• 
• 

Three Part Question

In [children presenting with head injury] does [a boggy, soft or large scalp haematoma] predict [a positive CT scan for significant cranial injury]?

Clinical Scenario

A seventeen-month-old boy attends the Emergency Department with his mother following a head injury after tripping over at home. He has an obvious large and "boggy" scalp haematoma. He appears very well and has no clinical signs to suggest intracranial injury.
You are unsure if a CT scan is needed and would like to know how much emphasis you should put on this one clinical sign.

Search Strategy

Medline and Embase databases were searched using the interface provided by NHS Evidence the week ending 15th September 2012. The references of relevant major research papers were cross referenced to ensure all relevant articles were found.
{(infant* OR baby OR babies OR newborn* OR pediatric* OR paediatric* OR child* OR teen* OR neonat* OR adolescen* OR toddler*).ti,ab;} AND {(boggy ADJ2 hematoma*).ti,ab; OR (boggy ADJ2 haematoma*).ti,ab; OR (soft* ADJ2 haematoma*).ti,ab; OR (soft* ADJ2 hematoma*).ti,ab; OR (soft* ADJ2 swell*).ti,ab; OR (boggy* ADJ2 swell*).ti,ab; OR boggy*.ti,ab; OR (scalp* ADJ2 haematoma*).ti,ab; OR (scalp* ADJ2 hematoma*).ti,ab; OR (head* ADJ2 haematoma*).ti,ab; OR (head* ADJ2 hematoma*).ti,ab;} AND {(brain* ADJ2 injur*).ti,ab OR (head* ADJ2 injur*).ti,ab; OR (cerebral ADJ2 injur*).ti,ab; OR (brain* ADJ2 traum*).ti,ab; OR (head* ADJ2 traum*).ti,ab; OR (cerebral ADJ2 traum*).ti,ab; OR exp HEMATOMA/; OR exp CRANIOCEREBRAL TRAUMA/; OR exp BRAIN INJURIES/;}

Search Outcome

This resulted in 31 results from Medline and 53 from Embase. Duplicates and articles clearly not relevant were removed, leaving 11 articles for full text evaluation. Of these papers, one directly addressed the clinical question using regression analysis and a second paper was published testing the decision rule produced. Six articles considered the question only as part of a clinical decision rule (CDR); however, three of these provided secondary outcome data, shown in the table. One of the CDRs (CHALICE study) considered scalp haematoma along with laceration and bruise and was excluded for this reason. Three papers attempted to validate previous CDRs and provided no relevant data.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Quayle et al. 1997 USA	321 Children under 18 years with non-trivial head injury. Scalp haematoma was considered a significant finding in those less than 2 years old. All children received CT	A descriptive 6-month prospective cohort study	Presence of ICI by skull radiograph or CT head	Univariate analysis indicates a non-significant OR of 1.82 (95% CI 0.82 to 4.02) for the clinical variable of scalp haematoma	No clear analysis of the subgroup of fully asymptomatic group. The significance of isolated scalp haematoma is therefore unclear. The size and nature of the haematomas were not discriminated further
Greenes DS, Schutzman SA. 1999 USA	608 Children aged less than 2 years with head trauma presenting to a tertiary centre paediatric ED	A descriptive 1-year prospective consecutive sample cohort study	Skull fracture or ICI (cerebral contusion, cerebral oedema, or intracranial haematoma) on skull radiograph or head CT	Significant scalp haematomas were noted in 77% of subjects with ICI. A significant scalp haematoma had an OR of 4.65 (95% CI 2.00 to 10.79) for ICI	Similar to Greenes and Schutzman (2001)
Greenes DS, Schutzman SA. 2001 USA	422 Asymptomatic children (absence of symptoms and signs of brain injury) aged less than 2 years with head trauma presenting to a tertiary centre paediatric ED. 172 Subjects received some form of head imaging (either skull radiograph and/or CT head)	A descriptive 1-year prospective consecutive sample cohort study. Greenes and Schutzman (1999) using decision rule for CT; regression analysis to determine independent risk factors	Skull fracture or ICI (cerebral contusion, cerebral oedema, or intracranial haematoma) on skull radiograph or head CT	Large scalp haematomas were associated with higher risk of skull fracture compared to no haematoma (OR 27.0 (95% CI 8.0 to 90.5). Of those with large haematomas 7/17 (41%, p=0.008) had ICI compared to 1/20 (5%) with no haematoma. Telephone follow-up of those not imaged revealed no unwell cases	Decision to CT was left to individual clinicians but they were encouraged to follow the following guidelines: (1) skull radiograph initially if large haematoma present; (2) head CT if skull fracture present on radiograph; (3) skull radiograph encouraged in the <1 year group with any non-trivial injury with or without haematoma. Clinician judgement was used to define ‘large’ haematoma, leading to inevitable variation
Palchak et al. 2003 USA	2043 Children with non-trivial head trauma aged less than 18 years. All those who had a CT head (n=1271) were analysed. Scalp haematoma only assessed in children less than 2 years old and receiving CT head imaging (n=194)	A descriptive 3-year prospective cohort study	Presence of TBI excluding isolated skull fracture	The presence of scalp haematoma and <2 years old confers a 2.6 RR (95% CI 1.5 to 4.3) of TBI. 14/77 (18.2%) of children <2 years with scalp haematoma had a positive CT head for TBI compared to (1/117) (0.9%) without	Data were presented as part of a CDR. The size and nature of the haematomas were not discriminated further
Kuppermann et al, 2009, (the PECARN study)	42 412 Children aged less than 18 years presenting to 25 North American ED with GCS 14–15	A prospective cohort study over 28 months	Presence of clinically important TBI (death, neurosurgery, intubation for >24 h, hospital admission for ≥2 nights associated with TBI on CT)	18/1126 (1.6%) of patients who had no altered mental status but had an occipital, parietal or temporal haematoma had a clinically important TBI. Presence of scalp haematoma was the 2nd predictor of the decision tree following CART analysis	Data were presented as part of a CDR. It is not possible to determine if scalp haematoma would be predictive for clinically important TBI as an independent variable

