The clinical decision rules for a cranial computed tomography (CCT) scan in a patient with minor head injury (MHI) remain controversial (1, 2). Although the total number of MHI patients is large (approximately 800,000/year for the United States) (2), potentially life-threatening intracranial lesions are detected by CCT scan in only a few such patients (3% - 20 %) (1, 3-5). However, patients with intracranial lesions, such as intracerebral hemorrhage, diffuse brain edema, or skull fractures, are predisposed to develop severe post-traumatic neurological deficits similar to those occurring after either moderate or severe head trauma (scores of 9 - 12 and 3-8, respectively, on the Glasgow Coma Scale [GCS]) (6). Therefore, in any clinical decision rules for MHI, rapid and reliable identification of patients with intracranial lesions is critical to avoid post-traumatic complications and secondary brain damage (7). Because clinical symptoms such as headache, vomiting, amnesia, and loss of consciousness are significantly correlated with radiologically detected intracerebral lesions, several studies have defined those clinical indicators as requiring a CCT scan (1, 2, 8). However, the low efficiency of the discrimination between CCT-positive (CCT+, 6% - 10%) and CCT-negative (CCT−, 90% - 94%) patients using the above-mentioned clinical parameters may be especially because of the multifactorial causes of these (e.g., nausea or vomiting). Although these neurological signs are often thought to be clear indicators of intracerebral lesions, the frequency of coactivators, (such as alcohol intoxication, is substantially higher (30% - 50%) (9,10) than that of actuating lesions (10%) themselves. Hence, an additional objective parameter that can be easily and rapidly measured would be of outstanding clinical value to optimize the clinical decision rules for an initial CCT scan.
In this context, the assessment of the astroglial-derived protein S-100B released into blood has been considered not only as a potential objective quantitative prognostic marker of severe head injury (11-15) but also as a reliable screening tool to identify those patients who indeed have post-traumatic lesions in the large pool of MHI patients (16-19).
Astroglial S-100B consists of a small dimeric protein (S-100BB) with a molecular weight of approximately 21 kDa, consisting of ββ-chains and belonging to a multigenic family of calcium-binding proteins expressing a great variety of homodimeric and heterodimeric proteins in diverse mammalian cells (20-22). S-100BB (ββ-chain) is predominantly expressed by cells of the central nervous system, mainly astroglial cells and also neuronal cells. However, it can also be detected in melanoma cells and, to some extent, in other tissues such as fat. The functional protein is implicated in a variety of intracellular and extracellular regulatory activities (e.g., protein phosphorylation, cell motility, neuronal differentiation, and proliferation) (21). Although several studies raised substantial hints that the measurement of S-100B in the systemic circulation of MHI patients might be an additional valuable tool to optimize clinical decision rules for an initial CCT scan, its value has not yet been shown in a large enough group of patients to be statistically reliable. Hence, the aim of this multicenter study was to clarify whether the serum S-100B concentration, measured in more than 1300 patients with MHI, might be a statistically reliable and safe means of identifying patients without CCT-detectable intracerebral lesions, thereby avoiding costs and unnecessary CCT scans.
We set the following targets: (1) to identify a physiological cutoff level by investigating serum S-100B concentrations in a large normal, healthy control group (n = 540); (2) to calculate sensitivity, specificity, and positive and negative predictive values for the classification of MHI patients into the clinically relevant group, CCT− or CCT+; and (3) to analyze the data by receiver operating characteristic (ROC) curve to determine the diagnostic value of S-100B.
Study design and groups
This prospective study was carried out from June 2002 to October 2003 in 3 study sites at level I trauma centers in Germany: the Department of Traumatology and Orthopedic Surgery-Central, University of Munich; the Department of Traumatology and Orthopedic Surgery, Academic Teaching Hospital München-Schwabing, Munich; and the Department of Traumatology, Hand Surgery and Reconstructive Surgery, University of Frankfurt.
The study was approved by the ethics committee of the University of Munich (study sites: Ludwig-Maximilians-University and Urban Hospital Munich-Schwabing, reference number 247/02) and based on this positive approval also by the University of Frankfurt. Informed consent was obtained from all healthy volunteers and from all patients (either immediately before inclusion in the study or immediately after they had regained vigilance after head trauma). If a patient with head trauma did not reach an appropriate level of consciousness to agree to participate in the study, a legal representative or close relative was asked for consent.
