Monitoring the cerebral status of patients with severe traumatic brain injury (TBI)
S100 proteins are a family of small, dimeric multigenic calcium-binding proteins comprising various combinations of α and β subunits. S100 proteins most commonly occur as S100A (α – α) or S100B (α – β [S100A1B] and β – β [S100BB]) subtypes.4 S100B is predominately confined to glial and Schwann cells and is the most well-studied subtype in traumatic brain injury (TBI). Both S100A1B and S100BB have been implicated in severe TBI. 5
Severe TBI invariably results in neuronal destruction and destabilization of the blood-brain barrier (BBB). These phenomena are accompanied by a release of S100B protein into the blood. S100B is measurable within minutes of a TBI, and can be detected for an extended period in the bloodstream. S100B is removed from the serum by the renal clearance pathway, with a half-life of 20 to 25 minutes. 4
In the USA alone, an estimated 1.7 million people sustain a TBI each year. Of these, 275,00 are hospitalized (16.3%), 1,365,000 are treated and released from emergency departments (80.7%) and 53,000 (3%) die.1
Patients with a severe TBI are at high risk of post-traumatic events, such as intracranial bleeding or brain edema which lead to a rise in intracranial pressure and secondary brain damage. In such cases rapid intervention is essential to relieve the intracranial pressure through medication and/or surgery (craniotomy).
Patients with severe TBI are typically treated under analgo-sedation in an intensive care unit. Although sedation provides an optimal environment for physical recovery, it limits the clinician’s options to examine the patient’s cognitive state. It is therefore common practice to monitor the stability of the patient’s neurological state using computer tomography (CT) imaging. While CT serves as a gold standard to monitor structural changes of TBI patients, two limitations restrict its usage to daily intervals or longer:
*Measure the S100 value within 3 hours after trauma. A negative result from a sample collected more than 3 hours after trauma should not be used.
S100B protein is a marker that displays high clinical sensitivity for severe TBI, and the extent of S100B elevation has been found to be useful in predicting clinical outcome after brain injury.7,8
Serial daily measurements of serum S100B were found to be a useful non-invasive means for identifying brain damage and could be used for prediction of mortality.7,9 Furthermore, some studies have reported that S100B levels aboce certain thresholds might have predictive value on trauma-induced brain death.4,9,10
Secondary neurological events are accompanied by an elevation in the serum levels of S100B, often visible earlier than when detected with diagnostic imaging.4,15,16 S100B levels have been shown to rise hours to days before changes in intracranial pressure or onset of cerebral hypoxia.4,17 S100B levels may be used to monitor comatose intensive care patients for neurological complications such as a new infarction, new hemorrhage, or a newly developed progressive disease.16
Serial measurements using the S100 assay may therefore help a clinician to recognize the onset of a neurological complication and enable early therapeutic intervention.16 The same study has shown that this information affected patient management in 21% of severe TBI cases.16
S100B concentrations in serum have been shown to be representative of the extent of primary brain damage, corroborated by clinical scales of neurological status,11 subsequent CT examination,10-13 and neurological outcome.14
For example, in 60 patients with S100B measured within 24 hours of trauma, levels correlated with neurological outcome as assessed by the Glasgow Outcome Scale (GOS).14 Another study of 102 adult patients with severe TBI demonstrated that initial serum S100B concentrations correlated with the severity of brain injury on CT imaging as determined by the Marshall classification.10
S100B is related to outcome prognosis.7,9,18 Serial measurement of S100B accurately predicts short-term mortality, with the strongest correlation with clinical outcome seen >84 hours after trauma.18 Comparison of time courses of S100B levels in patients with favorable and unfavorable outcomes also indicate that S100B values by day 2 after admission independently predict 12-month mortality.7
In a meta-analysis of 41 studies, serum S100B concentrations measured after moderate or severe TBI were significantly associated with prognosis in the short (<3 months), medium (3–6 months) or long term (6 months), as defined by mortality or a Glasgow Outcome Score ≤3. Furthermore, this association was unaffected by concomitant traumatic injuries.19
A CT scan has high sensitivity for detecting intracranial injuries in patients with head injury.20 However, the technique is costly, exposes the patient to high doses of radiation,21 and clinically relevant lesions are found in less than 10% og cases of minor head injury.2 Several studies have demonstrated that a normal S100B level reliably predicts normal CT findings after minor head injury in adults.22
In a prospective multicenter study of 1,309 patients with minor head injury, an S100B cut-off of 0.10 µg/L (the 95th percentile of healthy volunteers) identified those patients with trauma-relevant CT findings with a sensitivity of 99% and a negative predictive value of 99.68%.3
A meta-analysis of 12 studies of adults with minor head injury reported a pooled sensitivity for S100B for the prediction of CT findings of 97% (95% CI, 91-99%) and pooled specificity of 40% (95% CI, 30-51%).22 This equated to a negative predictive value of >99% (95% CI, 98-100%) at an average prevalence for intracranial findings after minor head injury of 8%.22 Omitting CT in adults with minor head injury and an S100B concentration of <0.10 μg/L would reduce the number of CTs by approximately one third.22
International guidelines advise that adult patients with mild head injury and no risk factors who have a serum S100B level <0.10 µg/L measured within 6 hours of injury may be discharged without the need for CT.23
Diagnosis of TBI in children with head injury can be more challenging than in adults, as obtaining a reliable patient history may be difficult and physical examination can be uncomfortable for the child.
Similar to studies in adult populations, S100B levels measured following TBI in pediatric populations predict CT findings24, 25 and outcomes.26, 27 In a study of 446 children who presented within 3 hours of mild TBI, S100B levels correlated with severity of TBI and predicted a poor clinical evolution with 100% sensitivity (95% CI, 84-100%).24 Furthermore, S100B levels had a 100% negative predictive value for ruling out trauma-relevant intracerebral lesions on CT. Hence, S100B determination during the first 3 hours after TBI can potentially reduce the number of CT scans in children.
|S100B||CCT +||CCT -|
||92||855||PPV 10% (95% CI, 7-13%)
||1||361||NPV 99.68% (95% CI, 99-100%)
|Sensitivity: 99% (95% CI, 96-100%)
||Specificity: 30% (95% CI, 29-30%)