White matter microstructure in early onset Obsessive-Compulsive Disorder and Tourette Syndrome. A diffusion tensor imaging study in a population of drug-naïve children and adolescents with long-term clinical follow-up

Background and Objective Early onset obsessive-compulsive disorder (OCD) and Tourette syndrome (TS) are frequently associated conditions. Beside the evidence of their high epidemiological cross-prevalence supported by a common genetic liability (Huisman-van Dijk et al., 2016; Yu et al., 2015), little is known on the nature of their close relationship on a pathophysiological level. By analyzing white matter (WM) microstructure through diffusion tensor imaging (DTI), the present study aimed to characterize and compare primary pathophysiological changes in drug-naïve children and adolescents with OCD, TS, and TS+OCD. Methods Fifty-one participants (mean age 10.2  2.0 years), including N=10 with OCD, N=16 with pure TS, N=14 with TS+OCD, and 11 age-matched controls were studied cross-sectionally through 3T MRI. We performed tractography and extracted DTI metrics in five WM tracts of interest, i.e., the cortico-spinal tract (CST), the anterior thalamic radiations (ATR), the inferior longitudinal fasciculus (ILF), the corpus callosum (CC), and the cingulum. Relationship between DTI changes and clinical severity was examined through correlational analyses. A clinical follow-up at mean 7.6 years after MRI examination was performed to evaluate clinical outcomes and association to neuroimaging findings. Results Significant between-group differences emerged in DTI metrics, specifically in fractional anisotropy (FA), an index of myelination and organization of axon fibers (Johansen-Berg & Rushworth, 2009; Toga et al., 2006). All analyzed tracts of interest except for the cingulum revealed a differential microstructure at group comparisons. The OCD group showed decreased FA within CST, ATR, ILF, and CC in respect to controls. A negative correlation was found between obsessive-compulsive symptoms and FA values in OCD, indicating that more severe clinical phenotypes are likely underpinned by less organized WM. Compared to controls, TS and TS+OCD groups both displayed remarkably different correlates from OCD and opposite DTI changes, i.e., increased FA in CST, ATR, ILF, and CC. Moreover, TS and TS+OCD had comparable DTI changes within all the investigated WM tracts and FA showed negative correlation with tic severity, revealing a shared pattern of WM organization in TS/TS+OCD with inverse relationship to symptom expression. At follow-up, no significant associations were found between FA values at baseline and long-term outcomes. Substantial symptom remission was achieved in 58.3% of TS, 63.6% of TS+OCD, and 70% of OCD patients, although a significant proportion of patient developed additional psychiatric disorders such as anxiety or depression. Conclusion The study highlights differential white matter involvement in pediatric OCD as opposed to TS/TS+OCD. Compared to neurotypical population, children with TS/TS+OCD showed an early increase in axons, fiber density, and/or myelination in WM bundles linking the frontal, occipital, and temporal cortices with each other and with the thalamus. Conversely, children with OCD showed widespread reduced organization of callosal, temporo-occipital, and fronto-thalamic WM tracts. Correlational analysis suggests that DTI changes in TS may reflect a compensatory reorganization in response to the disease pathophysiology, while in OCD they may represent a marker of the overall disease severity deriving from delay or damage to white matter development. Confirmation of these possibilities awaits longitudinal studies. The observation of shared DTI correlates of TS and TS+OCD strengthens the concept that at least some forms of OCD are etiologically related to TS and might therefore be a variant expression of the same etiologic factors that are important for the expression of tics (i.e., TS+OCD as a peculiar subtype of TS). By characterizing and differentiating early-stage neural underpinnings of OCD and TS, future targeted and neuroimaging-informed interventions may be developed.