Cellular morphology of chronic spinal cord injury in the cat: analysis of myelinated axons by line-sampling.

A systematic line-sampling method is described for counting and mapping myelinated axons in transverse sections of the spinal cord. Its advantages over random sampling of small areas are considered. The technique was applied to quantifying experimental weight-drop contusion injuries of cat spinal cord, from several months to more than a year after injury. Contusion of the mid-thoracic cord with a 20 g weight dropped 20 cm was usually sufficient to produce chronic hindlimb paralysis whilst allowing the survival of significant numbers (40,000-110,000) of myelinated axons passing through the lesion site. The axons which survived were concentrated towards the pial surface. There was a proportionally greater loss of larger diameter axons, but this was independent of distance from the pia, indicating that at least two independent factors contribute to selective axonal death following injury, one related to depth within the cord, the other to axon diameter. Myelin sheath thickness was decreased from normal and this deficit also increased with depth. There was overlap in all these quantitative morphological characteristics between animals showing some recovery of hindlimb locomotion and those with maintained spastic paralysis at more than six months after injury. Effective locomotion was found to recover in some cases with the maintenance of a small proportion (5-10%) of the original axonal population, largely concentrated in a rim only 200-300 microns thick. Morphological correlates of paralysis in chronic injuries included severe reduction of axonal number, selective elimination of large fibers, and sustained dysmyelination. Any one or combination of these may be responsible for chronic paralysis in individual animals.