity for each data set.
A total of 2,723 patients were analyzed in the SDEC data set and 4,202 patients in the ROC-CCC data set. A favorable neurologic status at hospital discharge occurred in 728 (27%) patients in the French data set and in 1,239 (29%) patients in the North American data set. Among the French data set, TTM32-36 was independently associated with better neurologic outcome in the tertile of patients with low (adjusted OR, 1.63; 95% CI, 1.15-2.30; P = .006) and high (adjusted OR, 1.94; 95% CI, 1.06-3.54; P = .030) severity according to mCAHP at ICU admission. Similar results were observed in the North American data set (adjusted ORs of 1.36 [95% CI, 1.05-1.75; P = .020] and 2.42 [95% CI, 1.38-4.24; P = .002], respectively). No association was observed between TTM32-36 and outcome in the moderate groups of the two data sets.
TTM32-36 was significantly associated with a better outcome in patients with low and high severity at ICU admission assessed according to the mCAHP score. Further studies are needed to evaluate individualized temperature control following out-of-hospital cardiac arrest.
Poisoning-related cardiac arrest: Why prognosis should be better?
2022, Resuscitation
Reply to “OHCA (Out-of-Hospital Cardiac Arrest) and CAHP (Cardiac Arrest Hospital Prognosis) scores to predict outcome after in-hospital Cardiac Arrest: Insight from a multicentric registry”
2021, Resuscitation
Blood biomarkers of brain injury after cardiac arrest – A dynamic field
2020, Resuscitation
Predictive Performance of Neuron-Specific Enolase (NSE) for Survival after Resuscitation from Cardiac Arrest: A Systematic Review and Meta-Analysis
2023, Journal of Clinical Medicine

View all citing articles on Scopus

¹: These authors contributed equally to this work and are considered as 2 first co-authors.

View full text

Clinical paperProtein S100B as a reliable tool for early prognostication after cardiac arrest

Abstract

Purpose

Methods

Measurements and main results

Conclusions

Introduction

Section snippets

Methods

Results

Discussion

Conclusions

Funding

Conflict of interest statement and disclosures

CRediT authorship contribution statement

Resuscitation

Resuscitation

Resuscitation

J Am Coll Cardiol

Resuscitation

Resuscitation

J Am Coll Cardiol

Resuscitation

Resuscitation

Resuscitation

J Am Coll Cardiol

Resuscitation

Resuscitation

Resuscitation

Med Hypotheses

Characteristics and prognosis of sudden cardiac death in Greater Paris: population-based approach from the Paris Sudden Death Expertise Center (Paris-SDEC)

Intensive Care Med

European Resuscitation Council Guidelines for Resuscitation 2015 Section 1. Executive summary

Resuscitation

Clinical paper
Protein S100B as a reliable tool for early prognostication after cardiac arrest