Serologic Behavior of S100B Protein in Patients Who Are Brain Dead: Preliminary Results J.J. Egea-Guerrero, J. Revuelto-Rey, E. Gordillo-Escobar, A. Rodríguez-Rodríguez, J. Enamorado-Enamorado, Z. Ruiz de Azúa López, T. Aldabó-Pallás, A. León-Justel, F. Murillo-Cabezas, and A. Vilches-Arenas ABSTRACT Objective. The objective of this study is to assess the S100B protein serum concentrations from brain dead (BD) donors to understand whether its level could provide clinical information during BD diagnosis as a potential confirmatory test. Methods. During 12 months, 26 patients declared BD were prospectively included in this study. Once the diagnosis of BD was achieved, serum S100B protein levels were measured using an electrochemiluminescence assay. For analytical purposes, we selected the maximum S100B serum value reached during the first 5 days of evolution from a historical cohort of 124 survived patients after a severe brain injury (SBI), as well as from 18 healthy donors (HD) and a subgroup of patients who had severe traumatic brain injuries (TBIs) without extracranial injuries. Results. Mean age was 53.48 years (SD, 18.91 years). The BD group had significantly higher S100B serum levels (1.44 mg/L; interquartile ratio [IR], 0.63e3.68) than the SBI (0.34 mg/L; IR, 0.21e0.60) and HD groups (0.06 mg/L; IR, 0.03e0.07; P < .001). Analysis of S100B levels depending on the main cause responsible for BD development showed significant differences between subgroups (P ¼ .012). S100B serum levels were higher in the isolated TBI BD group (P ¼ .004). The S100B value showed an odds ratio for BD diagnosis of 8.38 (95% confidence interval [CI], 1.16e60.45; P ¼ .035). Reciever operating characteristic analysis revealed an area under the curve of 0.92 (95% CI, 0.79e1.00; P ¼ .007). We set a cut-off value of 2 mg/L in S100B serum concentrations. At this level, the diagnostic properties of S100B would reach 100% of specificity and positive predictive value (PPV), and sensitivity and negative predictive value (NPV) of 60% and 86.7%, respectively. Conclusion. This preliminary analysis shows for the very first time that BD is associated with higher S100B serum levels, compared with other neurocritical care patients. We also found that the cause of BD development must be considered. Specifically, S100B serum levels in severe isolated TBI patientsdwith clinical exploration compatible with BDdcould be used in a future as confirmatory test.
From the NeuroCritical Care Unit (J.J.E.-G., J.R.-R., E.G.-E., J.E.-E., Z.R.A.-L., T.A.-P., F.M.-C.), and the Department of Clinical Biochemistry (A.R.-R., A.L.-J.), Virgen del Rocío University Hospital; and the Department of Preventive Medicine and Public Health (A.V.-A.), IBIS/CSIC/University of Seville, Spain. Supported by funding from the Andalucian Society of Transplant Organs and Tissues (SATOT) and through the generous
donation of Protein S100B Electrochemiluminescence Assay Kits from Roche Diagnostics, Mannheim, Germany. Address reprint requests to Juan José Egea-Guerrero, MD, PhD, NeuroCritical Care Unit, Virgen del Rocío University Hospital, IBIS/CSIC/University of Seville, Avda. Manuel Siurot s/n. PC 41013 Seville, Spain. E-mail: juanjoegea@hotmail. com
Crown Copyright ª 2013 Published by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 Transplantation Proceedings, 45, 3569e3572 (2013)
0041-1345/13/$esee front matter http://dx.doi.org/10.1016/j.transproceed.2013.10.021 3569
3570
EGEA-GUERRERO, REVUELTO-REY, GORDILLO-ESCOBAR ET AL
T
Table 1. Characteristics, Demographics and Clinical Data on 26 Brain Death Donors
ODAY, the diagnosis of brain death (BD) requires the establishment of an irreversible cause of coma upon a strict clinical confirmationdwithout confounding factorsdof cessation of all brainstem reflexes. After that, an instrumental test is performed depending on the clinical situation of the patient or the legal requirements of each country.1,2 The most usual tests are flow or neurophysiological tests, which confirm cerebral circulatory arrest or flat electroencephalogram (EEG), respectively.2 As far as we know, there is no evidence in the literature about the potential role of biomarkers of brain damage, such as the S100B protein, as potential