Elsevier

Resuscitation

Volume 84, Issue 8, August 2013, Pages 1083-1088
Resuscitation

Clinical paper
Brain biomarkers and management of uncertainty in predicting outcome of cardiopulmonary resuscitation: A nomogram paints a thousand words

https://doi.org/10.1016/j.resuscitation.2013.01.031Get rights and content

Abstract

Aim

Use of brain biomarkers for predicting death after cardiopulmonary resuscitation (CPR) is limited by a research focus on the discriminative ability of each biomarker and ethical/cultural controversy concerning the likelihood of misclassification of potential survivors. We illustrate an approach to address these limitations by creating a dynamic nomogram with four levels of sensitivity (0.8, 0.9, 0.95 and 1.0) selected to represent different degrees of certainty in correct identification of survivors.

Methods

A prolective observational study conducted in a single 850-bed hospital. Admission serum S100beta (S100B) and neuron-specific enolase (NSE) were determined for all adult survivors of non-traumatic out-of-hospital arrest and CPR.

Results

158 patients were included, 126 (80%) died in hospital, 32 (20%) survived. Non-survivors had higher admission biomarker levels than survivors (p  0.001 for both S100B and NSE). Presenting rhythm (VT/VF vs. other) and logarithmic-transformed S100B and NSE levels were statistically significant in the multivariable model predicting survival. The area under the model ROC curve was 0.868 (95%CI 0.80, 0.936). Plots for predicting survival for each combination of biomarker levels were generated for each sensitivity with and without VT/VF, allowing clinicians to select their option in terms of survival probability. In this modest-sized illustrative study the model misclassified 1/19 patients with Cerebral Performance Category 1–2 for sensitivity >0.80.

Conclusions

We demonstrate how brain biomarkers can serve as decision support tools after CPR despite ethical/cultural differences in defining futility. Data from larger and diverse samples are required for stable estimates prior to clinical implementation of such a tool.

Introduction

The realization that medical care must be tailored to prognosis has led to an increasing role for prediction models in patient management. Cardiologists were the first to combine clinical presentation, electrocardiography and cardiac biomarker data to direct physician care within the context of myocardial infarction.1, 2, 3 Although it is clear that neither creatine kinases nor troponins constitute “ideal” biomarkers,4, 5, 6, 7, 8, 9 use of standardized definitions and predefined biomarker laboratory measurements have led to significant advances in the treatment of ischemic heart disease by enabling more precise disease stratification and creation of tools for directing and measuring treatment.

Within the context of cardiopulmonary resuscitation (CPR) it is increasingly clear that patient treatment after ROSC (return of spontaneous circulation) should be directed by the likelihood of neurologically intact survival as eventual outcome is determined primarily by the extent of ischemic brain damage.10, 11 The reluctance of clinicians to use biomarkers as prognostic tools within this context has been limited by two main problems: research on brain biomarkers has been performed with an almost “ascetic” attitude in that it has concentrated solely on the discriminative ability of the isolated biomarker tested without integration of the clinical picture. This has led to a (justifiable) clinician vote of no confidence because clearly there is no perfect single predictor of a poor outcome, be it severe neurological injury or death. Secondly, despite studies performed on human subjects, there remains a clear gap between “bench” research and “bedside” implementation of biomarkers; in other words, we have had little idea of how to use the biomarker data when we actually need to discuss the likelihood of death/survival with the patient's family. Under these circumstances, post-ROSC treatment decisions mostly remain based upon complex tests, incidental clinical findings and expert opinion.10, 12

Nomograms serve as a simple visual tool which can be intuitively understood even by non-professionals yet they integrate several variables to produce individual risk assessment. Nomograms can be used to calculate the relative risk of death or severe morbidity13 and nomographic prediction has been demonstrated to be superior to expert opinion in several clinical situations.14, 15 In a previous paper we demonstrated that brain biomarkers add to the predictive strength of conventional clinical variables.16 In this paper we suggest how to apply this information in practical terms; i.e. how biomarker data can be translated into a decision support tool despite cultural diversity regarding acceptable levels of misclassification of a potential survivor. We demonstrate how a nomogram or computer program which includes both biomarker and clinical data could be used to reduce the likelihood of a premature/mistaken diagnosis of death in patients who may survive.

