Assaying structural changes after nerve damage, an essay on quantitative morphology

Morphologic changes, such as cell loss or sprouting after nerve lesions, presumably have an important bearing on chronic pain states. Accordingly, unbiased estimates of such things as cell or synapse numbers are necessary to evaluate these difficulties. Basic principles for obtaining such estimates are in dispute, however, so the present paper is an essay on these techniques with the aim of making it easier for non-devotees of these methods to make accurate assessments when such data are reported. The basic points are that, with the exception of serial reconstructions, which are very inefficient, older techniques to determine cell and synapse numbers from histological sections make geometric assumptions that are almost never verified or quantified. Accordingly, the geometrical biases that arise from these assumptions are not assessed. Fortunately, recent stereological techniques do not depend on these assumptions and so are free from the above biases. As such, these represent a technological advance. It must be remembered, however, that all methods are subject to potential observational, methodological and sampling biases.