confirmatory tests for the BD diagnosis process. We have recently proven that this protein, released from the astroglial cell after brain injury, could serve as screening tool for the early detection of patients at risk for BD development after severe traumatic brain injury (TBI).3e5 However, little is known about the serologic behavior of the S100B protein after the establishment of BD status. The objective of this study is to assess the S100B protein serum concentrations from BD donors to understand whether its level could provide clinical information during BD diagnosis as a potential confirmatory test. METHODS During 12 months, 26 patients declared BD in the Neurocritical Care Unit of the Virgen del Rocío University Hospital in Seville, Spain, were prospectively included in this study. The protocol was approved by the Virgen del Rocío Hospital Institutional Review Board and the Andalucian Society of Transplant Organs and Tissues. Written consent was obtained from the patients’ relatives. BD diagnosis was established following legal provisions for BD determination in Spain and internationally recognized clinical requirements. All patients fulfilled all the clinical criteria for BD and a mandatory instrumental test was performed (transcranial doppler or EEG). Once the diagnosis of BD was achieved, 9 mL of blood was extracted for sampling. Serum S100B protein levels were measured using an electrochemiluminescence assay, produced by Elecsys 2010 immunoassay systems (Roche Diagnostics, Germany). For analytical purposes, we selected the maximum S100B serum value reached during the first 5 days of evolution from a historical cohort of 124 survived patients after a severe brain injury (SBI) condition with no renal impairment, as well as S100B level from 18 healthy donors. To avoid bias about the origin of the S100B protein after the injury 6 we also studied a subgroup of patients who had severe TBI and without extracranial injuries (five patients from BD group vs. 13 survivors after isolated severe TBI).
Gender, male (%) Age, mean (SD) Cause of brain death, n (%) Traumatic brain injury Intracerebral hemorrhage Others Medical history Hypertension Diabetes Dyslipidemia Cardiac arrest before hospital admission, n (%) Catecholamine use, n (%) Intracranial pressure monitoring, n (%) Renal function Serum creatinine (mg/dL), (mean [SD]) Urine output (mL/h) (median [IR]) eGFR (mL/min/1.73 m2) (median [IR])
18 (69.23) 53.48 (18.91) 11 (42.31) 10 (38.46) 5 (19.23) 11 9 5 3
(42.30) (34.61) (19.23) (11.53)
22 (84.61) 2 (7.69) 0.75 (0.25) 150 (107.50e235.00) 94.25 (75.57e156.00)
Abbreviations: SD, standard deviation; IR, interquartile range; eGFR, estimated glomerular filtration rate.
subgroup were used to differentiate patients as BD or non-BD. This analysis was used to explore a preliminary cut-off value for patient classification. Statistical analyses were conducted using SPSS (20.0, IBM, Chicago, Ill, United States).
RESULTS
The mean age was 53.48 years (SD, 18.91 years). Patients evolutions to BD were mainly due to severe TBIs (42.31%). No renal impairment was found in this sample. Table 1 provides a summary of the BD patients’ demographic and clinical data. As shown in Fig 1, the BD group had significantly higher S100B serum levels (1.44 mg/L; IR, 0.63e3.68) than the SBI (0.34 mg/L; IR, 0.21e0.60) and healthy donor groups (0.06 mg/L; IR, 0.03e0.07; P < .001). Analysis of S100B levels depending on the main cause responsible for
Statistical Analysis Quantitative variables were expressed using mean and standard deviation (SD) or median and interquartile range (IR). Comparisons of means of quantitative variables between subgroups were made applying the Student t test or ManneWhitney U test. For comparing more than two groups, the KruskaleWallis test was applied. Analysis of relationships between variables was made by contingency tables using the chi-square test or the Fisher exact test. Inferential analyses were performed using multivariate logistic regression. Odds ratio (OR) with 95% confidence interval (CI) was calculated for variables. Receiver operating characteristic (ROC) curve analysis on S100B serum values from the isolated TBI
Fig 1. S100B serum levels distribution in study population.