Section snippets

Methods

All out-of-hospital cardiopulmonary arrests (OHCAs) in Israel are treated by the National EMS in accordance with ILCOR guidelines. The Shaare Zedek Medical Center (SZMC), a 850-bed university-affiliated acute care hospital in the Jerusalem district, has since 2003 become the dominant district resuscitation referral center because of its central location and its early implementation of therapeutic hypothermia protocols. Current protocols include cooling of all patients with witnessed VT/VF

Results

Among the 158 patients, 34 (21.5%) died in the ED and an additional 92 (58.2%) died later during admission. Hypothermia was induced in 21 patients. Thirty-two patients (20%) survived to hospital discharge. Survivors were younger, more likely to be men, and included a far higher proportion of patients presenting with VT/VF (Table 1). Median S100B levels at hospital arrival were 2.48 (IQR 0.79–6.75) for survivors and 7.80 (IQR 3.85–16.0) for non-survivors. Median NSE levels were 25.8 (IQR

Discussion

Within the uncertain art of medicine, clinicians, patients and families seek security in the certainty of science. This contrast is most bleak in our dealings with death, where science often fails to provide appropriate tools for prediction of the outcome. Failure in this context is, however, not only related to precision. In fact, it is often associated more with the failure of science to accommodate the diversity of cultural attitudes and beliefs. Thus, use of brain biomarkers for predicting

Conclusions

Predictive models are subject to interpretation and therefore can assist, not replace, clinical decision making. In real life the dilemma lies in the balance between sanctity and quality of life and economic constraints. Decision-assisting tools should enable clinicians and families to jointly select the degree of emphasis they place on either. In this paper we illustrate an approach to develop such a tool for patients after CPR. Using clinical and biomarker data, we created a dynamic nomogram

Authors’ contributions

SE and JDK conceived and designed the study, NK and NA contributed to data acquisition. Data were analyzed by NSL and SE with input from JDK. Interpretation of the data was undertaken by SE and JDK. SE drafted and JDK performed critical revision of the manuscript.

Funding

This study was supported by grant no. 3-00000-3160 from the Chief Scientist Office of the Ministry of Health, Israel. Study design, analyses and interpretation as well as writing of the manuscript and submission for publication were all performed by the authors.

Conflict of interest statement

The authors have no relevant financial information or potential conflicts of interest to disclose.

Acknowledgments

Our deepest thanks to Dr. Heftziba Ivgi who performed the laboratory work and to Hana Amsalem for her valuable assistance in the logistic aspects of this work. We are also indebted to the director of the department of Emergency Medicine, Dr. Todd Zalut, and his staff and to the director of the Cardiac Intensive Care Unit, Dr. Jonathan Balkin, and his staff for their ongoing personal support and willing collaboration throughout the study period.

References (33)

  • World Health Organization Regional Office for Europe

    Myocardial infarction community registers

    (1976)
  • H. Tunstall-Pedoe et al.

    Myocardial infarction and coronary deaths in the World Health Organization MONICA Project, Registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents

    Circulation

    (1994)
  • J.S. Alpert et al.

    Myocardial infarction redefined – a consensus document of The Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction

    J Am Coll Cardiol

    (2000)
  • H. Tunstall-Pedoe

    Redefinition of myocardial infarction by a consensus dissenter

    J Am Coll Cardiol

    (2001)
  • R.M. Norris

    Dissent from the consensus on the redefinition of myocardial infarction

    Eur Heart J

    (2001)
  • K. Thygesen et al.

    Universal definition of myocardial infarction

    Circulation

    (2007)
  • Cited by (0)

    A Spanish translated version of the abstract of this article appears as Appendix in the final online version at http://dx.doi.org/10.1016/j.resuscitation.2013.01.031.

    View full text