Intracranial Hemorrhage after Blunt Head Trauma in Children with

Bleeding Disorders
Lois K. Lee, MD, MPH, Peter S. Dayan, MD, MSc, Michael J. Gerardi, MD, Dominic A. Borgialli, DO, MPH,
Mohamed K. Badawy, MD, James M. Callahan, MD, Kathleen A. Lillis, MD, Rachel M. Stanley, MD, Marc H. Gorelick, MD,
MSCE, Li Dong, MSc, Sally Jo Zuspan, RN, MSN, James F. Holmes, MD, MPH, and Nathan Kuppermann, MD, MPH, and the
Traumatic Brain Injury Study Group for the Pediatric Emergency Care Applied Research Network (PECARN)

Objective To determine computerized tomography (CT) use and prevalence of traumatic intracranial hemorrhage
(ICH) in children with and without congenital and acquired bleeding disorders.
Study design We compared CT use and ICH prevalence in children with and without bleeding disorders in a mul-
ticenter cohort study of 43 904 children <18 years old with blunt head trauma evaluated in 25 emergency depart-
ments.
Results A total of 230 children had bleeding disorders; all had Glasgow Coma Scale (GCS) scores of 14 to 15.
These children had higher CT rates than children without bleeding disorders and GCS scores of 14 to 15 (risk ratio,
2.29; 95% CI, 2.15 to 2.44). Of the children who underwent imaging with CT, 2 of 186 children with bleeding disor-
ders had ICH (1.1%; 95% CI, 0.1 to 3.8) , compared with 655 of 14 969 children without bleeding disorders (4.4%;
95% CI, 4.1-4.7; rate ratio, 0.25; 95% CI, 0.06 to 0.98). Both children with bleeding disorders and ICHs had symp-
toms; none of the children required neurosurgery.
Conclusion In children with head trauma, CTs are obtained twice as often in children with bleeding disorders,
although ICHs occurred in only 1.1%, and these patients had symptoms. Routine CT imaging after head trauma
may not be required in children without symptoms who have congenital and acquired bleeding disorders.
(J Pediatr 2011;158:1003-8).

I
ntracranial hemorrhage (ICH) is a significant and potentially life-threatening
complication for children with congenital or acquired bleeding disorders.1-9
There is evidence that these children are at increased risk for sustaining ICH
even after minor blunt head trauma.2,6,10 Studies in children with hemophilia
have reported ICH rates of 2% to 16% after head trauma, including some children
with no signs or symptoms of trauma. The risk of ICH varies with the severity of
hemophilia, and children with severe hemophilia (factor level <1%) are at highest
risk, from spontaneous and traumatic ICH.2,10-13 Although there are few studies
on the risk of ICH after head trauma in children with von Willebrand disease, they
seem to be at less risk than children with hemophilia.2,12,14
The risk of ICH in patients with other congenital and acquired bleeding dis-
orders is less well described.1,2,15 In patients with immune (idiopathic) thrombo-
cytopenic purpura (ITP), ICH, including spontaneous and traumatic, is rare,
with a reported incidence of 0.1% to 1.0%.1,9,16,17 However, the prevention of
ICH has been a primary goal in the management of ITP, because ICH risk cor-
relates with the severity of thrombocytopenia.1,4,17 The ICH risk in patients who
have taken anti-coagulants has only been reported in adults, with differing con-
clusions about the risk of anti-coagulation therapy.18-22
CT Computerized tomography
ED Emergency department
GCS Glasgow Coma Scale
ICH Intracranial hemorrhage
ITP Immune (idiopathic) thrombocytopenic purpura
LOC Loss of consciousness
PECARN Pediatric Emergency Care Applied Research Network
RR Rate ratio
From the Department of Pediatrics, Harvard Medical
School, Boston, MA (L.L.); Department of Pediatrics,
Columbia University College of Physicians and
Surgeons, New York, NY (P.D.); Department of
Emergency Medicine, Atlantic Health System,
Morristown Memorial Hospital, Morristown, NJ
(M.Gerardi); Department of Emergency Medicine,
University of Michigan School of Medicine and Hurley
Medical Center, Flint, MI (D.B.); Departments of
Emergency Medicine and Pediatrics, University of
Rochester School of Medicine and Dentistry, Rochester,
NY (M.B.); Departments of Emergency Medicine and
Pediatrics, SUNY-Upstate Medical University, Syracuse,
NY (J.C.); Department of Pediatrics and Emergency
Medicine, SUNY-Buffalo School of Medicine and
Biomedical Sciences, Buffalo, NY (K.L.); Department of
Emergency Medicine, University of Michigan School of
Medicine, Ann Arbor, MI (R.S.); Department of Pediatrics,
Medical College of Wisconsin, Milwaukee, WI
(M.Gorelick); Department of Pediatrics, University of
Utah and PECARN Central Data Management and
Coordinating Center, Salt Lake City, UT (L.D., S.Z.); and
Department of Emergency Medicine, University of
California, Davis School of Medicine, Davis, CA (J.H.,
N.K.)
List of members of the Traumatic Brain Injury Study
Group for the Pediatric Emergency Care Applied
Research Network (PECARN) available at www.jpeds.
com (Appendix).
