Controlling for total brain volume, this study found no significant difference in amygdala volumes of the CPS compared to the normal subjects. Within the CPS group, children with an affective/anxiety disorder had significantly larger left amygdala volumes compared to those with no psychopathology. Amygdala volumes, however, were unrelated to seizure variables including age of seizure onset, duration of illness, seizure frequency, and temporal involvement vs extra-temporal disease, as well as history of perinatal problems or family history of psychopathology.
Previous adult TLE studies reported amygdala volume reduction (7
) associated with duration of epilepsy but not with other seizure variables. It is possible that our study found no significant difference between the CPS and normal subjects in amygdala volumes because of the shorter duration of illness 3.6 (2.55) in the children compared to the adults in the other studies.
Although there have been no structural MRI studies to date on amygdala volumes in children with epilepsy, supporting our findings, recent studies reported enlarged amygdala volumes in children without epilepsy who have MDD (26
) and generalized anxiety disorder (27
), adults with refractory partial epilepsy with comorbid affective disorders (23
). Similar to our study, a recent study by Richardson et al. (45
) showed a significant positive relationship between amygdala volumes and depression in adolescents and adults with TLE. However, unlike our study these authors (45
) found a positive relationship of both right and left amygdala volumes with depression whereas our study found this association only for left amygdala volumes and in a mixed group of affective/anxiety disorders.
Despite larger left amygdala volumes in the CPS subjects with depression and anxiety disorder diagnoses, a left focus on EEG did not appear to drive this finding. In fact, as suggested by several researchers in both imaging (11
) and neuropathology studies (18
), these preliminary findings in children with CPS who have affective/anxiety disorder might reflect the effects of the neuropathology underlying depression and anxiety disorders on amygdala development.
Our findings, however, are unlike those of previous studies that, in keeping with neuropathologic findings of significant reduction of glial cells and glial/neuron ratio in the left amygdala of MDD patients (18
), found amygdala volume reduction in children with MDD (19
) and anxiety disorders (20
), as well as in adults with MDD (15
). The inclusion of children with both comorbid anxiety and depression in one group, the relatively small study sample size, and differences in amygdala anatomical boundaries across studies underscore the importance of replicating our findings on larger samples of CPS patients both with and without comorbid psychopathology.
Limitations of the present study include its exploratory nature with a retrospective analysis, the small sample size which precluded differentiating between the subgroup of children with depression and anxiety symptoms, a greater, albeit not significantly higher proportion of girls than boys, possible parental memory bias for seizure-related information, few CPS subjects with right epileptic activity, missing EEG data in 2 CPS subjects, and significant differences in the Full Scale IQ of the CPS and normal groups. Although the normal children in the study had high mean Full Scale IQ scores, to ensure that we do not remove illness effect, we did not control for IQ differences in the group comparisons. The study limitations underscore the need for replication of our findings.
With these limitations in mind, children with CPS and comorbid affective/anxiety disorders appear to have enlarged left amygdala volumes compared to CPS children with no psychopathology. Involvement of the left amygdala and lack of association with seizure variables imply that amygdala volumes might reflect the underlying neuropathology of the comorbid affective illness rather than the epilepsy. However, these findings need to be replicated in a prospective study of a larger sample of CPS patients both with and without depression and anxiety disorder to determine the role played by epilepsy and/or the on-going depression and anxiety disorders on amygdala development in pediatric CPS.