Presentations and Outcomes of Children With Intraventricular Hemorrhages After Blunt Head Trauma

Objective To describe the clinical presentations and outcomes of children with intraventricular hemorrhages (IVHs) after blunt head trauma (BHT).

Design Subanalysis of a large, prospective, observational cohort study performed from June 1, 2004, through September 31, 2006.

Setting Twenty-five emergency departments participating in the Pediatric Emergency Care Applied Research Network.

Patients Children presenting with IVH after BHT.

Exposure Blunt head trauma.

Main Outcome Measures Clinical presentations and outcomes, including the Pediatric Overall Performance Category (POPC) and Pediatric Cerebral Performance Category (PCPC) scores at hospital discharge.

Results Of 15 907 patients evaluated with computed tomography, 1156 (7.3%) had intracranial injuries. Forty-three of the 1156 (3.7%; 95% CI, 2.7%-5.0%) had nonisolated IVHs (ie, with intracranial injuries on computed tomography), and 10 of 1156 (0.9%; 95% CI, 0.4%-1.6%) had isolated IVHs. Only 4 of 43 (9.3%) of those with nonisolated IVHs had Glasgow Coma Scale (GCS) scores of 14 to 15, and all 10 (100.0%) with isolated IVHs had GCS scores of 15. No patients with isolated IVHs required neurosurgery or died. One patient had moderate overall disability (by the POPC score), and no patient had moderate or severe disability at discharge (by the PCPC score). Of the 43 patients with nonisolated IVHs, however, 16 (37.2%) died and 18 (41.9%) required neurosurgery. In 27 patients (62.8%), injuries ranged from moderate overall disability to brain death by the POPC score.

Conclusions Children with nonisolated IVHs after BHT typically present with GCS scores of less than 14, frequently require neurosurgery, and have high mortality rates. In contrast, those with isolated IVHs typically present with normal mental status and are at low risk for acute adverse events and poor outcomes.

Blunt head trauma (BHT) is one of the most common reasons for children to be evaluated in emergency departments (EDs).¹ Depending on a variety of factors, including the specific mechanisms of BHT, a wide range of intracranial injuries (ICIs) and outcomes after BHT are observed.^1,2 Computed tomography (CT) is the diagnostic tool most commonly used to evaluate children promptly after BHT.^3-5 Although prompt CT provides specific information regarding the anatomical location of the ICI if present, it provides limited information on the functional consequences of identified lesions.^4,6-8

Intraventricular hemorrhage (IVH) is a type of injury uncommonly seen in adults or children undergoing cranial CT after head trauma.⁹ The reported prevalence of IVH in patients with BHT undergoing cranial CT is 1.4%.¹⁰ As a consequence of its infrequency, there is limited information regarding the clinical presentations and outcomes of children with traumatic IVHs. Most reported cases of traumatic IVHs are secondary to severe mechanisms of injury and have poor outcomes. Atzema et al,¹⁰ in a large multicenter study of adult and pediatric BHT, found that 70% of patients with IVHs had Glasgow Outcome Scale scores of 1, 2, or 3 (poor outcome). Similarly, Kobayashi et al¹¹ described 4 pediatric patients with IVHs after severe mechanisms of injury, 2 of whom died of their injuries.

Most reported cases of IVHs are also associated with other ICIs.^10,12,13 The cause of reported poor outcomes in patients with IVHs and other associated head injuries is unclear but may be a consequence of intraventricular blood migrating from nearby mechanically disrupted tissues. Intraventricular hemorrhages in association with other ICIs, such as subdural hematomas, shearing injuries, brain contusions, and intracerebral hemorrhages, may be a marker of greater injury to the brain and poor prognosis.¹² It is unclear, however, whether IVHs that are isolated (ie, without other ICIs visible on CT) have the same clinical significance. The goal of this study was to describe the prevalence, clinical presentations, and outcomes of children with traumatic IVHs in a large study of pediatric head trauma. We hypothesized that IVHs are seen uncommonly in children after minor BHT (defined by Glasgow Coma Scale [GCS] scores of 14-15), and children with isolated IVHs are at low risk for adverse outcomes.