Supported by a grant from the Health Resources and
Services Administration/Maternal and Child Health Bu-
reau, Division of Research, Education, and Training, and
the Emergency Medical Services of Children program
(R40MC02461). The Pediatric Emergency Care Applied
Research Network is supported by cooperative agree-
ments U03MC00001, U03MC00003, U03MC00006,
U03MC00007, and U03MC00008 from the Emergency
Medical Services of Children program of the Health Re-
sources and Services Administration/Maternal and Child
Health Bureau, Division of Research. The authors declare
no conflicts of interest.
0022-3476/$ - see front matter. Copyright ª 2011 Mosby Inc.
All rights reserved. 10.1016/j.jpeds.2010.11.036

1003
THE JOURNAL OF PEDIATRICS
www.jpeds.com
The objectives of this study were to determine the fre-
quency of computerized tomography (CT) imaging after
blunt head trauma in children with bleeding disorders com-
pared with children without bleeding disorders, and the prev-
alence of ICH in these children.
Methods
This was an a priori planned substudy conducted as part of
a larger prospective cohort study to derive and validate a neu-
roimaging decision rule for children after blunt head
trauma.23 The study was approved by the institutional review
boards at all participating institutions. Written or verbal con-
sent for this observational study was obtained at each institu-
tion as required by their institutional review boards.
The study was conducted in 25 emergency departments
(EDs) participating in the Pediatric Emergency Care Applied
Research Network (PECARN).24,25 Children <18 years old
who were evaluated for blunt head trauma resulting from
non-trivial mechanisms within 24 hours of injury at any of
the participating EDs between June 2004 and September
2006 were eligible for the main study.23 Congenital or ac-
quired bleeding disorder was defined as hemophilia, von Wil-
lebrand disease, congenital or acquired thrombocytopenia
(defined as platelet count <150 000/mL), a functional platelet
disorder, other bleeding disorder, or anti-coagulation therapy
(warfarin, heparin, low molecular weight heparin/enoxa-
parin, clopidogrel). Patients were excluded from both the
main study and this substudy when they had: (1) trivial mech-
anisms of injury (falls from standing height, walking, or run-
ning into stationary object) and (2) no signs or symptoms of
head injury besides a scalp laceration or abrasion. Patients
were also excluded from both studies when they sustained
penetrating trauma, when the injury occurred >24 hours be-
fore the ED evaluation, when they had a pre-existing neuro-
logical disease, known brain tumor, or history of ventricular
shunt placement, or when they had been transferred with cra-
nial imaging from an initial treating institution to the study
facility.
A full description of the main study protocol has been pub-
lished.23 In brief, the treating clinician conducted the exam-
ination and recorded the results on a structured case report
form before knowledge of any imaging studies, if performed.
Cranial CTs were obtained at the discretion of the treating
clinician. The case report form included information about
patient history (including history of bleeding disorders), in-
jury mechanism, symptoms, and physical examination find-
ings. When the patient had a bleeding disorder, the case
report form included check boxes for hemophilia, platelet
disorders, anticoagulation therapy, von Willebrand disease,
‘‘unknown,’’ and ‘‘other.’’ Three hundred-forty children
were indicated to have a bleeding disorder on the case report
forms. We performed a detailed secondary medical record re-
view of these cases to determine the specific type, and when
applicable, the severity of the bleeding disorder. Of these
340 children, 230 met criteria for analysis in this study. The
1004
Vol. 158, No. 6
other 110 were excluded from this study because they did
not meet the criteria for having a congenital or acquired
bleeding disorder at the time of the head injury. Patients
with hemophilia were categorized according to type of factor
deficiency and severity (mild, moderate, or severe). For chil-
dren with thrombocytopenia, platelet counts at the time of
the ED evaluation were obtained, when available from the
medical record.
Hospital admission was at the discretion of the treating ED
physician. To determine the clinical outcomes of the patients
hospitalized for their head trauma, we performed a medical
record review, and data were recorded on a structured case
report form. For children discharged home from the ED,
a follow-up telephone call was conducted by trained research
coordinators between 1 week and 3 months after the ED visit
to determine whether the patient had an unscheduled return
visit to a healthcare provider and whether any cranial imag-
ing was performed after their initial ED visit. When a missed
traumatic brain injury was suggested at follow-up, the med-
ical records and imaging results were obtained and reviewed,
and the patient’s outcome was recorded. When telephone
follow-up was not available, we mailed a follow-up survey
or reviewed the medical record, quality improvement re-
ports, trauma registries, or morgue reports at the respective
sites to obtain any missing clinical information.23
The primary outcomes were rates of CT use and presence
of an ICH on CT, as reported by an attending radiologist.
ICHs included epidural hematomas, subdural hematomas,
intraventricular hemorrhages, cerebral contusions, cerebral/
cerebellar hemorrhages, subarachnoid hemorrhages, or trau-
matic infarctions.
Data Analysis
We tabulated basic descriptive information for children with
and without bleeding disorders for the entire study popula-
tion. Because all the children with bleeding disorders pre-
sented with Glasgow Coma Scale (GCS) scores of 14 or 15,
as did 98% of the children without bleeding disorders, the re-
mainder of the analyses was performed only for children with
GCS scores of 14 and 15. We calculated rate differences with
95% CIs of the prevalence of signs and symptoms of head
trauma in children with bleeding disorders compared with
the reference population of children without bleeding disor-
ders. We also compared rates of CT imaging and ICH for pa-
tients with bleeding disorders versus patients without
bleeding disorders by using rate ratios (RRs) with 95% CIs.