In the study regions (Central Europe), MHI patients are generally seen first by surgeons on call in surgical or traumatological emergency departments. Because a neurosurgeon is consulted only if pathological intracranial lesions are detected in the CCT scan, most of MHI patients are not seen by a neurosurgeon in the first instance. The decision whether to order a CCT scan in MHI patients is made by the physician in the emergency department.
From June 2002 until the end of October 2003, all patients at the 3 study sites who fulfilled the following inclusion criteria were enrolled prospectively: history of isolated head trauma and admission within 3 h; GCS score of 13 to 15 upon admission; and one or more of 10 clinical risk factors: brief loss of consciousness, post-traumatic amnesia, nausea, vomiting, severe headache, dizziness, vertigo, intoxication, anticoagulation, and age of above 60 years. These inclusion criteria were chosen in accordance with recently published recommendations (1, 8) for clinical decision rules for an initial CCT scan in MHI patients (Table 1). Individuals underthe age of 18 years, pregnant women, prisoners, and multiple-injured patients were excluded.
Once a patient was enrolled in the study, a venous blood sample was drawn and the patient thereafter underwent a CCT scan.
Negative control group-
Healthy individuals (n = 540) without a history of head trauma were enrolled as a negative control group. These individuals were recruited from a group of voluntary blood donators who were checked on their health and potential head trauma status by a standardized set of questions. The demographic data are shown in Table 1. Venous blood samples were drawn and processed as described below.
Positive control group-
As positive control, 55 patients with moderate to severe head injury (moderate, 9 - 12 points on the GCS; severe, ≤8 points on the GCS) were studied; the epidemiological data are shown in Table 1. Blood samples were drawn upon admission, and all patients underwent a CCT scan.
Confirmation of diagnosis by CCT
An emergency CCT scan was performed in accordance with the following protocol: 2/2/4 mm and 8/8/8 mm kernel with additional bone window reconstruction (2/2/2 mm) for the skull base (23). To determine whether a patient had a trauma-relevant intracerebral lesion in accordance with data from literature (1, 2, 8), the radiological parameters given in Table 2 were recorded and the patients divided into 2 groups: CCT-negative (CCT−; MHI patients without any signs of trauma-relevant intracerebral lesions) and CCT-positive (CCT+; MHI patients with at least one of the pathophysiological trauma-relevant findings shown in Table 2).
Measurement of S-100B Concentrations
Venous blood samples were processed to serum and deep-frozen at −20°C until assay with an electrochemiluminescence immunoassay kit (Elecsys S100; Roche Diagnostics, Mannheim, Germany). According to the manufacturer's instructions, the test system requires 18 min and a probe volume of at least 20-μL serum. The lower detection limit is 0.005 μg/L, and concentrations of up to 39 μg/L can be measured without dilution. The results are reported in micrograms per liter and rounded off to 2 decimal places. Values of S-100B had no effect on the clinical management of the patients.
All statistical tests were performed using the SigmaStat 3.2 software package (Microsoft Corporation, Seattle, Wash). Because the demographic and epidemiological data were not normally distributed, they are reported as median and interquartile range. Comparison of S-100B values among study and control groups (the negative controls were healthy volunteers; the positive controls were patients with a GCS score from 3 to 12 points) were performed using nonparametric U test. The distribution in the negative control group was analyzed to determine the physiological S-100B concentration. The cutoff value was set at the 95th percentile of these data, which is based on the assumption that this is the limit between physiological and pathophysiological S-100B serum concentrations. Patients exhibiting serum concentrations below the said cutoff were considered S-100B-negative (S-100B−), and those above were considered S-100B-positive (S-100B+). For subgroup analysis, we compared the S-100B concentrations between patients with trauma-relevant intracerebral lesions (CCT+) with those without such lesions (CCT−) using U test. Moreover, we tested using analysis of variance (ANOVA) on ranks followed by the post hoc Dunn multiple comparisons procedure to see whether the S-100B concentrations varied in CCT+patients in the group with GCS scores from 13 to 15 and between the CCT− group with those scores.