S100B PROTEIN AND BRAIN DEATH
BD development showed significant differences between subgroups (P ¼ .012). Concretely, those patients who progressed to BD as a result of TBI had higher S100B serum concentrations (2.57 mg/L; IR, 1.33e6.30) compared with patients who presented BD development due to intracerebral hemorrhage (1.39 mg/L; IR, 0.50e4.83) and other causes (0.43 mg/L; IR, 0.23e0.63). A second analysis was performed focusing on isolated TBI patients (BD and non-BD groups). As shown in Table 2, S100B serum levels were higher in the isolated TBI BD group (P ¼ .004). No significant differences were found in median S100B values according to age, gender, and the presence of diffuse or mass lesions in computed tomographic scans at admission. After multivariate analysis, the only remaining variable was the S100B protein. We found that for each 1-mg/L increase in S100B value, the OR for BD diagnosis was 8.38 (95% CI, 1.16e60.45; P ¼ .035). Figure 2 shows the ROC curve for BD in patients after isolated TBI (area under the curve [AUC] 0.92 [95% CI, 0.79e1.00; P ¼ .007]). To maximize the relationship between sensitivity and specificity and avoid any false positive result, we considered from this preliminary study a cut-off value of 2 mg/L in S100B serum concentrations. At this level, the diagnostic properties of S100Bdas an ancillary test of BD after isolated TBIdwould reach 100% of specificity and positive predictive value, and sensitivity and negative predictive value of 60% and 86.7%, respectively. DISCUSSION
This preliminary analysis shows for the very first time that BD is associated with higher S100B serum levels compared
3571 Table 2. Main Patients’ Clinical Characteristics Divided Into Brain Death Vs. Nonebrain death Subgroups After Severe Isolated Traumatic Brain Injury BD After Isolated TBI, No 13 (72.2) Yes 5 (27.8) n (%) Age, mean (IR) 54 (47e60.50) 46 (30.50e71) Gender, male (%) 8 (61.5) 5 (38.5) CT-scan, n (%)* Diffuse lesion 3 (23.1) 1 (20) Mass lesion 10 (76.9) 4 (80) S100B protein, median 0.37 (0.22e0.68) 2.13 (1.04e12.39) (IR)**
P .712 .249 .999 .004
Abbreviations: BD, brain death; TBI, traumatic brain injury; IR, interquartile range; CT, computed tomography. *According to Traumatic Coma Data Bank. **S100B values (mg/L).
with other neurocritical care patients. We also found that the cause of BD development must be considered in this situation. Specifically, S100B serum levels in severe isolated TBI patientsdwith clinical exploration compatible with BDdcould be used in a future as a confirmatory test. Selecting patients with head trauma and no other injuries implies a strict inclusion criteria that led us to a very homogeneous population of BD patients.6 Our results should be interpreted with caution due to our small sample size during this 1-year period. To the authors’ knowledge, no evidence appears to exist about the role of biochemical serum markers as confirmatory tests in patients who have clinically fulfilled all BD criteria. Some authors have described previously that S100B protein levels are higher during the evolution of SBI patients who subsequently evolved to BD.4,5 Nevertheless, no study has focused specifically on serum concentration of this protein when the patient is declared dead under neurological criteria. According to these data, our results open a new future field of study in which the potential role of serum biomarkers that could cut down on unnecessary instrumental tests for BD confirmation. Most of those tests require specific physicians qualifications, are sometimes not easy to perform, and consequently cause a delay of BD diagnosis. However, whether a simple blood test may help physicians during the BD diagnosis must be investigated with larger sample sizes and under different neurocritical care conditions. REFERENCES
Fig 2. ROC analysis comparing sensitivity to specificity of S100B in serum to determine brain death after severe isolated traumatic brain injury.