We performed multivariable logistic regression analyses to
identify factors independently associated with the use of
CT imaging. Children <2 years old and $2 years old were an-
alyzed separately to optimize the inclusion of the presenting
signs and symptoms, because some symptoms (eg, headache)
cannot be accurately assessed in pre-verbal patients (<2 years
old). In these analyses, we adjusted for the diagnosis of coa-
gulopathy and the severity of mechanism of injury and other
signs and symptoms suggestive of traumatic brain injury
(history of loss of consciousness, headache, vomiting, acting
abnormally according to parent, altered mental status, signs
Lee et al
June 2011 ORIGINAL ARTICLES

of skull fractures, and scalp hematomas).23 Altered mental

status was determined a priori as a GCS score <15, agitation, Table II. Categorization of bleeding disorders (n = 230)
sleepiness, slow responses, or repetitive questioning. We per- Disorder n (%)
formed the data analysis with SAS software version 9.1.3 (SAS Hemophilia 129 (56.1%)
Institute, Cary, North Carolina). Severe 80 (62.0%)
Moderate 36 (27.9%)
Mild 13 (10.1%)
Results von Willebrand disease 45 (19.6%)
Thrombocytopenia (platelets/mL)* 34 (14.8%)
<5000 2 (5.9%)
A total of 43 904 patients with non-trivial blunt head trauma 5001-20 000 2 (5.9%)
were enrolled during the 28-month study period. Of these pa- 20 001-50 000 12 (35.3%)
50 001-150 000 17 (50.0%)
tients, 230 had a history of congenital or acquired bleeding Unknown 1 (2.9%)
disorders meeting criteria for this analysis, including 129 Anti-coagulation therapy 15 (6.5%)
(56.1%) with hemophilia. The demographic and injury sever- Functional platelet disorder 6 (2.6%)
Other bleeding disorder 1 (0.4%)
ity characteristics of the children with and without bleeding
disorders are presented in Table I. The group with bleeding *At the time of ED examination.
disorders was more likely to be male and less likely to have
severe mechanisms of injury than the children without hematomas, which was higher in children with bleeding
bleeding disorders. Because all patients with bleeding disorders.
disorders (and 98% of patients without bleeding disorders) Cranial CTs were obtained in 186 of the 230 children with
presented with seemingly minor head trauma, defined with bleeding disorders (80.9%) compared with 14 969 of the 42
GCS scores of 14 or 15, the remainder of the results 412 children without bleeding disorders (35.3%), but with
considers only children with GCS scores of 14 to 15. the same GCS scores of 14 to 15 (RR, 2.29; 95% CI, 2.15 to
After hemophilia, Von Willebrand disease and thrombocy- 2.44). Of the children with mild mechanisms of injury, chil-
topenia comprised the next largest categories of bleeding dis- dren with bleeding disorders were 3-times more likely to be
orders (Table II). Of the children with thrombocytopenia, 18 examined with CT scan (66/81, 81.5%) than children without
(54%) had ITP. Children with bleeding disorders presented bleeding disorders (1910/7106, 26.9%; RR, 3.03; 95% CI, 2.71
with a lower prevalence of symptoms and signs of ICH than to 3.39). Of the children with moderate or severe injury mech-
did the children with GCS scores of 14 to 15, but without anisms, children with bleeding disorders had twice the CT
bleeding disorders (Table III; available at www.jpeds.com). evaluation rate (114/143, 79.7%) of children without bleeding
The one exception to this was the prevalence of scalp disorders (12 895/34 993, 36.9%; RR, 2.16; 95% CI, 1.99 to
2.35).
In the multivariable logistic regression analysis, which
controlled for severity of mechanism and signs and symp-
Table I. Patient characteristics—all Glasgow Coma toms of head injury, children <2 years old with bleeding dis-
Scale scores orders had a 42-fold higher odds of having a head CT
Bleeding disorder No bleeding disorder performed than patients <2 years old without bleeding disor-
n = 230 n = 43 379*
ders (OR, 42.52; 95% CI, 19.77 to 91.45; Table IV). In
Mean age, years (SD) 6.4 (5.1) 7.1 (5.5) addition, children $2 years old with bleeding disorders had
Male sex (%) 192 (83.5) 27 040 (62.3)
Age <2 years (%) 65 (28.3) 10 901 (25.1) 23-fold higher odds of having a head CT performed than
Severity of mechanism† (%) children $2 years old without a bleeding disorder (OR,
Mild 81 (36.2) 7186 (16.7) 22.62; 95% CI, 14.75 to 34.67; Table V). The effect of GCS
Moderate 131 (58.5) 29 649 (68.9)
Severe 12 (5.4) 6210 (14.4) < 15 was captured in the multivariable logistic regression
GCS (%) analyses as it was included in the definition of the variable
#13 0 (0.0) 967 (2.2) altered mental status. Clinical characteristics of the children
14 5 (2.2) 1341 (3.1)
15 225 (97.8) 41 071 (94.7) with bleeding disorders who did not have a CT scan
CT obtained (%) 186 (80.9) 15 883 (36.6) examination for their head trauma are described in Table
ICH on CT (%) 2 (1.1) 993 (6.3) VI (available at www.jpeds.com).
