Elsevier

Pediatric Neurology

Volume 36, Issue 4, April 2007, Pages 217-226
Pediatric Neurology

Original article
Multimodality Comparison of Neuroimaging in Pediatric Traumatic Brain Injury

https://doi.org/10.1016/j.pediatrneurol.2007.01.003Get rights and content

Traumatic brain injury is a common cause of death and disability in children; early neuroimaging has assumed an increasingly important role in evaluating the extent and severity of injury. Several imaging methods were assessed in a study of 40 children with traumatic brain injury: computed tomography (CT), T2-weighted magnetic resonance imaging (MRI), fluid-attenuated inversion recovery (FLAIR) MRI, and susceptibility-weighted imaging (SWI) MRI to determine which were most valuable in predicting 6-12 month outcomes as classified by the Pediatric Cerebral Performance Category Scale score. Patients were subdivided into three groups: (1) normal, (2) mild disability, and (3) moderate/severe disability/persistent vegetative state. T2, FLAIR, and SWI showed no significant difference in lesion volume between normal and mild outcome groups, but did indicate significant differences between normal and poor and between mild and poor outcome groups. Computed tomography revealed no significant differences in lesion volume between any groups. The findings suggest that T2, FLAIR, and SWI MRI sequences provide a more accurate assessment of injury severity and detection of outcome-influencing lesions than does CT in pediatric traumatic brain injury patients. Although CT was inconsistent at lesion detection/outcome prediction, it remains an essential part of the acute traumatic brain injury work-up to assess the need for neurosurgic intervention.

Introduction

Traumatic brain injury is a major cause of morbidity and mortality in children and is responsible annually for more than 500,000 emergency room visits, approximately 95,000 hospital admissions, 7000 deaths, and 29,000 permanent disabilities. The majority of children with traumatic brain injury present with minor head trauma and a single, significant clinical sign or symptom indicating underlying neuroimaging identifiable pathology has not been shown [1]. Predicting long-term outcome after traumatic brain injury remains unreliable, and efforts to do so have generated much controversy. Because few overt clinical signs have been demonstrated to correlate with the underlying neuropathology, clinicians increasingly rely on radiologic evidence for the assessment of the severity and extent of neuronal injury. Because decisions about patient management often hinge on such diagnostic information, it is important to obtain the most accurate depiction possible of neuronal damage following traumatic brain injury.

In the acute trauma work-up of traumatic brain injury patients, computed tomography is important for the rapid detection of extra-axial hemorrhage (e.g., subdural or epidural hematomas), acute hydrocephalus, fractures, or other intracranial lesions that may require acute neurosurgical intervention [1], [2], [3], [4], [5]. Magnetic resonance imaging appears to be very sensitive for intraparenchymal lesion detection in traumatic brain injury patients, but frequently is not easily obtainable acutely after injury [2], [6], [7]. Magnetic resonance imaging sensitivity is understood to be superior to computed tomography for intracranial evaluation. However, despite the newer technological improvements for both modalities, there have been no reports that have directly compared the two modalities as to their relative sensitivity and specificity in detecting injury and in predicting outcome in children. Initial studies performed in the 1990s compared computed tomography to magnetic resonance imaging in adults [2], [3], [4], [6], [7], [8], [9], [10], [11].

Each of the magnetic resonance imaging sequences used in the present study has unique detection capabilities. T2 and fluid-attenuated inversion recovery (FLAIR) are common sequences in magnetic resonance imaging evaluation of traumatic brain injury patients. T2 is a more standard technique, generally able to detect a wide range of intraparenchymal lesions including edema, infarction, demyelination, moderate to large hemorrhages, and other parenchymal lesions [12]. FLAIR, a more recent technique, may be more sensitive in detecting some lesions within the parenchyma, including edema, particularly near the cortex or ventricles [13], [14], [15]. It is unknown whether there is any significance of intraparenchymal edema leading to more adverse outcomes compared with hemorrhage. In recent studies, we have demonstrated that a new gradient echo magnetic resonance imaging technique known as susceptibility-weighted imaging is also very useful in detecting hemorrhagic lesions associated with diffuse axonal injury [16], [17]. This sequence, which utilizes the paramagnetic properties of deoxyhemoglobin, is more sensitive for detection of extravascular blood products [18].

The objective of the present study was to determine which of these methods of detection of pediatric traumatic brain injury was most accurate at predicting outcome.