We performed a subanalysis of a large prospective cohort study of children younger than 18 years with BHT who presented to any of the 25 hospital EDs participating in the Pediatric Emergency Care Applied Research Network (PECARN) from June 1, 2004, through September 31, 2006. The study was approved by the institutional review boards at all participating sites and at the institutional review board of the university for the Data Coordinating Center. Study methods have been previously described in detail.¹⁴ In brief, we enrolled patients with BHT who presented to one of the participating EDs within 24 hours of the traumatic event. Patients with trivial mechanisms of injury (eg, ground-level falls or running into stationary objects) without signs of traumatic brain injury (TBI) were excluded. We also excluded patients with penetrating trauma, ventricular shunts, bleeding disorders, or previous neuroimaging. Specific methods pertinent to this subanalysis are provided in the following paragraphs.

We analyzed the subset of children who underwent cranial CT as part of their ED evaluation, and the site radiologist confirmed any abnormal collection of blood within the ventricular system. If there was more than 1 CT scan performed, we used the scan that most definitively described the IVH(s). Equivocal CT scans were adjudicated by a senior faculty study pediatric radiologist (S.L.W.-G.) at the primary study site. One patient was excluded because the Pediatric Overall Performance Category (POPC) and the Pediatric Cerebral Performance Category (PCPC) functional outcome scores were not available.

We compared patients with isolated IVHs with those with nonisolated IVHs with regard to clinical presentations and outcomes. We defined an isolated IVH as a CT-identified IVH alone without any other intracranial injury (with or without linear skull fractures). We defined a nonisolated IVH as an IVH in association with at least 1 other ICI on CT. We defined an ICI as any traumatic intracranial finding on CT except for isolated linear skull fractures. These ICIs could be any of the following: an intracranial hemorrhage or contusion, cerebral edema, traumatic infarction, diffuse axonal injury, shearing injury, sigmoid sinus thrombosis, midline shift of intracranial contents or signs of brain herniation, diastasis of the skull, pneumocephalus, and depressed skull fracture at least the width of the table of the skull.

We collected clinical data, including patient age, sex, mechanism of injury, history of loss of consciousness, history of vomiting, and initial presenting GCS score. Severe mechanism of injury was defined by motor vehicle crash with patient ejection, death of another passenger, or rollover; pedestrian or bicyclist without helmet struck by motorized vehicle; falls greater than 3 ft for children younger than 2 years and greater than 5 ft for children 2 years and older; or head struck by a high-impact object. The CT scans that showed IVHs were reviewed to describe location and number of ventricles containing hemorrhages.

The primary outcome of interest was an acute adverse event defined as death from TBI, neurosurgical intervention, or intubation for greater than 24 hours for management of the head injury. Patients who were hospitalized were followed up to determine outcomes; those discharged to home from the ED received telephone or mail follow-up to determine subsequent CT imaging, hospitalization, or neurosurgical intervention for a TBI. For those not available by telephone or mail follow-up, we reviewed the patients' medical records, ED quality improvement records, trauma registry records, and the county morgue records to identify any patients who were lost to telephone or mail follow-up and who may have had a clinically important TBI. The secondary outcome of interest was patient functional status at hospital discharge, which was determined by information collected during medical record review, using the POPC and the PCPC scores.¹⁵These scoring systems use a 6-point ordinal scale rating functional outcomes from 1 (normal; good overall performance) to 6 (brain death). We defined poor functional outcome as a score of 3 to 6 (ie, moderate or severe disability, coma or vegetative state, or brain death).^5,16

We described the study population using simple descriptive statistics, with 95% CIs where appropriate. We examined the association of each clinical presentation and outcome variable with the isolated and nonisolated IVH groups using the Fisher exact test. We calculated the rate differences of acute adverse and functional outcome events and their associated 95% CIs between the isolated and nonisolated IVH groups. We performed the data analysis using SAS statistical software, version 9.2 (SAS Institute, Inc).