1. Wijdicks EF. Brain death worldwide: accepted fact but no global consensus in diagnostic criteria. Neurology. 2002;58: 20e25. 2. Wijdicks EF. The diagnosis of brain death. N Engl J Med. 2001;344:1215e1221. 3. Egea-Guerrero JJ, Revuelto-Rey J, Murillo-Cabezas F, et al. Accuracy of the S100b protein as a marker of brain damage in traumatic brain injury. Brain Inj. 2012;26:76e82. 4. Egea-Guerrero JJ, Murillo-Cabezas F, Gordillo-Escobar E, et al. S100B Protein may detect brain death development after severe traumatic brain injury. J Neurotrauma. 2013;30:1762e1769.
3572 5. Dimopoulou I, Korfias S, Dafni U, et al. Protein S-100b serum levels in trauma induced brain death. Neurology. 2004;60: 947e951.
EGEA-GUERRERO, REVUELTO-REY, GORDILLO-ESCOBAR ET AL 6. Anderson RE, Hansson L-O, Nilsson O, et al. High serum S100B levels for trauma patients without head injuries. Neurosurgery. 2001;48:1255e1260.
From the NeuroCritical Care Unit (J.J.E.-G., J.R.-R., E.G.-E., J.E.-E., Z.R.A.-L., T.A.-P., F.M.-C.), and the Department of Clinical Biochemistry (A.R.-R., A.L.-J.), Virgen del Rocío University Hospital; and the Department of Preventive Medicine and Public Health (A.V.-A.), IBIS/CSIC/University of Seville, Spain. Supported by funding from the Andalucian Society of Transplant Organs and Tissues (SATOT) and through the generous
donation of Protein S100B Electrochemiluminescence Assay Kits from Roche Diagnostics, Mannheim, Germany. Address reprint requests to Juan José Egea-Guerrero, MD, PhD, NeuroCritical Care Unit, Virgen del Rocío University Hospital, IBIS/CSIC/University of Seville, Avda. Manuel Siurot s/n. PC 41013 Seville, Spain. E-mail: juanjoegea@hotmail. com
Crown Copyright ª 2013 Published by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 Transplantation Proceedings, 45, 3569e3572 (2013)
0041-1345/13/$esee front matter http://dx.doi.org/10.1016/j.transproceed.2013.10.021 3569
3570
EGEA-GUERRERO, REVUELTO-REY, GORDILLO-ESCOBAR ET AL
T
Table 1. Characteristics, Demographics and Clinical Data on 26 Brain Death Donors
ODAY, the diagnosis of brain death (BD) requires the establishment of an irreversible cause of coma upon a strict clinical confirmationdwithout confounding factorsdof cessation of all brainstem reflexes. After that, an instrumental test is performed depending on the clinical situation of the patient or the legal requirements of each country.1,2 The most usual tests are flow or neurophysiological tests, which confirm cerebral circulatory arrest or flat electroencephalogram (EEG), respectively.2 As far as we know, there is no evidence in the literature about the potential role of biomarkers of brain damage, such as the S100B protein, as potential confirmatory tests for the BD diagnosis process. We have recently proven that this protein, released from the astroglial cell after brain injury, could serve as screening tool for the early detection of patients at risk for BD development after severe traumatic brain injury (TBI).3e5 However, little is known about the serologic behavior of the S100B protein after the establishment of BD status. The objective of this study is to assess the S100B protein serum concentrations from BD donors to understand whether its level could provide clinical information during BD diagnosis as a potential confirmatory test. METHODS During 12 months, 26 patients declared BD in the Neurocritical Care Unit of the Virgen del Rocío University Hospital in Seville, Spain, were prospectively included in this study. The protocol was approved by the Virgen del Rocío Hospital Institutional Review Board and the Andalucian Society of Transplant Organs and Tissues. Written consent was obtained from the patients’ relatives. BD diagnosis was established following legal provisions for BD determination in Spain and internationally recognized clinical requirements. All patients fulfilled all the clinical criteria for BD and a mandatory instrumental test was performed (transcranial doppler or EEG). Once the diagnosis of BD was achieved, 9 mL of blood was extracted for sampling. Serum S100B protein levels were measured using an electrochemiluminescence assay, produced by Elecsys 2010 immunoassay systems (Roche Diagnostics, Germany). For analytical purposes, we selected the maximum S100B serum value reached during the first 5 days of evolution from a historical cohort of 124 survived patients after a severe brain injury (SBI) condition with no renal impairment, as well as S100B level from 18 healthy donors. To avoid bias about the origin of the S100B protein after the injury 6 we also studied a subgroup of patients who had severe TBI and without extracranial injuries (five patients from BD group vs. 13 survivors after isolated severe TBI).
