Hippocampal volume development in healthy siblings of childhood-onset schizophrenia patients

Abstract

Objective: Previous anatomic studies have established a reduction in hippocampal volume in schizophrenia, but few have investigated the progressive course of these changes and whether they are trait markers. In the present study, the authors examined hippocampal volumes in relation to age for patients with childhood-onset schizophrenia, their nonpsychotic healthy siblings, and healthy comparison subjects.

Method: Anatomic brain magnetic resonance scans were obtained in childhood-onset schizophrenia probands (N=89, 198 scans), their nonpsychotic full siblings (N=78, 172 scans), and matched healthy comparison subjects (N=79, 198 scans) between the ages of 10 and 29 years. Total, left, and right hippocampal volumes were measured using FreeSurfer software and analyzed using a linear mixed-model regression covarying for sex and intracranial volume.

Results: Childhood-onset schizophrenia probands had a fixed reduction in hippocampal volumes (total, left, and right) relative to both nonpsychotic siblings and healthy comparison subjects, whereas there were no significant volumetric or trajectory differences between nonpsychotic siblings and healthy comparison subjects.

Conclusions: Fixed hippocampal volume loss seen in childhood-onset schizophrenia, which is not shared by healthy siblings, appears to be related to the illness. Decreased hippocampal volume is not strongly genetically related but represents an important intermediate disease phenotype.

FIGURE 1. Longitudinal Trajectories (Slopes) of Total, Left and Right Hippocampal Volumes in Childhood-Onset Schizophrenia Probands, Their Healthy Siblings, and Healthy Comparison Subjects^a

^aThe graphs depict group-by-age interaction effects. Images represent the progression of hippocampal volume for the middle 80% of the data range, from age 12 to 24 years. Mixed-model linear regression for the total, left, and right hippocampal volumes are shown. Pairwise group differences in total hippocampal volume (top) at the centered age (17.58 years [SD=4.6]) were statistically compared using t tests. Statistically significant differences in volume were found when comparing the childhood-onset schizophrenia group with healthy comparison subjects (t=5.53, df=229.90, p<0.001) and the childhood-onset schizophrenia group with healthy siblings (t=−4.96, df=236.40, p=0.004). There were no statistically significant group differences in slope for any of the volume measures. Pairwise group differences in left hippocampal volume (center) at the centered age (17.58 years [SD=4.6]) were statistically compared using t tests. Statistically significant differences in volume were found when comparing the childhood-onset schizophrenia group with healthy comparison subjects (t=5.18, df=227.37, p<0.001) and the childhood-onset schizophrenia group with healthy siblings (t=−4.27, df=234.51, p=0.006). There were no statistically significant group differences in slope for any of the volume measures. Pairwise group differences in right hippocampal volume (bottom) at the centered age (17.58 years [SD=4.6]) were statistically compared using t tests. Statistically significant differences in volume were found when comparing the childhood-onset schizophrenia group with healthy comparison subjects (t=5.41, df=232.08, p<0.001) and the childhood-onset schizophrenia group with healthy siblings (t=−5.25, df=239.18, p=0.001). There were no statistically significant group differences in slope for any of the volume measures.