Efficient and Unbiased Sampling of Nerve Fibers for Estimating Fiber Number and Size

Publisher Summary Nerve fibers may be present in unifascicular or multifascicular peripheral nerve trunks or in central nervous tracts; they may be myelinated or non-myelinated; they may be homogeneous or heterogeneous in size, shape, and spatial distribution within trunks/tracts; they may be viewed by light and/or electron microscopy. This chapter focuses on the efficient and unbiased sampling of cross-sectioned nerve fibers to estimate their number and/or size. Any properly randomized sampling scheme is unbiased, by definition. Efficiency depends on how random selection is performed and on the number of fibers chosen. However, a systematic random scheme is usually preferable to simple random sampling because it does not oversample some areas of nerve trunk/tract at the expense of others. Systematic sampling can be based on strips, squares, or sectors and only about 200 fibers per specimen need to be chosen. No special measuring equipment is required; indeed, its use can introduce bias and its benefits can be wasted by employing it to measure all the fibers in a complete nerve cross section instead of just a sample. The chapter reviews the principal sources of error involved in nerve fiber sampling and presents a recommendation of more efficient ways to select nerve fibers for counting and sizing. Various methods in which specimens can be sampled once the nerve trunks or tracts have been taken through the preliminary processing, sectioning, and staining procedures are also illustrated in the chapter.