Comment(s)

Large CDRs such as PECARN and CHALICE are increasingly used to aid the diagnostic process of this group of emergency patients. The data regarding the implication of one clinical sign is difficult to disentangle from all factors in the CDR. The one study to answer the clinical question directly provides reasonably strong evidence supporting the finding of positive head imaging in the otherwise asymptomatic individual. The evidence within the derivation of other CDRs would appear to support proactive imaging strongly. Ideally, future studies of CDRs should be designed to answer the specific question, given current concerns about irradiation risk.

Editor Comment

CART, classification and regression trees; CDR, clinical decision rule; ED, emergency department; GCS, Glasgow coma scale; ICI, intracranial injury; TBI, traumatic brain injury.

Clinical Bottom Line

The presence of a large or soft (boggy) scalp haematoma appears to be associated with positive CT findings and, in an otherwise asymptomatic young child, was concluded to warrant CT imaging in the one study that addressed the three-part question. Since that time, however, concern regarding the risks associated with irradiation of children may challenge a strategy of imaging otherwise asymptomatic children with no other indication for CT imaging.

References

1. Quayle KS, Jaffe DM, Kuppermann N, et al. Diagnostic testing for acute head injury in children: When are head computed tomography and skull radiographs indicated? Pediatrics 1997;99:e1-8
2. Greenes DS, Schutzman SA. Clinical indicators of intracranial injury in head-injured infants. Pediatrics 1999;104:861-867
3. Greenes DS, Schutzman SA. Clinical significance of scalp abnormalities in asymptomatic head-injured infants. Pediatric Emergency Care 2001;17:88-92
4. Palchak MJ, Holmes JF, Vance CW, et al. A decision rule for identifying children at low risk for brain injuries after blunt head trauma. Annals of Emergency Medicine 2003;42:492-506
5. Kuppermann N, Holmes JF, Dayan PS, et al. Identification of children at very low risk of clinically important brain injuries after head trauma: a prospective cohort study (PECARN). Lancet 2009;374:1160–70.
6. Dunning J, Patrick Daly J, Lomas J-P, et al. Derivation of the children's head injury algorithm for the prediction of important clinical events decision rule for head injury in children. Arch Dis Child 2006;91:8858–91.