• 
• 

Three Part Question

In [patients with a head injury] do [levels of S-100B protein] predict [long-term disability]?

Clinical Scenario

A 17 year old male presents to the Emergency Department after a road traffic accident. His GCS was 8 on arrival but an immediate CT scan showed no focal abnormality. His GCS returned to 14 after 4 hours. You are talking to his mother who is reassured that he does not need urgent neurosurgery, but she asks whether he will suffer any long term consequences from this injury. You tell her that it is difficult to predict, but you have recently head that S-100 protein measurement is available in your hospital for research purposes. You wonder whether S-100 could help predict his long term prognosis.

Search Strategy

Medline 1966-Week 4 August 2005 using the OVID interface Embase 1980-2005 week 37
The Cochrane Library Issue 3 2005
Medline:[(exp S100 Proteins/ OR s100.mp OR s-100.mp) AND (exp Brain Injuries/ OR brain injury.mp OR exp Craniocerebral trauma/ OR head inj$.mp)]
Embase:[exp Protein S 100/ OR s100.mp OR s-100.mp] AND [exp Brain Injury/ OR brain injury.mp. OR craniocerebral trauma.mp. or exp Head Injury/] LIMIT to Human and English Language
The Cochrane Library: Exp Brain injuries [MeSH] OR exp Craniocerebral trauma [MeSH] AND exp S100 proteins [MeSH]

Search Outcome

200 papers were found of which 13 were found to be relevant. Two relevant papers described the same patient population. The remaining 12 papers are shown in the table

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Ingebrigtsen et al, 1999, Sweden	50 patients with minor head injury and LOC (GCS 13-15) referred to Neurosurgery dept after CT scan S-100 taken hourly up to 12 hours	Diagnostic Cohort study (3b)	Neuropsychological testing at 3 months (for attention, psychomotor speed, trail-making test, memory, digit span) In 36 patients	11/36 patients had S-100 >0.2mcg/l There were non significant trends to reduced impairment in the S-100 negative group	Very small study with no sample size estimates Non consecutive Only 36 of 50 patients followed up at 3 months
			MRI and CT scan findings within 48hrs	4 of 5 patients with brain contusion had S-100 >0.4mcg/l Sensitivity 80% (p=0.035)
Mussack T et al, 2000, Germany	80 patients presenting with a history of minor head trauma (GCS 13-15) Also 10pts with severe head injury (GCS<8) S-100 taken at 0h, 6h and 24hrs post admission 50 patients GCS 13-15 after normal CT scan	Diagnostic study (4)	S-100 in Minor Head Trauma pts	Patients discharged <=6hrs 0.29 +/- 0.11 ng/ml Patients discharged >= 24hrs 0.70 +/- 0.19 ng/ml Patients subsequently admitted to ICU 5.03 +/- 3.18 ng/ml	No gold standard outcome measures Non consecutive Results not clearly presented Non significant findings between groups Low number of patients
			Patients with Severe head Injury GCS<8	5.26 +/- 1.56ng/ml
Rothoerl et al, 1998, Germany	30 patients with a severe head injury (GCS<=9) and 11 with minor head injury (GCS 13-15) admitted to a neurosurgical unit S-100 levels measured mean 2.5 hrs after injury	Diagnostic Cohort study (4)	Glasgow Outcome Scale on discharge (Mean day 19 in severe group and mean day 1.3 in minor head injury group)	Patients with GOS 3-5 S-100 level mean 1.2mcg SD 1.8 Patients with GOS 1-2 (unfavourable) S-100 level mean 4.9mcg/l SD 5.3 P=0.0025	Non-independent gold standard Small, selected cohort of patients
			Detectable level of S-100 (>0.5mcg/l)	25 of 27 Elevated S-100 levels were found in the minor head injury group
Townend WJ et al, 2002, UK	148 adult head injury patients (GCS 4-15) in 4 hospitals. Most had a minor head injury S-100 levels taken within 6 hours of head injury	Diagnostic study (2b)	Extended Glasgow outcome score at 1 month	S-100>0.32mcg/l predicted severe disability (15 patients with GOSE<5): Sensitivity 93% (63%-100%) Specificity 72% (54%- 79%) NPV 99% (93%-100%)	Wide confidence intervals Non consecutive Wide definition of head injury (including no LOC) 80% follow up rate
Herrmann et al, 2001, Germany	69 patients admitted to a neurosurgical unit (mostly GCS >13) S-100 taken at 1, 2 and 3 days	Diagnostic study (3b)	Intracranial pathology on CT scan at 2 weeks and 6 months, or focal neurology	At 2 weeks, S-100 of >0.14mcg/l predicted positive outcome: Sensitivity 69% Specificity 90% At 6 month, S-100 of >0.14mcg/l predicted positive outcome: Sensitivity 65% Specificity 89%	Inclusion criteria for patients unclear Only 29 patients followed up to 6 months
Spinella et al, 2003, USA	27 children (<18yrs) with traumatic brain injury S-100 taken within 12 hours	Diagnostic cohort study (3b)	Pediatric Cerebral performance category score (PCPC) assessed at discharge and 6 months	For s-100 level of >2.0mcg/l, unfavourable outcome was predicted with Sensitivity 86% Specificity 95%	Very small study Confidence intervals not given Non consecutive
Savola O & Hillbom M, 2003, Finland	172 consecutive patients with mild head injury (GCS 13-15)	Diagnostic cohort study (2b)	Post concussional symptoms defined by Rivermead Post-Concussion Symptoms Questionnaire at 2-6 weeks	For s-100 level of >0.50mcg/l, PCS symptoms predicted with Sensitivity 27% Specificity 93%	No confidence intervals or sample size calculations
Chatfield DA et al, 2002, UK	20 patients with severe head injury (GCS<=8) admitted to neurosurgical unit s-100 on admission	Diagnostic cohort study (4)	Glasgow outcome score at 6 months after trauma (GOS 1-3 unfavourable)	Patients with GOS 1-3 S-100 mean level 2.46 +/-0.32mcg/l Patients with GOS 3-5 S-100 mean level 0.6 +/-0.1mcg P<0.05	Data not clearly presented Small study No cut off points or ROC curves calculated
Waterloo, k et al, 1997, Norway	7 patients with high S-100b after mild head injury matched with 7 patients with no detectable S-100b	Case control study	Overall cognitive function	No difference
			Reaction time	Increased in raised S-100b group
			Attention	Reduced in raised S-100b group
Raabe A et al, 1995, Germany	82 patients after severe head injury (GCS< = 8) s-100 taken at admission and every 24 hours	Diagnostic cohort study (2b)	Glasgow outcome score at 6 months Unfavourable outcome defined as severe disability or vegetative state	For S-100 level of >2.5mcg/l, unfavourable outcome was predicted with Sensitivity 44% Specificity 97%	No confidence intervals presented Non consecutive
Ingebrigsten et al, 2000, Scandinavia (3 centres Sweden, Denmark, Norway)	182 patients from 3 centres with GCS 13-15 and brief Loss of Consciousness. S-100 taken on admission	Diagnostic Cohort Study (2b)	Rivermead postconcussion symptoms questionnaire score (RPQ)	Patients with a positive S-100 had mean RPQ 6.0 vs 4.0 in S-100 negative group p = 0.07	No sensitivities or specificities given for prediction of long term disability
			Intracranial Pathology on CT scan at <24 hours	Detectable S-100 predicted intracranial pathology with: Sensitivity 90%, Specificity 65%
Woertgen et al, 1999, Germany	44 patients after severe head injury (GCS score < = 8) S-100 taken 1-6 hrs after injury	Diagnostic cohort study (3b)	Glasgow outcome score calculated at mean 11 months after trauma (GOS 1-3 unfavourable)	For S-100 level of >2mcg/l, PCS symptoms predicted with Sensitivity 95% Specificity 70%	Tables 2, 3 and 4 are incorrect, with erratum printed in a later edition

