LAUSR

Abstract Background Bipolar Disorder (BD) is a major psychiatric illness defined by episodic mood changes, which in approximately 50% of cases is associated with psychotic features. Over the past decades, a large amount of research has identified brain structural and functional alterations in patients with this mental disorder. Some findings have been found to be specific to patients with psychotic symptoms, raising suggestions that this could represent a biological subtype of the disorder. Recent interest has been addressed to Early-Onset Bipolar Disorder (EOBD, onset prior to age 18). Latest reviews in EOBD samples have pointed to abnormalities in the frontal lobe and limbic structures, with some inconsistencies in the reported results possibly caused by differences in the methodology. In addition, no study so far has examined the neural structural correlates of psychotic symptoms in adolescent-onset bipolar disorder (AOBD). The aim of the present study is to examine the impact of psychosis on the neurobiological architecture in a sample of patients with AOBD. To our knowledge, this is the first study comparing gray matter structure between AOBD patients with or without psychotic features. Methods We conducted a cross-sectional study collecting T1-weighted structural magnetic resonance neuroimaging (3T-MRI) data in patients diagnosed with Bipolar Disorder type I or II between 12 and 19 years old (N=46, mean age (SD)=15.89 (1.94), gender=52.2% females). All patients were recruited from child and adolescent mental health services of the Hospital Clinic of Barcelona, Spain. Diagnoses were confirmed with a semi-structured clinical interview (Kiddie-Sads present and lifetime version) by child and adolescent psychiatrists. Images were pre-processed employing FreeSurfer 5.3.0, and data corresponding to Cortical Thickness (CTH) and Subcortical Volumes (SCV) was obtained. Groups were compared according to whether patients had experienced psychotic symptoms at any point during their illness: Non-Psychotic Bipolar Disorder (NPSBD, N=25) and Psychotic Bipolar Disorder (PSBD, N=21). No differences in age (t=0.498, p=0.621) or sex (χ2=0.001, p=0.979). Group effects in relation to both CT and SCV were examined with a general linear model. The main effect of group on CTH and SCV, was performed for the whole brain, performing a correction for multiple comparisons (Montecarlo correction, threshold = 0.05). Results Between-group analyses showed smaller CTH in a cluster in the left medial orbitofrontal cortex (cluster size= 1142.58 mm2) in PSBD relative to NPSBD (x, y, z: 25.63, 89.61, -42.74; p=0.002). In addition, we observed a smaller right hippocampus volume (p=0.025) in PSBD relative to NPSBD. No other statistically significant differences were obtained. Discussion PSBD showed smaller cortical thickness in the left medial orbitofrontal cortex, as well as a volumetric reduction in the right hippocampal volume. Similar results have been reported in a study comparing adolescent patients with psychotic BD and healthy controls. These results add evidence about the role of these two structures in the genesis of psychotic symptoms in a population diagnosed with AOBD. Interestingly, one study has reported a surface area decreased of the orbitofrontal cortex in adolescent patients with a non-bipolar psychotic disorder, which suggests that they may be a common substrate to psychotic symptoms during adolescence regardless of co-occurring affective symptoms. In summary, this study points to the existence of a distinct biological nature between bipolar patients according to psychotic symptoms, underpinned by a different neurobiological architecture. Future research should focus on replication and on examining the clinical value of this finding.