Hippocampus, amygdala, and basal ganglia morphometrics in children after moderate-to-severe traumatic brain injury Elisabeth A Wilde* PhD, Baylor College of Medicine, Houston, TX; Erin D Bigler PhD, Brigham Young University, Provo, UT; Jill V Hunter MD, Texas Children’s Hospital, Houston, TX; Michael A Fearing MA, Brigham Young University, Provo, UT; Randall S Scheibel PhD; Mary R Newsome PhD, Baylor College of Medicine, Houston, TX; Jamie L Johnson BS, Brigham Young University, Provo, UT; Jocelyne Bachevalier PhD, Emory University, Atlanta, GA; Xiaoqi Li MS; Harvey S Levin PhD, Baylor College of Medicine, Houston, TX, USA. *Correspondence to first author at Cognitive Neuroscience Laboratory, Baylor College of Medicine, 1709 Dryden Road, Ste 725, Houston, TX, 77030 USA. E-mail:
[email protected]
While closed head injury frequently results in damage to the frontal and temporal lobes, damage to deep cortical structures, such as the hippocampus, amygdala, and basal ganglia, has also been reported. Five deep central structures (hippocampus, amygdala, globus pallidus, putamen, and caudate) were examined in 16 children (eight males, eight females; aged 9–16y), imaged 1 to 10 years after moderate-tosevere traumatic brain injury (TBI), and in 16 individuallymatched uninjured children. Analysis revealed significant volume loss in the hippocampus, amydala, and globus pallidus of the TBI group. Investigation of relative volume loss between these structures and against five cortical areas (ventromedial frontal, superomedial frontal, lateral frontal, temporal, and parieto-occipital) revealed the hippocampus to be the most vulnerable structure following TBI (i.e. greatest relative difference between the groups). In a separate analysis excluding children with focal hippocampal abnormalities (e.g. lesions), group differences in hippocampal volume were still evident, suggesting that hippocampal damage may be diffuse rather than focal.
See end of paper for list of abbreviations.
294
Developmental Medicine & Child Neurology 2007, 49: 294–299
Although frontal and temporal areas have been considered most susceptible to focal injury following traumatic brain injury (TBI), deep central brain structures, such as the hippocampal/amygdalar complex and the basal ganglia, are also vulnerable to TBI-related injury. Structural neuroimaging studies of adult patients with TBI, using T2-weighted fast field echo (FFE),1 T2-weighted2 and susceptibility-weighted3 gradient echo imaging pulse sequences, have recently identified increased evidence for diffuse axonal injury and ‘microbleeds’ in the basal ganglia and medial temporal lobes. Similarly, functional neuroimaging has implicated deep central areas, with evidence of long-term reduction in cerebral blood flow,4 hypoperfusion,5 and decreased cerebral metabolic rate6 in the basal ganglia/striatum following TBI. Medial temporal lobe vulnerability is also suggested by positron emission tomography and single-photon emission computed tomography,7 with an association between brain injuryinduced abnormalities and neurocognitive deficits. Finally, proton magnetic resonance spectroscopy has revealed a significantly decreased N-acetylaspartate/choline ratio in the basal ganglia and hippocampus of patients with TBI.8 Quantitative magnetic resonance imaging (MRI) studies of adult patients with TBI have found hippocampal atrophy which correlated with injury severity as measured by the Glasgow Coma Scale score.9–12 However, only three studies have examined hippocampal volume in individuals experiencing TBI as children or adolescents,13–15 and only a single study has examined caudate volume in this population.14 Despite growing evidence of injury to the amygdala and the basal ganglia from various mechanisms, there has been no examination of the amygdala, globus pallidus, or putamen using quantitative brain imaging in pediatric patients with TBI. This absence is notable, given that these structures are essential components of neuronal pathways supporting aspects of cognition and behavior commonly impacted in children with TBI. For example, neuronal projections between the ventral striatum (caudate, putamen, nucleus accumbens) and frontal lobes may play a role in mediating cognitive control, behavioral inhibition, and other aspects of executive functioning. Similarly, connections between the amygdala and prefrontal/temporal areas may mediate intermodal association and goal-directed behavior. Connections between the hippocampus and amygdala may be integral to memory, learning, and emotional functioning. The relative vulnerability of these deep central structures in comparison with cortical areas following TBI has not been reported in quantitative MRI literature. Therefore, the aims of this study were to: (1) compare volumetric measurements of the hippocampus, amygdala, globus pallidus, putamen, and caudate in children following moderate-to-severe TBI with demographically matched, typically developing (TD) children; (2) compare the relative volume loss in these structures with volume loss in cortical areas; and (3) examine whether volumetric changes in these structures are related to focal injury as opposed to diffuse or secondary injury. Method PARTICIPANTS
The TBI group consisted of 16 children (eight males, eight females) with moderate-to-severe TBI (initial Glasgow Coma Scale score 3–11) resulting from motor vehicle, bicycle, or pedestrian versus vehicle accident. Mean post-injury interval
was 3 years (SD 2y 5mo; range 1–10y), and mean age at time of scanning was 12 years 10 months (SD 2y 6mo; range 9–16y 9mo). Inclusion criteria also included having an Abbreviated Injury Scale16 score less than 4 for areas of the body other than the head, and no history of sustained (
