Testing the estrogen hypothesis of schizophrenia: Associations between cumulative estrogen exposure and cerebral structural measures

BACKGROUND Bone mineral density (BMD), as an indicator of cumulative estrogen exposure, may be reduced in female patients with psychotic disorder (van der Leeuw et al., 2013), possibly reflecting reduced cerebral exposure to estrogen and alterations in neuroprotective effects. To the degree that BMD is a marker of cumulative (endogenous) estrogen exposure, we hypothesized that BMD would be positively associated with cerebral gray and white matter indices. METHODS Dual X-ray absorptiometry (DEXA) and magnetic resonance (MRI) scans were acquired in fourteen female patients diagnosed with a psychotic disorder. BMD was expressed in total BMD (g/cm(2)), Z- and T-scores. Cerebral cortical thickness (CT) (as indicator of gray matter status) and fractional anisotropy (FA) (as indicator of white matter integrity) were measured and served as the dependent variables in multilevel random regression models. BMD measures were the independent variables. RESULTS Femoral BMD measures were positively associated with CT at trend significance (total BMD: B=0.266, 95% CI: -0.019-0.552, p=0.067; Z-score: B=0.034, 95% CI: 0.001-0.067, p=0.046; T-score: B=0.034, 95% CI: 0.000-0.068, p=0.052). There were no significant associations between femoral BMD measures and FA. CONCLUSIONS The data suggest that in women with psychotic disorder, alterations in the neuroprotective effect of estrogen (as measured by BMD) impact cortical gray matter, but not white matter integrity. These findings merit further investigation and, if replicated, would lend support to the estrogen hypothesis of schizophrenia.