*A total of 43 904 patients were enrolled in the main head trauma study. Number with no bleed- Of the patients with GCS scores of 14 to 15 for whom a CT
ing disorder and GCS 3-15 was 43 379 after the removal of 505 patients on the basis of ex- was obtained, an ICH was present in 2 of 186 children with
clusion criteria (230 with bleeding disorders; 110 initially categorized as bleeding disorder,
but on record review did not meet criteria for this analysis; 101 with ventricular shunts; 66 bleeding disorders (1.1%; 95% CI, 0.1 to 3.8) compared
with no GCS recorded; 2 had more than one exclusion) and 20 with no outcome recorded with 655 of 14 969 children without bleeding disorders
(2 with GCS 3-13).
†Severity of injury mechanism was defined as follows. Severe mechanisms included when the (4.4%; 95% CI, 4.1 to 4.7; RR, 0.25; 95% CI, 0.06 to 0.98). Fur-
patient was ejected in a motor vehicle crash; when the patient was a passenger in a motor thermore, in the children with bleeding disorders, there were
vehicle crash rollover; any passenger death in the motor vehicle crash; when the patient
was a pedestrian or un-helmeted bicyclist struck by automobile; fall >5 feet when $2 years no patients who initially had ED CT results that were negative
old or >3 feet when <2 years old; or when the head was struck by a high-impact object (eg, golf for ICH then subsequently had positive CT results. Similarly,
club). Mild injury mechanisms included falls to the ground from standing height or walking/run-
ning into stationary objects associated with signs or symptoms of blunt head trauma. All other none of the children with bleeding disorders who did not un-
mechanisms were considered moderate. dergo imaging in the ED had positive CT results on follow-up.
Intracranial Hemorrhage after Blunt Head Trauma in Children with Bleeding Disorders 1005
THE JOURNAL OF PEDIATRICS
www.jpeds.com Vol. 158, No. 6

dition who fell 3 to 5 feet. On presentation to the ED, he had

Table IV. Multivariable logistic regression analysis of a GCS score of 15, complaints of a mild headache, and had
factors associated with the use of computerized two episodes of emesis. Physical examination revealed a me-
tomography in children younger than 2 years with dium-size temporal scalp hematoma. He had an epidural he-
Glasgow Coma Scale scores of 14 or 15 matoma on CT, for which he was hospitalized for 3 nights
Odds ratio for CT examination with no neurosurgical intervention.
(95% CI)
Variable Unadjusted Adjusted
Presence of bleeding disorder 15.83 (7.55-33.23) 42.52 (19.77-91.45) Discussion
Mechanism severity
Mild 1.0 (reference) 1.0 (reference)
Moderate/Severe 1.73 (1.52-1.96) 2.11 (1.80-2.48) In this large, prospective, multicenter study of children with
History of loss of consciousness 6.04 (5.00-7.30) 6.55 (5.25-8.16) blunt head trauma, 1% of imaged patients with bleeding dis-
History of vomiting 3.18 (2.85-3.55) 3.18 (2.78-3.64) orders had ICHs. This study is unique in that it included chil-
Acting abnormally per 6.67 (5.91-7.53) 3.67 (3.15-4.27)
parent dren with acquired and congenital bleeding disorders, who
Altered mental status 9.26 (8.05-10.65) 4.85 (4.07-5.78) have varying degrees of risk for the development of traumatic
Signs of basilar skull 29.34 (10.63-81.01) 28.21 (8.25-96.53) ICH on the basis of the type and severity of the bleeding dis-
fracture
Palpable skull fracture 6.75 (5.32-8.56) 6.78 (5.07-9.08) order. Overall, patients with bleeding disorders (all with GCS
Scalp hematoma scores of 14-15) were twice as likely to have a cranial CT per-
None 1.0 (reference) 1.0 (reference) formed than children without bleeding disorders and with
Frontal 0.90 (0.81-0.99) 1.09 (0.96-1.23)
Non-frontal 2.55 (2.28-2.85) 2.77 (2.41-3.18) GCS scores of 14 to 15. The two children with bleeding dis-
orders and ICH after head trauma had other signs and symp-
toms of ICH that would have warranted cranial imaging.