Comment(s)

All studies were under 200 patients in size and most were under 100 patients. The studies find sensitivities from 27%–95% and specificities from 70% to 97%. The reasons for this great variation in findings may in large part be due to the small sample sizes. The specificities seem to perform better than the sensitivities and thus the finding of a high S-100 may indicate that your patient is at high risk of long term disability. The cut-points for a significant S-100 level differ between studies also and are generally much higher when applied to patients after a severe head injury. Most studies agree that S-100 levels must be taken within 6 hours of head injury.

Clinical Bottom Line

A high S-100 level is a marker of poorer long term outcome following minor and major head injury.

References

1. Ingebrigtsen T, Waterloo K, Jacobsen EA, et al. Traumatic brain damage in minor head injury: relation of serum S-100 protein measurements to magnetic resonance imaging and neurobehavioral outcome. Neurosurgery 1999;45:468-75.
2. Mussack T, Biberthaler P, Wiedemann E, et al. S-100b as a screening marker of the severity of minor head trauma (MHT)--a pilot study. Acta Neurochirurgica - Supplement 2000;76:393-6.
3. Rothoerl RD, Woertgen C, Holzschuh M, et al. S-100 serum levels after minor and major head injury. Journal of Trauma-Injury Infection & Critical Care 1998;45:765-7.
4. Townend WJ, Guy MJ, Pani MA, et al. Head injury outcome prediction in the emergency department: a role for protein S-100B? Journal of Neurology, Neurosurgery & Psychiatry 2002;73:542-6.
5. Herrmann M, Curio N, Jost S, et al. Release of biochemical markers of damage to neuronal and glial brain tissue is associated with short and long term neuropsychological outcome after traumatic brain injury. Journal of Neurology, Neurosurgery & Psychiatry 2001;70:95-100.
6. Spinella PC, Dominguez T, Drott HR, et al. S-100beta protein-serum levels in healthy children and its association with outcome in pediatric traumatic brain injury. Critical Care Medicine 2003;31:939-45.
7. Savola O, Hillbom M. Early predictors of post-concussion symptoms in patients with mild head injury. European Journal of Neurology 2003;10:175-81.
8. Chatfield DA, Zemlan FP, Day DJ, et al. Discordant temporal patterns of S100beta and cleaved tau protein elevation after head injury: a pilot study. British Journal of Neurosurgery 2002;16:471-6.
9. Waterloo K, Ingebrigtsen T, Romner B. et al. Neuropsychological function in patients with increased serum levels of protein S-100 after minor head injury. Acta Neurochirug 1997;139:26–32.
10. Raabe A, Grolms C, Seifert V. Serum markers of brain damage and outcome prediction in patients after severe head injury. British Journal of Neurosurgery 1999;13:56–9.
11. Ingebrigtsen T, Romner B, Marup-Jensen S. et al. The clinical value of serum S-100 protein measurements in minor head injury: a Scandinavian multicentre study. Brain Injury 2000;14:1047–55.
12. Woertgen C, Rothoerl RD, Metz C. et al. Comparison of clinical, radiologic, and serum marker as prognostic factors after severe head injury. Journal of Trauma-Injury Infection & Critical Care 1999;47:1126–30.