Of 43 904 patients enrolled in the primary study, 43 398 met inclusion criteria for the current analysis. A total of 15 907 patients (36.7%) had CTs performed, and 1156 (7.3%) had ICIs apparent on CT, including 53 (4.6%; 95% CI, 3.5%-6.0%) with IVHs, of whom 10 (0.9%; 95% CI, 0.4%-1.6%) had isolated IVHs. Therefore, of the 15 907 patients who underwent CT, the prevalence of any IVH was 53 of 15 907 (0.3%; 95% CI, 0.2%-0.4%) and the prevalence of isolated IVH was 10 of 15 907 (0.06%; 95% CI, 0.03%-0.12%). All subsequent analyses are limited to the 53 children with IVHs on CT.

We compared the clinical characteristics at ED presentation between the patients with isolated and nonisolated IVHs (Table 1). Children with isolated IVHs were more likely to have GCS scores of 14 to 15, whereas children with nonisolated IVHs were more likely to present with a history of loss of consciousness. No difference was found between the 2 groups in terms of sex, history of vomiting, and severity of the mechanism of injury.

Descriptions of the radiographic findings of the CT scans in the isolated and nonisolated IVH groups are provided in Table 2. Of the 43 patients with nonisolated IVHs, 23 (53.5%) had more than 1 ventricle with hemorrhage identified on CT compared with 3 of the 10 patients (30.0%) with isolated IVHs. Among the patients with isolated IVHs, the IVH was located in at least 1 of the lateral ventricles in 7 of the 10 patients (70.0%) compared with 23 of the 43 patients (53.5%) with nonisolated IVHs.

Patients with isolated IVHs had better outcomes than those with nonisolated IVHs. Table 3 details the acute adverse and functional outcomes in both patient groups. Patients with nonisolated IVHs were more likely to have intubation for more than 24 hours (rate difference, 55.8%; 95% CI, 21.7%-82.0%), neurosurgical intervention (rate difference, 41.9%; 95% CI, 7.1%-73.7%), or die due to their head injury (rate difference, 37.2%; 95% CI, 2.3%-67.2%).

Patients with isolated IVHs were also less likely to have poor functional outcomes than patients with nonisolated IVHs as assessed by both POPC and PCPC functional outcome scores. Only 1 of the 10 patients (10.0%) with isolated IVHs had a poor functional outcome based on the POPC score compared with 27 of 43 patients (62.8%) with nonisolated IVHs (rate difference, −52.8%; 95% CI, −81.3% to −19.1%). One patient with an isolated IVH with a poor functional outcome based on the POPC score had a functional outcome of moderate overall disability. None of the patients with isolated IVHs had poor functional outcomes based on the PCPC score compared with 25 of 43 patients (58.1%) with nonisolated IVHs (rate difference, −58.1%; 95% CI, −84.2% to −24.3%).

In this large, prospective series of children with BHT, we found the presence of IVH on CT to be an uncommon occurrence, with outcomes greatly predicated on whether the IVH was associated with other ICIs on CT. Among the patients with IVHs, 18.9% had isolated IVHs, and they typically presented with normal GCS scores and had good clinical and functional outcomes. This finding is in contrast to children with nonisolated IVHs, who presented to the ED predominantly with GCS scores of less than 14 and frequently experienced poor acute clinical and functional outcomes.

Few prospective, observational studies of patients (either children or adults) with IVHs exist. Most previous studies are case reports or case series of both adults and children with IVHs. In a prospective, observational cohort study of adults and children with BHT, Atzema et al¹⁰ described a 1.4% prevalence of IVH among those undergoing cranial CT. In that study, 7 (10%) of the patients were children. LeRoux et al¹⁷ also described 43 patients with traumatic IVHs, but that was an older case series in which the prevalence was reported as 2.8% of those who underwent CT, and only 8 (19%) of the patients were children. In the present study, of the children undergoing cranial CT after BHT, the prevalence of IVH was the lowest reported of any study and was 21.4% of the prevalence that was described by Atzema et al.¹⁰ However, this finding is not unexpected because our study included only children, in whom traumatic IVHs are rare, compared with adults and elderly patients. No lesions (eg, vascular malformations) were identified as an underlying cause of IVH in our cohort of children experiencing head trauma as may be expected in elderly patients.