Therefore, CT imaging may not routinely be needed in the
The two patients with bleeding disorders and ICHs on CT evaluation of children with bleeding disorders after blunt
scan presented with signs and symptoms of ICH. One was head trauma, particularly in those without signs and symp-
a 15- year-old boy with severe factor VIII deficiency who toms suggestive of ICH.
was involved in a motorcycle crash. On ED presentation, Head trauma-related ICH in patients with bleeding disor-
he had a GCS score of 15, complained of a moderate head- ders has been best described in patients with hemophilia (fac-
ache, and exhibited repetitive questioning. On physical ex- tor VIII and factor IX deficiency). ICH is the leading cause of
amination, he had a large parietal scalp hematoma. His CT mortality from bleeding in this population, and the reported
scan revealed a right frontal lobe parenchymal hemorrhagic prevalence of head trauma-related ICH in patients with he-
contusion. He was treated with factor VIII and hospitalized mophilia ranges from 2% to 16%. Current recommendations
for 2 nights because of his head injury. No neurosurgical in- for the management of head trauma in children with severe
tervention was required. The second patient was a 6-year-old hemophilia are to initiate treatment with factor replacement
boy receiving warfarin therapy for a congenital cardiac con- as soon as possible after the traumatic event.26,27 The signif-
icant decrease in mortality rate from ICH (spontaneous and
traumatic) in patients with hemophilia has been attributed to
the wide availability of factor concentrates for replacement.10
Table V. Multivariable logistic regression analysis of
The factor level at the time of injury and the use of factor cor-
factors associated with the use of computerized
rection were not evaluated in this study, because our primary
tomography in children 2 years or older with Glasgow
outcomes were the use of CT and presence of ICH on CT in
Coma Scale scores of 14 or 15
this population and not the effect of treatment on the clinical
Odds ratio for CT examination complications from the bleeding disorder. Although institu-
(95% CI)
tional guidelines for the use of CT after head trauma are
Variable Unadjusted Adjusted
available, we were not able to identify any published standard
Presence of bleeding disorder 6.17 (4.26-8.93) 22.62 (14.75-34.67) clinical practice guidelines on the topic. In our prospective
Mechanism severity
Mild 1.0 (reference) 1.0 (reference) cohort, only one of 129 patients with hemophilia who under-
Moderate/Severe 1.53 (1.44-1.63) 1.71 (1.56-1.88) went imaging with CT had an ICH. He was treated with fac-
History of loss of consciousness 12.73 (11.85-13.68) 14.75 (13.53-16.07) tor replacement alone.
Headache 3.38 (3.21-3.56) 2.54 (2.37-2.71)
History of vomiting 5.14 (4.78-5.53) 5.70 (5.16-6.29) Our results for children with hemophilia differ somewhat
Altered mental status 14.01 (12.83-15.31) 10.55 (9.38-11.87) from earlier studies, which have mainly been retrospective or
Signs of basilar skull 20.45 (12.63-33.10) 23.70 (13.65-41.16) small. One retrospective study of children with hemophilia
fracture
Palpable skull fracture 4.51 (3.81-5.34) 6.79 (5.42-8.52) identified 374 ED visits for head trauma in 11 years, with 9
Scalp hematoma episodes of ICH (2.4%). Five of these patients had no re-
None 1.0 (reference) 1.0 (reference) ported signs or symptoms of ICH during the time of the
Frontal 0.78 (0.74-0.83) 0.91 (0.83-1.00)
Non-frontal 1.29 (1.22-1.37) 1.25 (1.15-1.36) ED evaluation, which ranged from 1 to 10 hours after the
head trauma event. All patients were treated with factor
1006 Lee et al
June 2011
replacement for 10 to 14 days; none of the patients died or
required neurosurgical intervention.10 A slightly higher prev-
alence of ICH was noted in a retrospective study of children
with hemophilia and von Willebrand disease, with ICH oc-
curring in 5 of 109 episodes of blunt head trauma (4.6%).
All 5 patients had hemophilia and, similar to our study, pre-
sented with symptoms suggestive of ICH, including abnor-
mal neurological examinations.2 A prospective study from
1981 of children and adults with hemophilia reported 6 pa-
tients with ICH out of 47 episodes of head trauma; however,
few details are available to determine the symptoms and signs
of those with ICH.11
The incidence of ICH is low in children with thrombocy-
topenia and has primarily been described in children with
ITP.1,4,9,16,17 One review of the literature from 1954 to 1998
identified 75 published cases of ICH in children with ITP;
however, only 9 of these children had a history of head
trauma.1 A recent case control study of 40 children with
ITP who sustained ICH reported 33% (13/40) had a preced-
ing history of head trauma, compared with 80 children with
ITP and no ICH where only one child had a history of head
trauma (1.2 %).17 There were 34 children in our study pop-
ulation with thrombocytopenia (<150 000 platelets/mL);
none of them sustained an ICH.
Only a small percentage of our study patients were receiv-
ing anti-coagulation therapy. Of the 16 patients receiving
anti-coagulation therapy in this study, only one child (receiv-
ing warfarin) sustained an ICH after a 3- to 5-foot fall and
had symptoms (headache and vomiting). Traumatic ICH
in patients receiving anti-coagulation therapy has primarily
been described in the adult literature, with conflicting con-
clusions about the associated risks of ICH. One prospective
case-control study of adults medicated with warfarin found
a trend toward increased mortality after head trauma.21 A
retrospective study of 144 adult patients receiving warfarin
who were defined as low-risk for ICH by symptoms (no
symptoms, dizziness, or headache), identified 10 patients
with clinically important CT findings.19 In contrast, a smaller
retrospective study of 65 patients receiving warfarin with mi-
nor head trauma without loss of consciousness found that
none had ICH on CT.18 With a growing number of children
surviving with chronic medical conditions requiring anti-
coagulation (eg, congenital heart disease), caution should
still be advised about the potential risk of ICH with head
trauma while more data are gathered.