To determine the sensitivity, specificity, and positive and negative predictive values at the cutoff level, a contingency table was constructed. Patients were judged in accordance with their serum S-100B concentrations as S-100B+ or S-100B−, and this discrimination was cross-checked against results of the CCT scan (CCT+ or CCT−). To determine the diagnostic accuracy, the positive and negative likelihood ratios were calculated. For all of these values the 95% confidence interval (CI) was calculated using the normal approximation method.
To determine the discriminative ability of S-100B serum measurements in MHI patients, an ROC curve was calculated according to the dichotomous variable CCT− or CCT+.
For a more detailed insight into the characteristics of S-100B serum concentrations, we stratified the MHI study group according to whether their initial GCS score upon admission was 13, 14, or 15.
Negative control group (n = 540)
The demographic data of the control group are given in Table 1. The median concentration of S-100B was 0.05 μg/L and the interquartile range was 0.03 to 0.06 μg/L (Table 3). The 95th percentile was at 0.10 μg/L (Fig. 1). This concentration was considered the cutoff level to discriminate between S-100B+ and S-100B− MHI patients.
Positive control group, GCS 3 to 12 (n = 55)
Fifty-five patients with moderate or severe head injury showed an initial GCS below 13 points. Their demographic data and clinical characteristics are given in Table 1. Forty-two percent of these patients (n = 23) had trauma-relevant intracranial lesions according to the criteria given in Table 2, and 20% required neurosurgical intervention - 2 for implantation of intraventricular catheters for drainage of cerebrospinal fluid and 9 for decompressive craniotomies.
The S-100B serum concentrations are reported in Table 3 and were significantly higher (P < 0.001 on the U test) in the negative control and the MHI groups. From these, only 6 patients (11%) had an S-100B serum concentration below the cutoff, and none of these 6 had a positive CCT scan. Most (89%) of this collective was above the cutoff value.
MHI group (n = 1309)
A total of 1309 patients fulfilled the inclusion criteria of the study group. The demographic and clinical baseline characteristics are given in Table 1. The median interval between trauma and blood sampling was 60 min (range, 40 - 80 or 25% - 75%); the median interval between blood sampling and CCT scan was 30 min (range, 16 - 52 or 25% - 75%).
Of these MHI patients, 1216 (93%) were proved to be CCT− and 93 (7%) were proved to be CCT+ on the initial CCT scan. Of the latter group, 11 individuals required immediate neurosurgical intervention such as implantation of an intraventricular catheter for drainage of cerebrospinal fluid or decompressive craniotomy.
The serum concentrations of S-100B are depicted in Figure 1 as a distribution curve according to the frequencies of S-100B concentrations (y axis) between 0 μg/L and 0.5 μg/L (x axis). Median values and interquartile ranges are shown in Table 3.
About 30% (specificity) of the MHI group identified as CCT− by radiological examination exhibited S-100B concentrations below the cutoff value of 0.10 μg/L (Fig. 1B). Hence, these patients could have been discharged safely without CCT scanning. Only one patient with an S-100B serum concentration below the cutoff concentration had a positive CCT scan, which revealed a small fissure (Fig. 1C).
In addition, as shown in the contingency table (Table 4), measurement of S-100B identified patients correctly as CCT+ with a sensitivity of 99% (95% CI, 96% - 100%) and a specificity of 30% (95% CI, 29% - 31%). Based on these values of sensitivity and specificity, the positive likelihood ratio was calculated at 1.4 and the negative likelihood ratio at 0.03.
The ROC curve (Fig. 2) depicts the sensitivity and 1−specificity values calculated for each individual serum S-100B concentration with respect to the radiological findings in the initial CCT scan. The quality of the discriminative potential of a new test system is expressed by the area under the curve given as a ratio between 0 (no discrimination) and 1 (complete discrimination). With an area under the curve value of 0.80 (95% CI, 0.75 - 0.84) in the ROC analysis, the assessment of S-100B in patients with MHI allowed the correct stratification of each patient into the groups CCT− or CCT+ at a significant level (P < 0.001).
