In 2009, the Pediatric Emergency Care Applied Research Network (PECARN) derived 2 age-specific prediction rules to identify children with minor blunt head trauma, at low risk of clinically important traumatic brain injuries (ciTBIs) for whom computed tomography (CT) could safely be avoided on the basis of the absence of predictor variables.1 In the presence of PECARN predictors, the risk of ciTBI differs according to the type and number of predictors.2–8 Of the 6 PECARN predictors, altered mental status (AMS) was found to be associated with a higher risk of ciTBI, justifying performing CT scans on patients with AMS.9 To our knowledge, however, the risk of ciTBI in children with isolated AMS (ie, no other PECARN ciTBI predictors present) has not been investigated. We aimed to assess the risk of ciTBI and TBI on CT in children with isolated AMS.
Methods
This was a secondary analysis of a prospective cohort study at 25 emergency departments (EDs) between June 2004 and September 2006. The study included 42 412 children aged <18 years with Glasgow Coma Scale (GCS) scores ≥14 within 24 hours of blunt head trauma. Approval for the parent study was obtained from the institutional review board at each participating site. A detailed description of the parent study methods has been previously published.1 Because we were interested in identifying patients with isolated AMS, patients missing documentation of any of the PECARN risk factors or the ciTBI outcome were excluded from our analysis.
As per the parent study, AMS was defined by a GCS score of 14 and/or other minor signs of AMS (ie, agitation, somnolence, repetitiveness, slow response to speech at ED presentation). Our outcome measures were:
ciTBI, defined as death, neurosurgery, intubation >24 hours, or hospitalization ≥2 nights associated with TBI on CT; and
TBI on ED CT.
To determine the ciTBI outcome, hospitalized patients were followed until discharge. For patients discharged from the ED, follow-up telephone or mail surveys were conducted within 90 days of the ED visit to identify children with initially undiagnosed ciTBI. For discharged patients unreachable for follow-up, the patient’s medical record, process improvement records, trauma registries, and county morgue records were reviewed.1
Results
A total of 5084 children with AMS and complete PECARN predictors and outcome status were included. Of these, 1245 (24.5%) had isolated AMS, of whom 194 (15.6%) were aged <2 years, and 1051 (84.4%) were aged 2 years or older. Of the 1245 children, 17 (1.4%, 95% CI 0.8–2.2) had ciTBI, and 31 (4.0%, 2.7–5.6) of 785 patients who underwent CT had TBIs. Of the 17 children with isolated AMS who had ciTBIs, none underwent neurosurgery. Results stratified by age are presented in Table 1. Study outcomes in children by components of isolated AMS are reported in Table 2.
Study Outcomes in Children With PECARN Isolated AMS Compared With Children With AMS and Other PECARN Predictors, Stratified by Age Group
| . | . | . | n of N (%) [95% CI] . | . | |
|---|---|---|---|---|---|
| Outcome . | Age Group . | No. . | AMS With No Other PECARN Predictors (Isolated) n = 1245 . | AMS With Other PECARN Predictors (Nonisolated) n = 3839 . | Rate Difference (95% CI) . |
| ciTBI | <2 y | 1153 | 2 of 194 (1.0) [0.1–3.7] | 46 of 959 (4.8) [3.5–6.4] | 3.8% (1.0–5.6) |
| ≥2 y | 3931 | 15 of 1051 (1.4) [0.8–2.3] | 145 of 2880 (5.0) [4.3–5.9] | 3.6% (2.5–4.7) | |
| TBI on CTa | <2 y | 876 | 4 of 88 (4.6) [1.3–11.2] | 97 of 788 (12.3) [10.1–14.8] | 7.8% (0.9–11.6) |
| ≥2 y | 3345 | 27 of 697 (3.9) [2.6–5.6] | 216 of 2648 (8.2) [7.1–9.3] | 4.3% (2.3–5.9) | |
| . | . | . | n of N (%) [95% CI] . | . | |
|---|---|---|---|---|---|
| Outcome . | Age Group . | No. . | AMS With No Other PECARN Predictors (Isolated) n = 1245 . | AMS With Other PECARN Predictors (Nonisolated) n = 3839 . | Rate Difference (95% CI) . |
| ciTBI | <2 y | 1153 | 2 of 194 (1.0) [0.1–3.7] | 46 of 959 (4.8) [3.5–6.4] | 3.8% (1.0–5.6) |
| ≥2 y | 3931 | 15 of 1051 (1.4) [0.8–2.3] | 145 of 2880 (5.0) [4.3–5.9] | 3.6% (2.5–4.7) | |
| TBI on CTa | <2 y | 876 | 4 of 88 (4.6) [1.3–11.2] | 97 of 788 (12.3) [10.1–14.8] | 7.8% (0.9–11.6) |
| ≥2 y | 3345 | 27 of 697 (3.9) [2.6–5.6] | 216 of 2648 (8.2) [7.1–9.3] | 4.3% (2.3–5.9) | |
TBI on CT, defined as any intracranial bleeding, pneumocephalus, cerebral edema, skull fracture depressed by at least the width of the skull, or diastasis of the skull. Patients with isolated skull fractures not depressed by the skull width were not classified as having TBI on CT.