This study has some limitations. Although this was a very
large prospective study of pediatric head trauma, there were
limited numbers of patients with congenital or acquired
bleeding disorders in our study population, especially those
with severe hemophilia, severe thrombocytopenia, or receiv-
ing anti-coagulation therapy. Therefore, the precision of our
findings is limited. Analysis of the larger head trauma study
from which this subanalysis was performed demonstrated
that the missed eligible population was similar to the study
population, making it unlikely that a substantial number of
children with bleeding disorders were not enrolled.23 Al-
though only two patients in our cohort sustained ICHs,
Intracranial Hemorrhage after Blunt Head Trauma in Children with Bleeding Disorders
ORIGINAL ARTICLES
both had physical signs and symptoms of ICH, suggesting
that symptomatically silent ICHs in children with bleeding
disorders are very uncommon. Children with hemophilia
are also at risk for a delayed presentation of ICH after head
trauma; however, children presenting for ED evaluation >24
hours after head trauma were not included in this study, be-
cause they were eligible only when they presented within 24
hours. However, of the children with bleeding disorders
who presented within 24 hours of the injury, none had de-
layed bleeding, which we would have detected at follow-up.
Despite these limitations, it is unlikely that a prospective
study larger than this one will be conducted in the near fu-
ture, and this study provides the largest prospectively con-
ducted study on the topic. In this prospective study of
blunt head trauma in children with congenital and acquired
bleeding disorders, the prevalence of ICH was very low. The
two patients with bleeding disorders and ICH had signs and
symptoms suggestive of ICH, which would have warranted
cranial CT evaluation. Although patients with congenital or
acquired bleeding disorders are at risk for ICH, the low rate
of ICH suggests that they may not routinely require cranial
CT imaging after minor blunt head trauma in the absence
of signs or symptoms of ICH. n
We thank Rene Enriquez at the PECARN Data Center (University of
Utah) for his dedicated and diligent work, the research coordinators in
PECARN, without whose dedication and hard work this study would
not have been possible, and all the clinicians around the PECARN
who enrolled children in this study.
Submitted for publication Aug 16, 2010; last revision received Sep 20, 2010;
accepted Nov 15, 2010.
Reprint requests: Lois K. Lee, MD, MPH, Division of Emergency Medicine,
Children’s Hospital, Boston, 300 Longwood Ave, Boston, MA 02115. E-mail:
lois.lee@childrens.harvard.edu
References
1. Butros LJ, Bussel JB. Intracranial hemorrhage in immune thrombocyto-
penic purpura: a retrospective analysis. J Pediatr Hematol Oncol 2003;
25:660-4.
2. Dietrich AM, James CD, King DR, Ginn-Pease ME, Cecalupo AJ. Head
trauma in children with congenital coagulation disorders. J Pediatr Surg
1994;29:23-32.
3. Klinge J, Auberger K, Auerswald G, Brackmann HH, Mauz-Korholz C,
Kreuz W, et al. Prevalence and outcome of intracranial haemorrhage
in haemophiliacs—a survey of the paediatric group of the German Soci-
ety of Thrombosis and Haemostasis (GTH). Eur J Pediatr 1999;
158(Suppl 3):S162-5.
4. Lilleyman JS, Paediatric Haematology Forum of the British Society for
Haematology. Intracranial haemorrhage in idiopathic thrombocytope-
nic purpura. Arch Dis Child 1994;71:251-3.
5. Ljung RCR. Intracranial haemorrhage in haemophilia A and B. Br J Hae-
matol 2007;140:378-84.
6. Lutschg J, Vassella F. Neurological complications in hemophilia. Acta
Paediatr Scand 1981;70:235-41.
7. Martinowitz U, Heim M, Tadmor R, Eldor A, Rider I, Findler G, et al.
Intracranial hemorrhage in patients with hemophilia. Neurosurgery
1986;18:538-41.
8. Stieltjes N, Calvez T, Demiguel V, Torchet MF, Briquel ME,
Fressinaud E, et al. Intracranial haemorrhages in French haemophilia
1007
THE JOURNAL OF PEDIATRICS
www.jpeds.com
patients (1991-2001): clinical presentation, management and prognosis
factors for death. Haemophilia 2005;11:452-8.
9. Woerner SJ, Abildgaard CF, French BN. Intracranial hemorrhage in chil-
dren with idiopathic thrombocytopenic purpura. Pediatrics 1981;67:
453-60.
10. Witmer CM, Raffini L, Manno CS. Utility of computed tomography of
the head following head trauma in boys with haemophilia. Haemophilia
2007;13:560-6.
11. Andes WA, Wulff K, Smith WB. Head trauma in hemophilia. Arch In-
tern Med 1981;144:1981-3.
12. Hennes H, Losek JD, Sty JR, Gill JC. Computerized tomography
in hemophiliacs with head trauma. Pediatr Emerg Care 1987;3:
147-9.
13. Nelson MD, Maeder MA, Usner D, Mitchell WG, Fenstermacher MJ,
Wilson DA, et al. Prevalence and incidence of intracranial haemor-
rhage in a population of children with haemophilia. Haemophilia
1999;5:306-12.
