Stereologic analysis of the lateral geniculate nucleus of the thalamus in normal and schizophrenic subjects

Abstract

Reduction of volume and neuronal number has been found in several association nuclei of the thalamus in schizophrenic subjects. Recent evidence suggests that schizophrenic patients exhibit abnormalities in early visual processing and that many of the observed perceptual deficits are consistent with dysfunction of the magnocellular pathway, i.e. the visual relay from peripheral retinal cells to the two ventrally located magnocellular layers of the lateral geniculate nucleus (LGN). The present study was undertaken to determine whether abnormalities in cell number and volume of the LGN are associated with schizophrenia and whether the structural alterations are restricted to either the magnocellular or parvocellular subdivisions of the LGN. Series of Nissl-stained sections spanning the LGN were obtained from 15 schizophrenic and 15 normal control subjects. The optical disector/fractionator sampling method was used to estimate total neuronal number, total glial number and volume of the magnocellular and parvocellular subdivisions of the LGN. Cell number and volume of the LGN in schizophrenic subjects were not abnormal. Volume of both parvocellular and magnocellular layers of the LGN decreased with age. These findings do not support the hypothesis that early visual processing deficits in schizophrenic subjects are due to reduction of neuronal number in the LGN.

Figure 1

Representative contour drawings of the LGN showing parvocellular layers (black outlines) and magnocellular layers (solid black) from two normal comparison brains (Norm1, Norm2) and two schizophrenic brains (Schizo1, Schizo2).

Figure 2

Low power photomicrographs of the LGN. Three representative sections through the (A) anterior, (B) middle, and (C) posterior LGN are shown. The LGN is comprised of six cellular layers: two magnocellular layers (marked with asterisks) lying ventrally and four parvocellular layers located dorsally, as well as cell-sparse intralaminar layers and a cell-poor superficial layer that is ventral to the magnocellular layers. Note that not all four parvocellular layers were visible in all sections and that individual layers often appeared to be fused. Scale bar = 1 mm.

Figure 3

High power photomicrographs of (A) parvocellular and (B) magnocellular subdivisions of the LGN. Typical neurons are indicated by arrowheads; typical glial cells are indicated by small arrows. Scale bar = 10 μm

Figure 4

Graphs of total neuronal number, total glial number, and volume in the parvocellular layers (above) and magnocellular layers (below) in normal control subjects (Norm, open circles) and schizophrenic subjects (Schizo, filled circles).

Figure 5

Graphs of correlations between volume and age but not neuronal number of glial number and age in the parvocellular subdivision of the LGN. A similar correlation between volume and age was found for the magnocellular subdivision (not shown).