ciTBI and TBI on CT Stratified by AMS Component, Regardless of Age
| . | n of N (%) [95% CI] . | |
|---|---|---|
| . | All Children With AMS and No Other PECARN Predictors . | |
| Isolated AMS Component . | ciTBI . | TBI on CT . |
| GCS of 14 alone (and no other signs of AMS)a | 1 of 97 (1.0) [0.0–5.6] | 2 of 47 (4.3) [0.5–14.5] |
| GCS of 15 but at least 1 other sign of AMSa | 12 of 1035 (1.2) [0.6–2.0] | 20 of 633 (3.2) [1.9–4.8] |
| Both GCS 14 and at least 1 other sign of AMSa | 4 of 113 (3.5) [1.0–8.8] | 9 of 105 (8.6) [4.0–15.7] |
| . | n of N (%) [95% CI] . | |
|---|---|---|
| . | All Children With AMS and No Other PECARN Predictors . | |
| Isolated AMS Component . | ciTBI . | TBI on CT . |
| GCS of 14 alone (and no other signs of AMS)a | 1 of 97 (1.0) [0.0–5.6] | 2 of 47 (4.3) [0.5–14.5] |
| GCS of 15 but at least 1 other sign of AMSa | 12 of 1035 (1.2) [0.6–2.0] | 20 of 633 (3.2) [1.9–4.8] |
| Both GCS 14 and at least 1 other sign of AMSa | 4 of 113 (3.5) [1.0–8.8] | 9 of 105 (8.6) [4.0–15.7] |
Other signs of AMS were defined as agitation, somnolence, repetitive questioning, or slow response to verbal communication.
Discussion
Our study demonstrated that ciTBIs are uncommon in children with AMS and no other PECARN risk factors, and none required neurosurgery. For patients with no other PECARN risk factors, we found that children with GCS scores of 14 and no other signs of AMS, and children with GCS scores of 15 but with other signs of AMS, had substantially lower risks of ciTBIs than children with both GCS scores of 14 and other signs of AMS. These data are important for clinicians facing the decision of whether to obtain neuroimaging in the nonnegligible number of children with minor blunt head trauma presenting with isolated AMS. Although based on an older data set, ours is the largest prospective study of isolated AMS and has not been duplicated since the time the parent study was conducted.
Limitations include possible interrater variability in assessment of the GCS score10 and a likely higher risk of ciTBI in patients with associated signs and symptoms other than PECARN predictors.
However, given our results and depending on clinician and parent comfort, among children with no other PECARN risk factors, those with either (1) isolated GCS scores of 14, or (2) GCS scores of 15 and isolated other signs of AMS, shared decision-making with parents about close observation in the ED until symptoms resolve versus CT may be considered. In the presence of a GCS of 14 with other signs of AMS, however, clinicians should have a low threshold to obtain a CT scan.
Dr Bressan helped conceive the study, interpreted the data, and drafted the initial manuscript; Dr Tancredi helped conceive the study design, analyzed and interpreted the data, and critically revised the final manuscript; Dr Heidt helped conceive the study design, contributed to interpreting the data, and critically revised the final manuscript; Dr Wang contributed to interpreting the data and critically revised the manuscript; Dr Kuppermann conceived and designed the present and parent study, obtained funding for the parent study, supervised training of study personnel, supervised patient enrollment, contributed to data acquisition, analysis, and interpretation, and drafted the initial manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspect of the work.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLAIMER: The authors have indicated they have no conflicts of interest relevant to this article to disclose.
Comments