14. Mizoi K, Onuma T, Mori K. Intracranial hemorrhage secondary to von
Willebrand’s disease and trauma. Surg Neurol 1984;22:495-8.
15. Kyrnetskiy EE, Kun LE, Boop FA, Sanford RA, Khan RB. Types, causes,
and outcome of intracranial hemorrhage in children with cancer. J Neu-
rosurg 2005;102:31-5.
16. Iyori H, Bessho F, Ookawa H, Konishi S, Shirahata A, Miyazaki S, et al.
Intracranial hemorrhage in children with immune thrombocytopenic
purpura. Ann Hematol 2000;79:691-5.
17. Psaila B, Petrovic A, Page LK, Menell J, Schonholz M, Bussel JB. Intra-
cranial hemorrhage (ICH) in children with immune thrombocytopenia
(ITP): study of 40 cases. Blood 2009;114:4777-83.
1008
Vol. 158, No. 6
18. Garra G, Nashed AH, Capobianco L. Minor head trauma in anticoagu-
lated patients. Acad Emerg Med 1999;6:121-4.
19. Li J, Brown J, Levine M. Mild head injury, anticoagulants, and risk of in-
tracranial injury. Lancet 2001;357:771-2.
20. Mina AA, Knipfer JF, Park DY, Bair HA, Howells GA, Bendick PJ. Intra-
cranial complications of preinjury anticoagulation in trauma patients
with head injury. J Trauma 2002;53:668-72.
21. Mina AA, Bair HA, Howells GA, Bendick PJ. Complications of preinjury
warfarin use in the trauma patient. J Trauma 2003;54:842-7.
22. Cohen DB, Rinker C, Wilberger JE. Traumatic brain injury in anticoagu-
lated patients. J Trauma 2006;60:553-7.
23. Kuppermann N, Holmes JF, Dayan PS, Hoyle JD, Atabaki SM,
Holubkov R, et al. Identification of children at very low risk of
clinically-important brain injuries after head trauma: a prospective co-
hort study. Lancet 2009;374:1160-70.
24. The Pediatric Emergency Care Applied Research Network. The Pediatric
Emergency Care Applied Research Network (PECARN): rationale, de-
velopment, and first steps. Acad Emerg Med 2003;10:661-8.
25. Dayan P, Chamberlain J, Dean JM, Maio RF, Kuppermann N. The Pedi-
atric Emergency Care Applied Research Network: progress and update.
Clin Pediatr Emerg Med 2006;7:128-35.
26. Witmer CM, Manno CS, Butler RB, Raffini LJ. The clinical management
of hemophilia and head trauma: a survey of current clinical practice
among pediatric hematology/oncology physicians. Pediatr Blood Cancer
2009;53:406-10.
27. Hoots WK, Shapiro AD. Treatment of hemophilia. In: Fleisher GR, Marx
JA, Walls RM, editors. UpToDate. www.uptodate.com. Accessed January
11, 2010.
Lee et al
June 2011 ORIGINAL ARTICLES

Appendix Louis Children’s Hospital (K. Quayle, D. Jaffe); Women

Participating centers and site investigators of the Trau-
matic Brain Injury Study Group for the Pediatric Emer-
gency Care Applied Research Network (PECARN) are
listed in alphabetical order: Atlantic Health System/Morris-
town Memorial Hospital (M. Gerardi); Bellevue Hospital
Center (M. Tunik, J. Tsung); Calvert Memorial Hospital
(K. Melville); Children’s Hospital Boston (L. Lee); Chil-
dren’s Hospital of Michigan (P. Mahajan); Children’s Hos-
pital of New York–Presbyterian (P. Dayan); Children’s
Hospital of Philadelphia (F. Nadel); Children’s Memorial
Hospital (E. Powell); Children’s National Medical Center
(S. Atabaki, K. Brown); Cincinnati Children’s Hospital
Medical Center (T. Glass); DeVos Children’s Hospital (J.
Hoyle); Harlem Hospital Center (A. Cooper); Holy Cross
Hospital (E. Jacobs); Howard County Medical Center (D.
Monroe); Hurley Medical Center (D. Borgialli); Medical
College of Wisconsin/Children’s Hospital of Wisconsin
(M. Gorelick, S. Bandyopadhyay); St Barnabas Health
Care System (M. Bachman, N. Schamban); SUNY–Upstate
Medical Center (J. Callahan); University of California Davis
Medical Center (N. Kuppermann, J. Holmes); University
of Maryland (R. Lichenstein); University of Michigan
(R. Stanley); University of Rochester (M. Badawy, L.
Babcock-Cimpello); University of Utah/Primary Children’s
Medical Center (J. Schunk); Washington University/St.
and Children’s Hospital of Buffalo (K. Lillis).
We acknowledge the efforts of these individuals participat-
ing in PECARN at the time this study was initiated: PECARN
Steering Committee: N. Kuppermann, Chair; E. Alpern, J.
Chamberlain, J. M. Dean, M. Gerardi, J. Goepp, M. Gorelick,
J. Hoyle, D. Jaffe, C. Johns, N. Levick, P. Mahajan, R. Maio,
K. Melville, S. Miller (deceased), D. Monroe, R. Ruddy, R.
Stanley, D. Treloar, M. Tunik, A. Walker. MCHB/EMSC liai-
sons: D. Kavanaugh, H. Park; Central Data Management and
Coordinating Center (CDMCC): M. Dean, R. Holubkov, S.