Patients with IVHs often have serious sequelae. In adults, traumatic IVH is associated with a mortality rate of 22% to 62.5%.^13,18,19 Children with posttraumatic IVHs may experience equally serious outcomes, including death.^11,12,20 However, to our knowledge, no previous prospective, pediatric, observational cohort studies of traumatic IVH exist; therefore, the true prevalence and outcomes have been unknown. In one case series, 2 of 4 children with posttraumatic IVHs died, and 1 child was left in a persistent vegetative state after severe lateral impact.¹¹ Our experience was similar because more than 50% of children with nonisolated IVHs in the present study had severe disability or death.^10,17 In our series, more than 72.1% of children with nonisolated IVHs presented to the ED with GCS scores of 8 or lower. Clinical, pathologic, and radiologic studies of traumatic IVH implicate brainstem and other cerebral parenchymal contusion or hemorrhage as the source of ventricular system blood, and this finding would support the severe clinical presentations and adverse outcomes in children with IVHs and evidence of other ICIs on CT.^7,17,21

The few reports of isolated IVH after head trauma in children demonstrate that children with isolated IVHs have different clinical findings and prognoses than those with IVHs in association with ICI on CT.^10,13 Similar to our findings, presenting GCS score and prognosis have been described to be related to the severity of the head trauma and presence of other ICIs on CT.^10,17,19 Atzema et al¹⁰ reported that children and adults with isolated IVHs who appeared clinically well at ED presentation had good functional outcomes. In addition, IVH has been described as a low-risk CT finding in children and adults when no signs of clinical deterioration are present.^18,22 Isolated IVH may simply be secondary to vascular disruption localized to the ventricular system itself. Therefore, the degree of ICI may be limited, which is consistent with better clinical presentations and outcomes.

Our findings differ from previous investigations in several regards. To our knowledge, this is the only purely pediatric cohort study of IVH after BHT. Most existing reports^10,13,17,18 describe primarily adult patients with IVH and are mostly case series rather than cohort studies. The large study by Atzema et al¹⁰ only contained 7 patients with IVH who were younger than 10 years of age.²³ Therefore, it is difficult to extrapolate the prevalence and outcomes of pediatric traumatic IVH from preexisting research studies. In other studies involving adults, poor outcome was associated with abnormal GCS scores at presentation, as well as with CT findings of deep-seated intraparenchymal hemorrhage and/or IVHs, which is similar to our findings with children.^12,24 In addition to specific CT findings, the mechanisms of injury and clinical symptoms of the patients in the present study were more varied compared with previous studies^11,12 that included predominantly patients with severe clinical presentations. Although 2 patients with isolated IVHs presented with substantially decreased levels of consciousness in another study, all the patients with isolated IVH in our study had GCS scores of 15.¹² This finding may also reflect the difference between case reports or case series rather than a large observational cohort such as that reported in our study. Furthermore, there were no deaths among the patients with isolated IVHs in our study, although of note, the 10 patients with isolated IVHs in our study had mechanisms of injury (motor vehicle related and falls) similar to those with nonisolated IVHs.

We also provide new information regarding functional outcomes of children with posttraumatic IVHs. Using the POPC and PCPC scores, we documented good functional outcome at hospital discharge of children with isolated IVHs and frequent poor functional outcome of those with IVHs in association with other ICIs on CT.

To our knowledge, this is the largest prospective cohort study to date describing children with IVHs after BHT. Our large sample size made it possible to separate children with isolated IVHs from those with nonisolated IVH and to demonstrate differences in functional outcomes between these 2 groups. Children with isolated IVHs appear to represent a distinct population compared with what is generally described in patients with traumatic IVHs. Children with isolated IVHs had normal GCS scores at presentation and had either normal functional outcomes or mild disability at hospital discharge. These data highlight that children with isolated IVHs have less severe outcomes than previously ascribed to the overall group of children with traumatic IVHs.

This study has several limitations. Most data were collected prospectively; however, to evaluate functional outcomes, we retrospectively reviewed medical records of children with IVH at the time of hospital discharge to obtain POPC and PCPC scores. It is possible that subtle functional abnormalities were missed and that functional status could worsen after hospital discharge. In addition, despite a large sample size in the parent study, there were only 53 children with IVHs, reflecting the infrequency of this finding and limiting the power to make statistical inferences when comparing those with isolated IVHs with those with nonisolated IVHs. However, these data represent the largest number of pediatric patients with posttraumatic IVHs studied in a prospective fashion to date. Finally, CT scans were interpreted by individual site faculty radiologists, and findings were not validated independently. Equivocal CT scans, however, were adjudicated by a senior faculty pediatric radiologist at the primary study site.

