Unbiased Stereology in Schizophrenic Brains

Many cortical and subcortical regions in brains from people who had schizophrenia have been anatomically investigated. The symptoms are varied and different from person to person, and may include hallucinations, thought and language disorder, paranoia, psychosis, and problems with attention, working memory, auditory perception, response monitoring and inhibition. It is not surprising so many anatomical areas may be involved and have been studied. These regions have been examined for changes in volume, cell number and cell density. Some of the results from these studies have been contradictory and inconclusive.

In the paper Kreczmanski P, Heinsen H, Mantua V, Woltersdorf F, Masson T, Ulfig N, Schmidt-Kastner R, Korr H, Steinbusch HWM, Hof PR and Schmitz C, Volume, neuron density and total neuron number in five subcortical regions in schizophrenia, Brain, 2007, 130, 678-692, the authors used stereological techniques to compare well-matched groups of brains from schizophrenic patients to brains from controls, and demonstrate a pattern of ‘rather subtle neuropathological alterations’. Stereo Investigator software was used to operate various stereological probes estimating volume, cell number and cell density in the striatum and certain limbic structures as well as the cortical gray matter. The Cavalieri estimator was used to determine volumes, either on the live image or from a tracing of the anatomical area. The Optical Fractionator was used to estimate the number of neurons, and in one group of neurons the nearest neighbor probe was used to compare any tendency of the neurons to clump together. Virtual Slide, a montage-making module that is available for Stereo Investigator, was used to stitch fields together to illustrate the locations that were examined (The following figure shows montages of brain tissue from people who had schizophrenia, stereological estimates were done at higher magnification, scale bar = 23 mm).

The densities of cells in the regions examined were not different between schizophrenic patients and controls. The volumes of the cortical gray matter, putamen, and lateral nucleus of the amygdala, however, were smaller and the number of cells in the putamen, lateral nucleus of the amygdala, and caudate nucleus were fewer for the brains from the schizophrenic patients vs. the controls. No differences were found for the medial dorsal nucleus of the thalamus. Some of these results disagree with previous studies. The authors propose various reasons including smaller sample sizes, confounding pathologies and different stereological design. The authors say that defining the anatomical areas on thicker sections than one previous study may have resulted in better volume estimates. Another reason given for discrepancies is the greater number of counting spaces used in the present study (see table below).