Bipolar disorder (BD) is associated with excess cardiovascular burden and anomalous cerebral blood flow (CBF). We examined coronary artery disease (CAD) and myocardial infarction (MI) polygenic risk scores (PRS) in relation to CBF among youth with BD. Participants (mean age 17.2 years; n = 63 BD, n = 41 healthy controls HC) underwent magnetic resonance imaging to quantify CBF using arterial spin labelling. CAD-PRS and MI-PRS were calculated using adult genome-wide association summary statistics. Covariate-adjusted voxel-wise analyses examined associations of CAD-PRS and MI-PRS with CBF in the combined sample, and within subgroups. Sex-stratified analyses and PRS-by-diagnosis interaction effects were also examined. Sensitivity analyses controlled for psychotropic medications, cardiovascular risk factors, and mood. Higher MI-PRS was associated with lower CBF in one occipital cluster (β = −0.44, p = 0.003) in the combined sample, and two frontal and one occipital cluster within BD (β range = −0.47 to −0.51, p range = 0.03 to <0.001). Higher CAD-PRS was associated with higher caudal CBF within HC (β = 0.41, p ≤ 0.001; n.s. after controlling for cardiovascular risk factors). PRS-by-diagnosis interactions revealed three significant clusters for MI-PRS (β range = −0.34 to −0.35, p range = 0.03 to <0.001) and two significant clusters for CAD-PRS (β range = −0.32 to −0.33, p range = 0.001 to <0.001). In post-hoc analyses, higher PRS was associated with lower CBF within BD and higher CBF within HC. Higher polygenic risk for cardiovascular disease was associated with lower regional CBF in youth with BD, convergent with prior findings in adults and divergent from healthy youth in the current study. Future longitudinal studies incorporating putative mediators of the observed findings are warranted. • Cardiovascular genetics are linked to cerebral perfusion in youth bipolar disorder. • Findings in bipolar youth are independent of traditional risk factors. • Early cardiovascular considerations are crucial in BD research and treatment.
Kulkarni et al. (Sun,) studied this question.