In conclusion, IVHs are uncommon in children after BHT. However, children with IVHs in association with other ICIs after BHT typically present with GCS scores of less than 14, frequently require neurosurgery, and often die or have functional disabilities at hospital discharge. In contrast, children with isolated IVHs typically present with normal GCS scores and have good clinical and functional outcomes. Therefore, children with isolated IVHs after BHT may be candidates for early hospital discharge or even extended ED observation for clinical signs of deterioration rather than definite hospital admission.

Correspondence: Richard Lichenstein, MD, Department of Pediatrics, University of Maryland School of Medicine, 22 S Greene St, Baltimore, MD 21201 (rlichenstein@peds.umaryland.edu).

Accepted for Publication: December 20, 2011.

Published Online: April 2, 2012. doi:10.1001/archpediatrics.2011.1919

Author Contributions:Study concept and design: Lichenstein, Holmes, and Kuppermann. Acquisition of data: Lichenstein, Glass, Quayle, Wootton-Gorges, Badawy, Atabaki, Holmes, and Kuppermann. Analysis and interpretation of data: Lichenstein, Miskin, and Kuppermann. Drafting of the manuscript: Lichenstein and Kuppermann. Critical revision of the manuscript for important intellectual content: Glass, Quayle, Wootton-Gorges, Wisner, Miskin, Muizelaar, Badawy, Atabaki, and Holmes. Statistical analysis: Lichenstein, Miskin, and Kuppermann. Obtained funding: Kuppermann. Administrative, technical, and material support: Lichenstein, Quayle, Wisner, Muizelaar, Badawy, Holmes, and Kuppermann. Study supervision: Lichenstein, Holmes, and Kuppermann.

Funding/Support: This work was supported by grant R40MC02461 from the Health Resources and Services Administration/Maternal and Child Health Bureau, Division of Research, Education, and Training, and the Emergency Medical Services for Children Program. The PECARN is supported by cooperative agreements U03MC00001, U03MC00003, U03MC00006, U03MC00007, and U03MC00008 from the Emergency Medical Services of Children program of the Health Resources and Services Administration/Maternal and Child Health Bureau.

Role of the Sponsors: The sponsors are not responsible for the design and conduct of the study; in the collection, analysis, and interpretation of the data; and in the preparation, review, or approval of the manuscript.

Previous Presentations: Presented in part at the Pediatric Academic Societies Annual Meeting; May 4, 2009; Vancouver, British Columbia, Canada; and the Society for Academic Emergency Medicine Meeting; May 15, 2009; Washington, DC.

