Clinical Relevance of MRI Scanning in a Sample of 101 ASD Individuals: Evidences from EEG and MRI Findings

Background: EEG abnormalities are much frequent in ASD subjects, especially with Intellectual Disability(ID) (Volkmar and Nelson, 1990). On the contrary, the clinical relevance of MR scanning in individual with ASD is quite controversial. Moreover, little is still known about the relation between ID and MRI abnormalities in ASD. Brambilla et al. (2003) reviewed the structural MRI studies published from 1966 to May 2003: the most replicated abnormalities in ASD were increased total brain, parieto-temporal lobe, and cerebellar hemisphere volumes. Moreover, recent findings indicated that the size of amygdala, hippocampus, and corpus callosum(mkght be abnormal. Furthermore, ASD individuals might be characterized by abnormalities in neural netwozk involving fronto-temporo-parietal cortex, limbic system, and cerebellum. However, Filipek (1999) found that the prevalence of lesions detected by MRI in children with ASD has been reported to be similar to typically devmloping controls. Recently, a new study demonstrated structural atypicalities in the temporal cortex and adjacent cortical structures, associated with EEG abnormalities ( Ozlem et al.,2009). Thus, the debate on the clinical relevance of MR scanning in children with ASD is still open. New studies are necessary with sizeable and well-characterized groups of ASD individuals. Moreover, the co-occurrence of MRI and EEG abnormalities and ID might be of interest. 

Objectives: We aimed to evaluate the MRI findings and EEG and their relation with ID in a large group of individuals with non-syndromic  Autism Spectrum Disorder (ASD). 

Methods:  MRI inspection of 101 individuals with non-syndromic ASD (75M; age range: 2yrs 8 months- 23yrs and 8 moths; mean= 9 yrs 10 months; SD=51) was performed under pharmacological sedation. All met the DSM-5, ADOS and ADI –R criteria for autism. All MRI scans were acquired with a 1.5-T Signa GE (DW1,FSE T1, FSE T2, 3-DT1-FSPGR, Gd ev). EEG investigations were recorded according to the 10–20 system. The EEG was performed during sleep and it lasted1 hour. Unless the sleep was spontaneous, we used melatonin mg 5 per dose. All patients were evaluated for cognitive development with one of the following scales: Griffith Scales, Wechsler Intelligence Scale for Patients-Revised or Leiter-R.

Results:  The results indicated that 46/101 patients (45%) had abnormal EEG recordings. MRI abnormalities were detected in 47/101 patients (46%). The most frequent MRI abnormalities were bilateral periventricular leukomalacia 15/47 (32%), malformations of the brain 12/47 (25%), and mild cortical atrophy 10/47 (21%). 25/47 (53%) had abnormal EEG recordings together with abnormal MRI scans. ID was detected in 69/101 patients (68%). 31 of the 69 patients with ID (44%) had abnormal EEG recordings while 28/69 had abnormal MRI (40%). 15 of  the 69  patients with ID  (21% ) had abnormal EEG recordings and had abnormal MRI scans.

Conclusions:  The results highlight the clinical relevance of MRI in ASD evaluation, especially in patients with EEG abnormalities and in the ones with ID. The co-occurrence of EEG abnormalities, ID and MRI abnormalities is present in a considerable percentage of these patients and therefore needs to be specifically investigated.