In the MHI group, GCS scores of 13, 14, and 15 were assigned to 35, 122, and 1152 patients, respectively. Within the same 3 groups, the respective numbers of CCT+ patients were 6 (17%), 24 (20%), and 63 (5.5%). As depicted in Figure 3, the S-100B concentrations (median [interquartile range] in the group with a score of 13 were significantly different between CCT+ (1.3 μg/L [0.9 - 1.6 μg/L] and CCT− (0.14 μg/L [0.1 - 0.3 μg/L] patients (U test, P < 0.001). A similar significant difference was observed in the groups with GCS scores of 14 (CCT−, 0.16 μg/L [0.1 - 0.3 μg/L], vs CCT+, 0.9 μg/L [0.4 - 1.7 μg/L] and 15 (CCT−, 0.16 μg/L [0.1 - 0.3 μg/L], vs CCT+, 0.4 μg/L [0.2 - 0.9 μg/L]. Moreover, the S-100B concentrations in the 13-, 14-, and 15-point groups of patients without radiological findings (CCT−) were not significantly different from one another using ANOVA on ranks calculation; whereas the S-100B concentrations in the 15-point group of patients with positive scans (CCT+) were lower than those in the 13- or 14-point groups as determined using ANOVA on ranks.
Our results show that incorporation of S-100B concentration into the clinical decision rules for CCT in MHI patients could reduce the total number of negative CCT scans by as much as 30%.
So far, the indication for CCT scans in MHI patients has been based mainly on clinical risk factors such as neurological symptoms (amnesia, loss of consciousness, etc.) or coincidental intoxication mostly with alcohol. However, these clinical decision rules result in negative scans in approximately 90% of the cases, although it is well recognized that CCT scans impose a radiation-associated risk especially on individuals who do not benefit from this additional procedure and that unnecessary scans cause substantial financial and logistical burdens in emergency departments worldwide. On the other hand, a reliable, examiner-independent parameter for improved indication of CCT scans is missing to date. Hence, recent promising results of several groups about the diagnostic value of the brain damage marker S-100B, measured in the systemic circulation, raised the question whether the measurement of this astroglial protein in MHI patients might be able to improve the indication for CCT scans. Therefore, in this study, we tried to clarify this question using a sufficient number of patients to ensure statistical reliability of the outcome. The study design and the results are discussed in detail.
Study design and groups
Consideration of the study design requires a critical confrontation with the inclusion criteria and thereby with the definition of MHI in general. Unfortunately, the term „minor head injury“ has been used in the literature with various meanings (24, 25) In our investigation, we set up priorities on clinical criteria that represent the typical appearance of MHI patients in emergency departments according to recently published data (1, 8) in the context of enrolling patients with history of recent brain trauma: a GCS score of 13 to 15 in combination with at least one of the clinical variables (amnesia, loss of consciousness, medication with anticoagulant, etc).
In this context, several other authors have suggested considering head trauma patients with a GCS score of 13 as having „moderate“-not „minor“-injury, which would require a routine initial CCT scan (26, 27). However, because the aim of our study was to find out whether serum measurements of a brain-derived protein might provide valuable information in addition to the common neurological information used for the indication for initial CCT scan, we intended to include a preferable broad spectrum of patients to compare our results with previously published information from clinical studies on several thousands of patients with a GCS score of 13 to 15 (1, 8, 25). Therefore, we also included MHI patients with a GCS score of 13 points. Interestingly, the subgroup analysis (Fig. 3) revealed that the S-100B concentrations of CCT+ patients were highest in the 13-point subgroup and showed decreasing values in the 14- and 15-point subgroups, whereas the corresponding S-100B values of the CCT− patients were equally low in all 3 subgroups. Moreover, the 13- and 14-point subgroups contained similar percentages (17% and 20%, respectively) of CCT+ patients, whereas the corresponding value of the 15-point subgroup amounted to only 5.5%. Hence, the measurement of S-100B should provide the same valuable information for the indication of an initial CCT scan in patients presenting a GCS score of 13, 14, or 15.