Gender, male (%) Age, mean (SD) Cause of brain death, n (%) Traumatic brain injury Intracerebral hemorrhage Others Medical history Hypertension Diabetes Dyslipidemia Cardiac arrest before hospital admission, n (%) Catecholamine use, n (%) Intracranial pressure monitoring, n (%) Renal function Serum creatinine (mg/dL), (mean [SD]) Urine output (mL/h) (median [IR]) eGFR (mL/min/1.73 m2) (median [IR])
18 (69.23) 53.48 (18.91) 11 (42.31) 10 (38.46) 5 (19.23) 11 9 5 3
(42.30) (34.61) (19.23) (11.53)
22 (84.61) 2 (7.69) 0.75 (0.25) 150 (107.50e235.00) 94.25 (75.57e156.00)
Abbreviations: SD, standard deviation; IR, interquartile range; eGFR, estimated glomerular filtration rate.
subgroup were used to differentiate patients as BD or non-BD. This analysis was used to explore a preliminary cut-off value for patient classification. Statistical analyses were conducted using SPSS (20.0, IBM, Chicago, Ill, United States).
RESULTS
The mean age was 53.48 years (SD, 18.91 years). Patients evolutions to BD were mainly due to severe TBIs (42.31%). No renal impairment was found in this sample. Table 1 provides a summary of the BD patients’ demographic and clinical data. As shown in Fig 1, the BD group had significantly higher S100B serum levels (1.44 mg/L; IR, 0.63e3.68) than the SBI (0.34 mg/L; IR, 0.21e0.60) and healthy donor groups (0.06 mg/L; IR, 0.03e0.07; P < .001). Analysis of S100B levels depending on the main cause responsible for
Statistical Analysis Quantitative variables were expressed using mean and standard deviation (SD) or median and interquartile range (IR). Comparisons of means of quantitative variables between subgroups were made applying the Student t test or ManneWhitney U test. For comparing more than two groups, the KruskaleWallis test was applied. Analysis of relationships between variables was made by contingency tables using the chi-square test or the Fisher exact test. Inferential analyses were performed using multivariate logistic regression. Odds ratio (OR) with 95% confidence interval (CI) was calculated for variables. Receiver operating characteristic (ROC) curve analysis on S100B serum values from the isolated TBI
Fig 1. S100B serum levels distribution in study population.