Knight, A. Donaldson; Data Analysis and Management Sub-
committee (DAMS): J. Chamberlain, Chair; M. Brown, H.
Corneli, J. Goepp, R. Holubkov, P. Mahajan, K. Melville, E.
Stremski, M. Tunik; Grants and Publications Subcommittee
(GAPS): M. Gorelick, Chair; E. Alpern, J. M. Dean, G. Foltin,
J. Joseph, S. Miller*, F. Moler, R. Stanley, S. Teach; Protocol
Concept Review and Development Subcommittee
(PCRADS): D. Jaffe, Chair; K. Brown, A. Cooper, J. M.
Dean, C. Johns, R. Maio, N. C. Mann, D. Monroe, K.
Shaw, D. Teitelbaum, D. Treloar; Quality Assurance Sub-
committee (QAS): R. Stanley, Chair; D. Alexander, J. Brown,
M. Gerardi, M. Gregor, R. Holubkov, K. Lillis, B. Nordberg,
R. Ruddy, M. Shults, A. Walker; Safety and Regulatory Affairs
Subcommittee (SRAS): N. Levick, Chair; J. Brennan, J.
Brown, J. M. Dean, J. Hoyle, R. Maio, R. Ruddy, W. Schalick,
T. Singh, J. Wright.

Table III. Presenting findings in children with Glasgow Coma Scale scores of 14 and 15
Bleeding disorder No bleeding disorder
(n = 230) (n = 42 412) Rate difference*
Symptoms n/n (%) n/n (%) (95% CI)
History of loss of consciousness 11/229 (4.8) 6286/40 693 (15.4) 10.6 (13.4 to 7.9)
Headache† 57/157 (36.3) 12 700/28 518 (44.5) 8.2 (15.8 to 0.7)
History of vomiting 8/226 (3.5) 5557/42 112 (13.2) 9.7 (12.1 to 7.2)
Acting abnormally according to parent 24/225 (10.7) 6197/39 406 (15.7) 5.1 (9.1 to 1.0)
Altered mental status 9/228 (3.9) 5487/42 096 (13.0) 9.1 (11.6 to 6.5)
Signs of basilar skull fracture 0/229 (0.0) 287/41 991 (0.7) 0.7 (0.8 to 0.6)
Palpable skull fracture (or unclear exam) 4/230 (1.7) 1044/42 311 (2.5) 0.7 (2.4 to 1.0)
Scalp hematoma
Frontal 59/228 (25.9) 8753/41 919 (20.9) 5.0 (0.7 to 10.7)
Non-frontal 55/228 (24.1) 7761/41 919 (18.5) 5.6 (0.04 to 11.2)
Seizure 2/228 (0.9) 494/41 692 (1.2) 0.3 (1.5 to 0.9)
*Rate difference calculated from bleeding disorder and no bleeding disorder cohorts with GCS scores of 14 to 15, because all 230 bleeding disorder subjects had GCS scores of 14 to 15.
†Only recorded for children $2 years old.

Intracranial Hemorrhage after Blunt Head Trauma in Children with Bleeding Disorders 1008.e1
THE JOURNAL OF PEDIATRICS
www.jpeds.com Vol. 158, No. 6

Table VI. Characteristics of patients with bleeding disorders who were not examined with computerized tomography
Bleeding disorder
with no CT (n = 44) Mild injury mechanism Moderate/severe injury mechanism
n (%) n (%) n (%)
Bleeding disorder
Hemophilia 17 (38.6) 7 (41.2) 10 (58.8)
Mild 5 (29.4) 1 (14.3) 4 (40.0)
Moderate/severe 12 (70.6) 6 (85.7) 6 (60.0)
von Willebrand disease 11 (25.0) 2 (18.2) 9 (81.8)
Thrombocytopenia 10 (22.7) 5 (50.0) 5 (50.0)
<20 000 plts/mL 1 (10.0) 1 (20.0) 0 (0.0)
$20 000 plts/mL 8 (80.0) 4 (80.0) 4 (80.0)
Unknown 1 (10.0) 0 (0.0) 1 (20.0)
Anti-coagulation therapy 5 (11.4) 0 (0.0) 5 (100.0)
Other 1 (2.3) 1 (100.0) 0 (0.0)
Severity of injury mechanism
Mild 15 (34.1)
Moderate 27 (61.4)
Severe 2 (4.5)
History of loss of consciousness 0 (0.0)
Headache* 6 (17.6)
History of vomiting 0 (0.0)
Acting abnormally per parent 0 (0.0)
GCS
14 0 (0.0)
15 44 (100)
Altered mental status 0 (0.0)
Signs of basilar skull fracture 0 (0.0)
Palpable skull fracture† 1 (2.3)
Scalp hematoma
Frontalz 9 (20.5)
Temporal/parietalx 3 (6.8)
Occipital{ 5 (11.4)
*Ten children had missing data (6 were <2 years old).
†This child had hemophilia.
zSix children had hemophilia.
xOne child had hemophilia.
{Two children had hemophilia.

1008.e2 Lee et al