Moreover, we enrolled only those patients presenting in our emergency department within 3 h of the traumatic event because it has been reported from animal and human studies that the half-life of the relatively small S-100B protein (21 kDa) is estimated to be between 25 and 120 min (16,28,29). Hence, we intended to exclude the potential situation that in an individual with a small intracranial lesion reaching the emergency department after a delay, the S-100B serum concentrations might have declined again to below the cutoff level. Nevertheless, further studies still have to answer the question of how long the maximum interval between the trauma and blood sampling might be without missing intracerebral lesions because of decreasing S-100B concentrations.
Furthermore, in our study, 7% of the MHI group presented trauma-relevant intracranial findings categorized as CCT+. This is in line with the 8% reported by Stiell et al (8) Haydel. et al. (1) reported approximately 6% CCT+ patients in their group; however, they included only patients presenting an initial GCS score of 15. Similarly, we found a frequency of 5.5% CCT+ patients in our subgroup with GCS scores of 15. In 1% of our group, the patients required immediate neurosurgical intervention, a finding that supports data published elsewhere (8, 25, 30, 31). Hence, our study group can be considered to be representative of patients with MHI.
Finally, it is noteworthy that the median age in the negative control group was significantly lower (8 years; Table 1) than that in the study group (GCS score, 13 - 15) or the positive control group (GCS score, 3 - 12). However, no age-dependent differences in S-100B concentrations of adults have been detected in any study published so far, and therefore, the age difference in our groups can be considered clinically not relevant for our results.
Diagnostic value of CCT
One major critical point of our investigation is the technique of confirming the diagnosis MHI-that is, the „gold standard“ relative to which the new technique has to prove its sensitivity and specificity. In our study, we used initial CCT scanning for that purpose because it is the preferred imaging technique in emergency departments (1), although it is clear that magnetic resonance imaging (MRI) is more sensitive in detecting small lesions (32). Yet, in the routine management of emergency departments, the application of MRI is still jeopardized by its limited availability, substantially longer scanning and analysis intervals, and higher costs as compared with CCT. Because the aim of our study was to compare a new laboratory test against the conventionally standard technique routinely used in emergency departments, CCT definitively represents the gold standard. Our study revealed a specificity of 30% for S-100B measurement to exclude definitely intracerebral lesions. This means that 70% showed S-100B serum concentrations above the cutoff at 0.10 μg/L (Fig. 1), although no radiological findings were detectable by CCT scans. This low specificity may be because a small lesion might be overlooked by CCT but might still induce a clear increase of systemic S-100B concentrations. Therefore, further studies on this special group (i.e., MHI patients with increased S-100B concentration and negative CCT scan) should be performed using MRI to detect even smaller lesions and thus potentially reevaluating CCT as a gold standard for the definitive diagnosis of brain damage after MHI. Moreover, another potential reason for an increase of S-100B without positive CCT scan might be peripheral injuries that have been elegantly demonstrated by Pelinka et al (33-35).
Diagnostic value of S-100B measurements
The serum concentrations of S-100B in our control groups (540 healthy volunteers; 55 patients with moderate to severe head injuries with GCS scores of 3 - 12) and in the study group (MHI; n = 1309) were similar to those of investigations in smaller groups of patients already reported by others (14, 18, 36) and by our research group (10, 17, 37). In addition, the S-100B serum concentrations of CCT+ and CCT− MHI patients support the findings in previous investigations (14, 15, 17, 18).
The same held true for the ascertainment of a cutoff level and the results of the ROC analysis (38). However, the considerable new information from the present study is that at a serum S-100B cutoff level of 0.10 μg/L, intracerebral lesions are ruled out at a sensitivity level of 99% and a negative predictive level of 100%, respectively. Because of the large number of patients studied, these results are achieved at 95% CIs of 96% to 100% for the sensitivity level and of 99% to 100% for the negative predictive value, respectively. Interestingly, the identification of our cutoff level at 0.10 μg/L by taking the 95th percentile of the data of 540 normal healthy controls is strengthened by other researchers, who found exactly the same cutoff level at 0.10 μg/L, although they used another device for measurement (39).