Additional Contributions: We acknowledge the efforts of the following individuals participating in PECARN at the time this study was initiated. PECARN Steering Committee: Nathan Kuppermann, MD, MPH (chair); Elizabeth Alpern, MD, MSCE, James Chamberlain, MD, J. Michael Dean, MD, MBA, Michael Gerardi, MD, Julius Goepp, MD, Marc Gorelick, MD, MSCE, John D. Hoyle Jr, MD, David Jaffe, MD, Christina Johns, MD, Nadine Levick, MD, MPH, Prashant Mahajan, MD, MPH, MBA, Ronald Maio, DO, MS, Kraig Melville, MD, Steven Miller, MD,† David Monroe, MD, Richard Ruddy, MD, Rachel Stanley, MD, MHSA, David Treloar, MD, MS, Michael Tunik, MD, Allen Walker, MD. Maternal and Child Health Bureau/Emergency Medical Services for Children Program liaisons: Dan Kavanaugh, MSW, LCSW-C, Hae Young Park, MPH. Central Data Management and Coordinating Center: J. Michael Dean, MD, MBA, Richard Holubkov, PhD, Stacy Knight, MStat, Amy Donaldson, MS. Data Analysis and Management Subcommittee: James Chamberlain, MD (chair); Morton Brown, PhD, Howard Corneli, MD, Julius Goepp, MD, Richard Holubkov, PhD, Prashant Mahajan, MD, MPH, MBA, Kraig Melville, MD, Ernest Stremski, MD, MBA, Michael Tunik, MD. Grants and Publications Subcommittee: Marc Gorelick, MD, MSCE (chair); Elizabeth Alpern, MD, MSCE, J. Michael Dean, MD, MBA, George Foltin, MD, Jill Joseph, MD, PhD, MPH, Steven Miller, MD,† Frank Moler, MD, MS, Rachel Stanley, MD, MHSA, Stephen Teach, MD, MPH. Protocol Concept Review and Development Subcommittee: David Jaffe, MD (chair); Kathleen Brown, MD, Arthur Cooper, MD, MS, J. Michael Dean, MD, MBA, Christina Johns, MD, Ronald Maio, DO, MS, N. Clay Mann, PhD, MS, David Monroe, MD, Kathy Shaw, MD, MSCE, Daniel Teitelbaum, MD, David Treloar, MD, MS. Quality Assurance Subcommittee: Rachel Stanley, MD, MHSA (chair); Diana Alexander, MD, Jeri Burr, RNC, CCRC, CPCCRN, Michael Gerardi, MD, Mary Ann Gregor, DrPH, Richard Holubkov, PhD, Kathleen Lillis, MD, Brian Nordberg, BS, Richard Ruddy, MD, Mike Shults, MA, Allen Walker, MD. Safety and Regulatory Affairs Subcommittee: Nadine Levick, MD, MPH (chair); John Brennan, MD, Jeri Burr, RNC, CCRC, CPCCRN, J. Michael Dean, MD, MBA, John D. Hoyle Jr, MD, Ronald Maio, DO, MS, Richard Ruddy, MD, Walton Schalick, MD, PhD, Tasmeen Singh, MPH, NREMT-P, Joseph Wright, MD, MPH. We thank Rene Enriquez, BS, at the PECARN Data Center (University of Utah) and Sally Jo Zuspan, RN, MSN, for their 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. Participating centers and site investigators are listed in alphabetical order: Atlantic Health System/Morristown Memorial Hospital (Michael Gerardi, MD); Bellevue Hospital Center (Michael Tunik, MD, and James Tsung, MD); Calvert Memorial Hospital (Kraig Melville, MD); Children's Hospital Boston (Lois Lee, MD, MPH); Children's Hospital of Michigan (Prashant Mahajan, MD, MPH, MBA); Children's Hospital of New York–Presbyterian (Peter Dayan, MD, MSc); Children's Hospital of Philadelphia (Frances Nadel, MD); Children's Memorial Hospital (Elizabeth Powell, MD, MPH); Children's National Medical Center (Shireen Atabaki, MD, MPH, and Kathleen Brown, MD); Cincinnati Children's Hospital Medical Center (Todd F. Glass, MD, MS); DeVos Children's Hospital (John D. Hoyle Jr, MD); Harlem Hospital Center (Arthur Cooper, MD, MS); Holy Cross Hospital (Elizabeth Jacobs, MD); Howard County Medical Center (David Monroe, MD); Hurley Medical Center (Dominic Borgialli, DO, MPH); Medical College of Wisconsin/Children's Hospital of Wisconsin (Marc Gorelick, MD, MSCE, and Subhankar Bandyopadhyay, MD); St Barnabas Health Care System (Michael Bachman, MD, and Neil Schamban, MD); SUNY Upstate Medical Center (James Callahan, MD); University of California–Davis Medical Center (Nathan Kuppermann, MD, MPH, and James F. Holmes, MD, MPH); University of Maryland (Richard Lichenstein); University of Michigan (Rachel Stanley, MD, MHSA); University of Rochester (Mohamed Badawy, MD, and Lynn Babcock-Cimpiello, MD, MPH); University of Utah/Primary Children's Medical Center (Jeffrey Schunk, MD); Washington University, St Louis Children's Hospital (Kimberly S. Quayle, MD, and David Jaffe, MD); and Women and Children's Hospital of Buffalo (Kathleen Lillis, MD).

†Deceased.