The frequency of MRI-detected brain abnormalities with first febrile seizures (FS) and their association with FS type and with specific features of complex FS were determined in a prospective study at the Pediatric Emergency Department of New York-Presbyterian Children’s Hospital, Columbia University, New York. MRI performed within 1 week of the first FS showed abnormalities in 12.6% of 159 children affected. The number and ratio of simple to complex FS was 105:54 or 2:1. Imaging abnormalities occurred in 11.4% with simple FS and 14.8% of complex FS (n.s.). Of 54 complex FS cases, those with both focal and prolonged FS (N=14, 26%) were more likely to have MRI abnormalities than simple FS cases. These included focal cortical dysplasia and gray matter heterotopia (known to be associated with seizures) and subcortical focal hyperintensities (=/> 5 mm) and delayed myelination (not typically associated with seizures). Focal hyperintensities, the most common abnormality with first FS, were more frequent in children with complex (N=7, 13%) compared to simple FS (N=l, 0.9%, p=0.001). In comparison, the NIH Study of Normal Brain Development found no brain abnormalities on baseline MRI scans. Brain abnormalities may be associated with a lower seizure threshold in febrile children, predisposing to the development of FS. The findings did not affect our clinical management of FS, and MRI is unnecessary in FS, without some other neurological indication. 
COMMENT. AAP Practice Parameters on the evaluation and treatment of first simple febrile seizures (1996) advise against investigation with MRI, CT, or EEG, MRI abnormalities are reported in patients with febrile status epilepticus (Scott RC et al, 2003), but studies following first FS are understandably lacking, having regard to the hazards of heavy sedation in infants. Most neurologists consider MRI and EEG are indicated in children with recurrence of complex FS, prolonged impairment of consciousness following a seizure, or abnormal neurological signs. In the above prospective study of first FS, the incidence of abnormal MRIs in simple FS is unexpected, given the presumed benign nature of the FS. Although the finding did not change the clinical management of FS in this institution, the brain lesions are important in our understanding of the mechanism of FS. Authorities are divided regarding inclusion of children with a history of birth injury as FS, and some have grouped such patients as having epilepsy (Livingston (1954) and Friderichsen and Melchior (1954)). In unselected series of patients, however, evidence suggests that the threshold to FS may be lowered by both inherited and acquired factors. In addition to the essential role of fever, birth injury or anoxia and structural cerebral pathology may be factors in etiology and should not negate the diagnosis of FS. (Millichap JG et al, 1960). The necessity to guard against selection bias by such arbitrary exclusions is stressed by Baumann RJ and others. 
In the Columbia University study, the proportion of complex FS (both focal and prolonged) is high (25%), accounting for the higher frequency of MRI abnormalities among this group. An 11% incidence of MRI abnormalities in patients with simple FS is a novel finding, however, and suggests that febrile seizures are less benign than generally assumed. Without some other neurological indication, MRI is not recommended in children with FS. In support of this conclusion, a recent report of the management of FS in an unselected series of 100 patients found no structural brain abnormality in CT performed in 18% and MRI performed in 4% of FS patients. Head CT was obtained in 6% (4/77) of simple FS cases (1 had mastoiditis) and 61% (14/23) of complex cases. In keeping with AAP guidelines, no simple FS patients were examined with MRI; 17% of complex FS received MRI and all were normal . The frequency of negative CT scans in FS studies is noteworthy, and more specific indications for neuroimaging in children with complex FS should be determined.