In contrast, Ingebrigtsen et al. (14) published a Scandinavian multicenter study on 182 patients after MHI presenting with very similar inclusion criteria but using a test that detects S-100B in serum only above a detection level of 0.2 μg/L. They thereby identified CCT+ patients (10 of 182) at a sensitivity level of 90% and a specificity level of 65%. Although this study was not performed on a large enough number of patients to warrant a change in the clinical decision rules, the results are roughly in line with our findings. However, because the technology we used allows measurement of S-100B at a lower concentration, we were able to further differentiate in the clinically relevant sector of serum concentration below 0.2 μg/L. By reduction of the cutoff level to 0.10 μg/L, the sensitivity level is increased to 99%, and the substantially higher number of included patients gives a 95% CI of 96% to 100%.
Nevertheless, one person in our study exhibited an S-100B concentration below 0.10 μg/L but had a positive CCT. The detailed analysis of the CCT finding revealed that a small fissure was considered a fracture of the left frontal skull without any signs of brain cell damage. Because S-100B is a marker of damaged brain cells, it is not surprising that no increase in S-100B serum concentration became apparent in this person. The patient was a 58-year-old male individual. The interval between trauma and blood sampling was 35 min, and he complained of amnesia as a clinical symptom. His GCS was 15 points.
It is also noteworthy that none of the CCT+ patients with an S-100B serum concentration below 0.5 μg/L required surgical intervention. This is in line with data from several other investigators who compared the life quality of brain trauma patients with S-100B serum concentrations (40) or signs of secondary neurological complications in brain trauma patients and thereby described 0.5 μg/L as a cutoff level for severe complications (41). However, for incorporation into clinical decision rules for MHI patients, the cutoff level at 0.5 μg/L seems too high (39).
As previously mentioned, the detailed analysis of S-100B serum concentrations in subgroups of initial GCS scores (Fig. 3) revealed a significant difference between the elevated S-100B serum concentrations in CCT+ patients with a score of 15 and those in the subgroups with scores of 14 and 13, whereas the CCT− patients had similar low S-100B concentrations. Moreover, in each subgroup, the CCT+ patients had significantly higher serum concentrations than those without lesions. From these data, it can be concluded with a very high level of statistical reliability that measurement of S-100B might be a safe and effective criterion for the decision to perform an initial CCT scan in patients with GCS scores of 13 to 15. We suggest that addition of S-100B concentrations to the clinical decision rules might be able to reduce the amount of CCT scans by at least 30% for the group of patients with GCS scores of 13 to 15 upon admission.
With regard to the diagnostic value of S-100B measurement in serum, we note that Anderson et al. (39) published a study on patients with multiple trauma (n = 17) without head injuries, finding increased serum concentrations when there were accompanying bone fractures (2 - 10 μg/L) or thoracic contusions (0.5 - 4 μg/L). Those authors concluded that, in case of multiple trauma, S-100B might be released from tissues other than those of the brain. This assumption is supported by Pelinka et al. (15, 33-35) and other authors who investigated systemic concentrations of S-100B in multiple trauma patients with and without cranial injury (42). Thus, it might be thought that the release of S-100B from other sources would result in reduced specificity of the test system with regard to intracerebral lesions. However, because of the high negative predictive level in our study, acute intracerebral lesions are ruled out if the systemic concentration of S-100B in an individual falls below the cutoff level.
Finally, 45% of the patients in the MHI group in our study were intoxicated with alcohol. A similar finding has been reported in previous studies about coincidence of brain trauma and alcohol intoxication, which described a frequency of approximately 50% (9). Measurement of an examiner-independent objective parameter in systemic circulation of intoxicated MHI patients is a major point because-especially in these patients-neurological symptoms (amnesia, vertigo, nausea, etc.) are most frequently induced by the concomitant intoxication rather than intracerebral lesions. Because alcohol intoxication does not at all rule out these life-threatening complications, the present clinical decision rules advise emergency physicians to perform CCT scans on all intoxicated patients (1). We have shown in previous studies that alcohol intoxication does not affect S-100B serum concentrations or the ability to discriminate between CCT+ and CCT− patients (10, 37). Hence, measurement of S-100B serum concentrations seems an ideal technology to distinguish between neurological deficits caused by intoxication and those caused by intracerebral lesions in the intoxicated patient.
