Chronic exposure to haloperidol and olanzapine leads to common and divergent shape changes in the rat hippocampus in the absence of grey-matter volume loss

Abstract

Background: One of the most consistently reported brain abnormalities in schizophrenia (SCZ) is decreased volume and shape deformation of the hippocampus. However, the potential contribution of chronic antipsychotic medication exposure to these phenomena remains unclear.

Method: We examined the effect of chronic exposure (8 weeks) to clinically relevant doses of either haloperidol (HAL) or olanzapine (OLZ) on adult rat hippocampal volume and shape using ex vivo structural MRI with the brain retained inside the cranium to prevent distortions due to dissection, followed by tensor-based morphometry (TBM) and elastic surface-based shape deformation analysis. The volume of the hippocampus was also measured post-mortem from brain tissue sections in each group.

Results: Chronic exposure to either HAL or OLZ had no effect on the volume of the hippocampus, even at exploratory thresholds, which was confirmed post-mortem. In contrast, shape deformation analysis revealed that chronic HAL and OLZ exposure lead to both common and divergent shape deformations (q = 0.05, FDR-corrected) in the rat hippocampus. In particular, in the dorsal hippocampus, HAL exposure led to inward shape deformation, whereas OLZ exposure led to outward shape deformation. Interestingly, outward shape deformations that were common to both drugs occurred in the ventral hippocampus. These effects remained significant after controlling for hippocampal volume suggesting true shape changes.

Conclusions: Chronic exposure to either HAL or OLZ leads to both common and divergent effects on rat hippocampal shape in the absence of volume change. The implications of these findings for the clinic are discussed.

Keywords: antipsychotic; hippocampus; magnetic resonance imaging; schizophrenia; shape; volume.

Fig. 1.

(a) Anatomical criteria for creation of the binary mask for tensor-based morphometry (TBM) and shape analysis of the rat hippocampus using a mean MR image of the entire dataset (n = 24 scans). The orientation of the displayed scans is in the coronal plane. (b) Three-dimensional rendering of the binary mask of the rat hippocampus. (c) Processing pipeline for automated TBM analysis.

Fig. 2.

(a). Image processing pipeline for elastic surface-based shape deformation analysis. (b) The geometric error (Euclidean distance) between the original surface and the registered surface. This error is calculated for every surface and the average error is projected on the average surface and displayed with a colour map.

Fig. 3.

Elastic shape deformation analysis reveals common and divergent effects of chronic haloperidol and olanzapine exposure on hippocampal shape metrics. Data shown are the mean inward and outward displacement vectors (mm), shown both at an exploratory threshold of p < 0.05 uncorrected for multiple comparisons and q = 0.05 FDR-corrected, to illustrate significant differences in hippocampal shape metrics between (a) vehicle and haloperidol; (b) vehicle and olanzapine and (c) haloperidol and olanzapine exposed rats. Directionality of statistically significant hippocampal shape deformations, either inward (blue) or outward (red), may be found by comparing the p value maps with the mean displacement vector maps in each treatment group comparison. A, Anterior; P, posterior; S, superior; I, inferior; L, left; R, right.

Fig. 4.

Normalization of hippocampal volume does not alter the pattern of significant voxels identified by elastic shape deformation analysis, suggesting true shape changes. Data shown are size and shape and shape only (volume normalized to 1) for (a) vehicle v. haloperidol; (b) vehicle v. olanzapine and (c) haloperidol v. olanzapine.

Fig. 5.

Post-mortem confirmation of the lack of hippocampal volume changes in rats chronically exposed to either haloperidol or olanzapine using the Cavalieri probe. Data shown are individual hippocampal volume (scatter plot with mean) in vehicle (V), haloperidol (H) and olanzapine (O) treated rats.