A Proton MRS Study of the Basal Ganglia and Dorsolateral Prefrontal Cortex in Adults with Autistic Spectrum Disorders

Background: There is emerging evidence that individuals with Autistic Spectrum disorders have abnormalities in neuronal integrity (and perhaps also in glutamate/glutamine(Glx)) as measured using 1H MRS. However findings have varied – possibly because some included people with co-morbid medical problems and/or mental retardation. Also, some included children, whereas other studied adults. For example we reported that young adults with ASD have a significant increase in amygdala hippocampal Glx, and medial frontal N-Acetyl Aspartate (NAA), Choline (Cho), and Creatine and phosphocreatine (Cre) (Murphy et al 2002, Page et al 2006). In contrast others reported a reduction in NAA and Glx in other regions (for example DeVito et al 2007). NAA is correlated with mitochondrial function and neuronal integrity, Cho is a measure of membrane synthesis and turnover, and Cre is a measure of cellular energy metabolism. Prior structural and functional imaging studies have reported abnormalities in the basal ganglia and dorso-lateral pre-frontal regions in people with Autistic Spectrum Disorders. However nobody has specifically investigated the neuronal integrity of those regions using single voxel 1H MRS. Objectives: Hence we compared the neuronal integrity and concentration of glutamate/glutamine (using 1H MRS ) in the basal ganglia and the dorsolateral pre-frontal cortex (together with a control (parietal) region), in non-learning disabled adults with autistic spectrum disorder and controls. Methods: We included 30 individuals with an ADI confirmed ASD (8 Asperger syndrome, 9 with Autism, and 11 with Atypical autism) and compared them to 20 age, gender and IQ matched controls. MR spectroscopy studies were performed with a 1.5T GE HDx MRI scanner (GE Medical Systems, Milwaukee, WI, USA) equipped with TwinSpeed gradients. Single voxel 1H MR spectra were acquired with a PRESS sequence with repetition time (TR) = 3000 ms and echo time (TE) = 30 ms. Voxels of interest (VOI) were positioned in the left medial parietal lobe (20x20x20 mm3) and left dorso-lateral prefrontal cortex (16x24x20 mm3), using previously described methods (Murphy et al 2002). A VOI was also positioned in the left basal ganglia (20x20x15 mm3), to include the head of the caudate, putamen, and internal capsule. MRS concentrations were corrected for VOI CSF and grey/white matter content. Results: Individuals with ASD had a significant reduction; 1) in basal ganglia concentration of Cre, NAA, and Glx, and; 2) in dorsolateral pre-frontal cortex Cre and NAA. No differences from controls were found in the parietal region. Conclusions: Adults with ASD have regionally specific abnormalities in neuronal integrity of brain regions previously implicated in the disorder, and perhaps also in the glutamate/GABA system. However, taken in combination with our prior work and that of others (Murphy et al 2002, Page et al 2006, DeVito et al 2007), our findings suggest that people with ASD do not simply have an increase, or decrease, in measures of neuronal integrity and Glx. Rather, they have a complicated mixture of both – depending upon the brain region investigated and the age of the person included.