The results from our multicenter study undoubtedly show at a high level of statistical reliability that measurement of S-100B serum concentrations provides substantial information for the management of MHI patients and, in addition to conventional clinical decision rules, might allow the reduction of CCT scans by approximately 30% of cases. However, this study also shows that no single laboratory value alone can be taken as a basis for any clinical decision rule. If the individual shows significant neurological deterioration, especially in the absence of chemical intoxication, a CCT scan might be indicated even if the S-100B value is below the cutoff level.
The invaluable technical help of Christopher Hauser, Ruza Hell, Maria Maier, Christina Melzer, Tobias Vogel, Birgit Wehnl, Martin Oremek, and Herbert Sauer is gratefully appreciated.
1. Haydel MJ, Preston CA, Mills TJ, Luber S, Blaudeau E, DeBlieux PM: Indications for computed tomography in patients with minor head injury [see comments]. N Engl J Med
2. Nagy KK, Joseph KT, Krosner SM, Roberts RR, Leslie CL, Dufty K, et al: The utility of head computed tomography after minimal head injury. J Trauma
3. Jeret JS: Clinical predictors of abnormality disclosed by computed tomography after mild head trauma. Neurosurgery
4. Borczuk P: Predictors of intracranial injury in patients with mild head trauma. Ann Emerg Med
5. Stein SC: Indications for computed tomography after minor head injury. N Engl J Med
6. Klauber MR, Marshall LF, Luerssen TG, Frankowski R, Tabaddor K, Eisenberg HM: Determinants of head injury mortality: importance of the low risk patient [see comments]. Neurosurgery
7. Graham ID, Stiell IG, Laupacis A, O'Connor AM, Wells GA: Emergency physicians' attitudes toward and use of clinical decision rules
for radiography. Acad Emerg Med
8. Stiell IG, Wells GA, Vandemheen K, Clement C, Lesiuk H, Laupacis A, et al: The Canadian CT head rule for patients with minor head injury. Lancet
9. Kelly DF: Alcohol and head injury: an issue revisited. J Neurotrauma
10. Biberthaler P, Mussack T, Wiedemann E, Gilg T, Soyka M, Koller G, et al: Elevated serum levels of S-100B reflect the extent of brain injury in alcohol intoxicated patients after mild head trauma. Shock
11. Woertgen C, Rothoerl RD, Metz C, Brawanski A: Comparison of clinical, radiologic, and serum marker as prognostic factors after severe head injury. J Trauma
12. Raabe A, Grolms C, Sorge O, Zimmermann M, Seifert V: Serum S-100B protein in severe head injury. Neurosurgery
13. Pleines UE, Morganti-Kossmann MC, Rancan M, Joller H, Trentz O, Kossmann T: S-100beta reflects the extent of injury and outcome, whereas neuronal specific enolase is a better indicator of neuroinflammation in patients with severe traumatic brain injury. J Neurotrauma
14. Ingebrigtsen T, Romner B, Marup-Jensen S, Dons M, Lundqvist C, Bellner J, et al: The clinical value of serum S-100 protein measurements in minor head injury: a Scandinavian multicentre study. Brain Inj
15. Pelinka LE, Toegel E, Mauritz W, Redl H: Serum S 100 B: a marker of brain damage in traumatic brain injury with and without multiple trauma. Shock
16. Ingebrigtsen T, Romner B: Serial S-100 protein serum measurements related to early magnetic resonance imaging after minor head injury. J Neurosurg
85:945-948, 1996[Case report].