S100B PROTEIN AND BRAIN DEATH
BD development showed significant differences between subgroups (P ¼ .012). Concretely, those patients who progressed to BD as a result of TBI had higher S100B serum concentrations (2.57 mg/L; IR, 1.33e6.30) compared with patients who presented BD development due to intracerebral hemorrhage (1.39 mg/L; IR, 0.50e4.83) and other causes (0.43 mg/L; IR, 0.23e0.63). A second analysis was performed focusing on isolated TBI patients (BD and non-BD groups). As shown in Table 2, S100B serum levels were higher in the isolated TBI BD group (P ¼ .004). No significant differences were found in median S100B values according to age, gender, and the presence of diffuse or mass lesions in computed tomographic scans at admission. After multivariate analysis, the only remaining variable was the S100B protein. We found that for each 1-mg/L increase in S100B value, the OR for BD diagnosis was 8.38 (95% CI, 1.16e60.45; P ¼ .035). Figure 2 shows the ROC curve for BD in patients after isolated TBI (area under the curve [AUC] 0.92 [95% CI, 0.79e1.00; P ¼ .007]). To maximize the relationship between sensitivity and specificity and avoid any false positive result, we considered from this preliminary study a cut-off value of 2 mg/L in S100B serum concentrations. At this level, the diagnostic properties of S100Bdas an ancillary test of BD after isolated TBIdwould reach 100% of specificity and positive predictive value, and sensitivity and negative predictive value of 60% and 86.7%, respectively. DISCUSSION
This preliminary analysis shows for the very first time that BD is associated with higher S100B serum levels compared
3571 Table 2. Main Patients’ Clinical Characteristics Divided Into Brain Death Vs. Nonebrain death Subgroups After Severe Isolated Traumatic Brain Injury BD After Isolated TBI, No 13 (72.2) Yes 5 (27.8) n (%) Age, mean (IR) 54 (47e60.50) 46 (30.50e71) Gender, male (%) 8 (61.5) 5 (38.5) CT-scan, n (%)* Diffuse lesion 3 (23.1) 1 (20) Mass lesion 10 (76.9) 4 (80) S100B protein, median 0.37 (0.22e0.68) 2.13 (1.04e12.39) (IR)**
P .712 .249 .999 .004
Abbreviations: BD, brain death; TBI, traumatic brain injury; IR, interquartile range; CT, computed tomography. *According to Traumatic Coma Data Bank. **S100B values (mg/L).
with other neurocritical care patients. We also found that the cause of BD development must be considered in this situation. Specifically, S100B serum levels in severe isolated TBI patientsdwith clinical exploration compatible with BDdcould be used in a future as a confirmatory test. Selecting patients with head trauma and no other injuries implies a strict inclusion criteria that led us to a very homogeneous population of BD patients.6 Our results should be interpreted with caution due to our small sample size during this 1-year period. To the authors’ knowledge, no evidence appears to exist about the role of biochemical serum markers as confirmatory tests in patients who have clinically fulfilled all BD criteria. Some authors have described previously that S100B protein levels are higher during the evolution of SBI patients who subsequently evolved to BD.4,5 Nevertheless, no study has focused specifically on serum concentration of this protein when the patient is declared dead under neurological criteria. According to these data, our results open a new future field of study in which the potential role of serum biomarkers that could cut down on unnecessary instrumental tests for BD confirmation. Most of those tests require specific physicians qualifications, are sometimes not easy to perform, and consequently cause a delay of BD diagnosis. However, whether a simple blood test may help physicians during the BD diagnosis must be investigated with larger sample sizes and under different neurocritical care conditions. REFERENCES
Fig 2. ROC analysis comparing sensitivity to specificity of S100B in serum to determine brain death after severe isolated traumatic brain injury.
1. Wijdicks EF. Brain death worldwide: accepted fact but no global consensus in diagnostic criteria. Neurology. 2002;58: 20e25. 2. Wijdicks EF. The diagnosis of brain death. N Engl J Med. 2001;344:1215e1221. 3. Egea-Guerrero JJ, Revuelto-Rey J, Murillo-Cabezas F, et al. Accuracy of the S100b protein as a marker of brain damage in traumatic brain injury. Brain Inj. 2012;26:76e82. 4. Egea-Guerrero JJ, Murillo-Cabezas F, Gordillo-Escobar E, et al. S100B Protein may detect brain death development after severe traumatic brain injury. J Neurotrauma. 2013;30:1762e1769.
3572 5. Dimopoulou I, Korfias S, Dafni U, et al. Protein S-100b serum levels in trauma induced brain death. Neurology. 2004;60: 947e951.
EGEA-GUERRERO, REVUELTO-REY, GORDILLO-ESCOBAR ET AL 6. Anderson RE, Hansson L-O, Nilsson O, et al. High serum S100B levels for trauma patients without head injuries. Neurosurgery. 2001;48:1255e1260.