Section snippets

Patient Selection and Data Collection

We studied 40 children and adolescents with traumatic brain injury admitted to Loma Linda University Children’s Hospital from March 2001 to September 2002. This was an inception cohort of patients with traumatic brain injury without sufficient neurologic recovery in the first day who warranted magnetic resonance imaging during hospitalization for additional neurologic assessment of their injury. We excluded children with (1) previous head injuries; (2) pre-existing neurological disorders; (3)

General Clinical Data

Patient ages ranged from 18 months to 18 years (mean age, 12 years). No statistical differences in age were detected between the outcome groups. Most patients were passengers or drivers involved in motor vehicle accidents (n = 23) or struck by a car (n = 14). In addition, one patient each had the following types of injury: sports-related injury, assault, and bicycle fall.

Imaging Modalities

All patients involved in this study had a computed tomographic scan within 24 hours and a magnetic resonance imaging study 7

Interpretation of Data

The present data indicate that the T2, FLAIR, and susceptibility-weighted imaging sequences provided more accurate assessment of injury severity in pediatric traumatic brain injury patients and more accurately predicted outcome. It was not surprising that lesion numbers and volumes were not significantly different in comparison of the normal group vs the mild group. Patients with normal and mild disability likely have relatively similar degrees of neuronal injury. However,

References (27)

  • S.A. Schutzman et al.

    Pediatric minor head trauma

    Ann Emerg Med

    (2001)
  • Z. Groswasser et al.

    Magnetic resonance imaging in head injured patients with normal late computed tomography scans

    Surg Neurol

    (1987)
  • D.H. Fiser

    Assessing the outcome of pediatric intensive care

    J Pediatr

    (1992)
  • A.B. Kelly et al.

    Head trauma: comparison of MR and CT—Experience in 100 patients

    AJNR Am J Neuroradiol

    (1988)
  • J.R. Hesselink et al.

    MR imaging of brain contusions: a comparative study with CT

    AJR Am J Roentgenol

    (1988)
  • R.A. Zimmerman et al.

    Head injury: early results of comparing CT and high-field MR

    AJR Am J Roentgenol

    (1986)
  • M. Wintermark et al.

    Admission perfusion CT: prognostic value in patients with severe head trauma

    Radiology

    (2004)
  • W.W. Orrison et al.

    Blinded comparison of cranial CT and MR in closed head injury

    AJNR Am J Neuroradiol

    (1994)
  • L.R. Gentry et al.

    Prospective comparative study of intermediate-field MR and CT in the evaluation of closed head injury

    AJR Am J Roentgenol

    (1988)
  • R.L. Mittl et al.

    Prevalence of MR evidence of diffuse axonal injury in patients with mild head injury and normal head CT findings

    AJNR Am J Neuroradiol

    (1994)
  • H.S. Levin et al.

    Magnetic resonance imaging and computerized tomography in relation to the neurobehavioral sequelae of mild and moderate head injuries

    J Neurosurg

    (1987)
  • T. Ogawa et al.

    Comparative study of magnetic resonance and CT scan imaging in cases of severe head injury

    Acta Neurochir

    (1992)
  • A.S. Smith et al.

    Magnetic resonance with marked T2-weighted images: improved demonstration of brain lesions, tumor, and edema

    AJR Am J Roentgenol

    (1985)
  • Cited by (77)

    • ACR Appropriateness Criteria® Head Trauma-Child

      2020, Journal of the American College of Radiology
    • Susceptibility-Weighted Imaging and Magnetic Resonance Spectroscopy in Concussion

      2018, Neuroimaging Clinics of North America
      Citation Excerpt :

      In a separate evaluation of 18 children with different severities of trauma, lesion volume, and to a lesser extent lesion number, were negatively correlated with neuropsychological outcomes and intelligence quotient 1 to 4 years after injury.22 Another SWI study in 40 children with TBI showed a significant difference in lesion extension between normal/mild outcome and poor outcome groups 6 to 12 months after injury using the Pediatric Cerebral Performance Category Scale score.36 However, in a different cohort of patients including mild, moderate, and severe TBI, Chastain and colleagues45 did not find SWI to be predictive using the Glasgow Outcome Scale score, although the data were heterogeneous and the length of follow-up was widely variable.

    View all citing articles on Scopus
    View full text