17. Biberthaler P, Mussack T, Wiedemann E, Kanz KG, Koelsch M, Gippner-Steppert C: Evaluation of S-100b as a specific marker for neuronal damage due to minor head trauma. World J Surg
18. Rothoerl RD, Woertgen C, Holzschuh M, Metz C, Brawanski A: S-100 serum levels after minor and major head injury. J Trauma
19. Ingebrigtsen T, Romner B: Biochemical serum markers for brain damage: a short review with emphasis on clinical utility in mild head injury. Restor Neurol Neurosci
20. Zimmer DB, Cornwall EH, Landar A, Song W: The S100 protein family: history, function, and expression. Brain Res Bull
21. Donato R: S100: a multigenic family of calcium-modulated proteins of the EF-hand type with intracellular and extracellular functional roles. Int J Biochem Cell Biol
22. Marenholz I, Heizmann CW, Fritz G: S100 proteins in mouse and man: from evolution to function and pathology (including an update of the nomenclature). Biochem Biophys Res Commun
23. Linsenmaier U, Brandl R, Akkerman B, Kanz KG, Rock C, Papst E, et al: CCT in minor head trauma-comparison of two low dose spiral and sequential scanning protocols in vitro and in vivo. Emerg Radiol
24. Servadei F, Teasdale G, Merry G: Defining acute mild head injury in adults: a proposal based on prognostic factors, diagnosis, and management. J Neurotrauma
25. Jagoda AS, Cantrill SV, Wears RL, Valadka A, Gallagher EJ, Gottesfeld SH, et al: Clinical policy: neuroimaging and decisionmaking in adult mild traumatic brain injury in the acute setting. Ann Emerg Med
26. American College of Surgeons: Head trauma. In ATLS Advanced Trauma Life Support
, 6th ed. Chicago, IL: American College of Surgeons, pp 181-213, 2004.
27. Stein SC: Minor head injury: 13 is an unlucky number. J Trauma
28. Woertgen C, Rothoerl RD, Brawanski A: Early S-100B serum level correlates to quality of life in patients after severe head injury. Brain Inj
29. Jonsson H, Johnsson P, Hoglund P, Alling C, Blomquist S: Elimination of S100B and renal function after cardiac surgery. J Cardiothorac Vasc Anesth
30. Zee CS, Go JL: CT of head trauma. Neuroimaging Clin N Am
31. Shackford SR: The clinical utility of computed tomographic scanning and neurologic examination in the management of patients with minor head injuries. J Trauma
32. Doezema D, King JN, Tandberg D, Espinosa MC, Orrison WW: Magnetic resonance imaging in minor head injury. Ann Emerg Med
33. Pelinka LE, Bahrami S, Szalay L, Umar F, Redl H: Hemorrhagic shock induces an S 100 B increase associated with shock severity. Shock
34. Pelinka LE, Toegel E, Mauritz W, Redl H: Serum S 100 B: a marker of brain damage in traumatic brain injury with and without multiple trauma. Shock
35. Pelinka LE, Harada N, Szalay L, Jafarmadar M, Redl H, Bahrami S: Release of S100B differs during ischemia and reperfusion of the liver, the gut, and the kidney in rats. Shock
36. Ingebrigtsen T, Romner B, Kongstad P, Langbakk B: Increased serum concentrations of protein S-100 after minor head injury: a biochemical serum marker with prognostic value? J Neurol Neurosurg Psychiatry
37. Mussack T, Biberthaler P, Kanz KG, Heckl U, Gruber R, Linsenmaier U, et al: Immediate S-100B and neuron-specific enolase plasma measurements for rapid evaluation of primary brain damage in alcohol-intoxicated, minor head-injured patients. Shock
38. Biberthaler P, Mussack T, Wiedemann E, Kanz KG, Mutschler W, Linsenmaier U, et al: Rapid identification of high-risk patients after minor head trauma (MHT) by assessment of S-100B: ascertainment of a cut-off level. Eur J Med Res
39. Anderson RE, Hansson LO, Nilsson O, Dijlai-Merzoug R, Settergren G: High serum S100B levels for trauma patients without head injuries. Neurosurgery
40. Rosengren D, Rothwell S, Brown AF, Chu K: The application of North American CT scan criteria to an Australian population with minor head injury. Emerg Med Australas
41. Raabe A, Kopetsch O, Woszczyk A, Lang J, Gerlach R, Zimmermann M, et al: Serum S-100B protein as a molecular marker in severe traumatic brain injury. Restor Neurol Neurosci
42. Savola O, Pyhtinen J, Leino TK, Siitonen S, Niemela O, Hillbom M: Effects of head and extracranial injuries on serum protein S100B levels in trauma patients. J Trauma