Dung Hoang, Victor Zeng, Jothini Sritharan, Henk J M M Mutsaerts, Jan Petr, Rebekah Trotti, Elena Ivleva, Weiying Dai, Scot Hill, Elliot Gershon, Sarah Keedy, Brett Clementz, Carol Taminga, Godfrey Pearlson, Nicolas Bolo, Matcheri Keshavan, Paulo Lizano
Distinct perfusion patterns exist in PSD, with widespread hypoperfusion in BP and Biotype 1 and hyperperfusion associated with greater symptom severity, supporting perfusion as a potential biomarker to stratify PSD subtypes.
BACKGROUND: Cerebral perfusion abnormalities may contribute to the pathophysiology of psychosis spectrum disorders (PSD), including schizophrenia (SZ), schizoaffective disorder (SAD), and bipolar disorder with psychotic features (BP), yet large-scale investigations remain limited.
METHODS: Pseudo-continuous arterial spin labeling scans were acquired from 2412 SZ, 2278 SAD, 158 BP, 401 neurotypical controls (NC) and processed using ExploreASL and FreeSurfer to quantify resting-state cerebral blood flow (CBF) across total gray matter and 60 brain regions. General linear regression compared CBF between NC and diagnostic groups or biologically informed biotypes derived from neurophysiological clustering. Within patient groups, Spearman correlations assessed associations between cognitive/clinical measures and CBF. Analyses were corrected for multiple comparisons.
RESULTS: Compared to NC, patient groups showed significantly lower total gray matter CBF (PSD: Cohen's d=-0.124, p-adjusted=0.037; BP: d=-0.259, p-adjusted=0.004; the most severe biotype, Biotype 1: d=-0.222, p-adjusted=0.006). BP and Biotype 1 exhibited significant hypoperfusion in 25 regions spanning bilateral temporal, occipital, inferior frontal cortices, and subcortical structures (BP d= -0.21 to -0.33; Biotype 1 d=-0.19 to -0.26). No significant CBF-cognition associations were observed. Within PSD, higher total gray matter CBF correlated with greater psychotic symptom severity, while right superior temporal sulcus hyperperfusion was associated with greater psychotic, manic, and depressive symptoms.
CONCLUSIONS: Distinct perfusion patterns exist in PSD, with widespread hypoperfusion in BP and Biotype 1 and hyperperfusion associated with greater symptom severity, supporting perfusion as a potential biomarker to stratify PSD subtypes.