Quantitative MR analysis of caudate abnormalities in pediatric ADHD: Proposal for a diagnostic test

Abstract 

Most morphometric magnetic resonance imaging (MRI) studies of pediatric attention-deficit/hyperactivity disorder (ADHD) with appropriate sample sizes reveal a decreased right caudate nucleus volume. Recently, our group reported that this decrease is mainly due to a diminished right caudate body volume (rCBV). Here, we hypothesize that, employing either the total bilateral caudate volume (tbCV) or the bilateral caudate body volume (bCBV) as scaling variables, the rCBV/tbCV and rCBV/bCBV ratios could be found diminished and used as a basis of an imaging diagnostic test. Volumetric caudate nucleus data were obtained from a case-control morphometric MRI study with 39 ADHD subjects and 39 handedness- and IQ-matched controls, using a novel semi-automated caudate segmentation procedure. Student t-tests comparing each relevant ratio were conducted between the two samples. After splitting the samples into two groups, a receiving operator characteristic (ROC) analysis was conducted on the training group to determine the optimal cut-off. Its performance was then examined on the test group. The rCBV/bCBV ratio was found to be statistically different. For a value equal or inferior to 0.48, the specificity was 95.00%. We propose using the rCBV/bCBV ratio to assist in the diagnosis of ADHD in children.

Keywords: Attention-deficit/hyperactivity disorder, Magnetic resonance imaging, Diagnostic test, Diagnostic accuracy, Receiver operating characteristic analysis, Caudate nucleus