To make certain that the results from transection experiments are correctly interpreted and to minimize the variability in results, it is important that transection methods be standardized and that control animals be prepared at the same time that the experimental animals are treated. For example, to ensure that the recovery of function is due to axonal regeneration and not spared spinal cord circuitry, researchers must precisely perform transections of the spinal cord and must be sure that the axons projecting from the neurons are completely severed. If not all of the axons are severed, sparing and sprouting from uninjured axons become issues. It is important to note that damage to the dura mater as a result of a penetrating injury (including experimental transection) provides a route for the invasion of fibroblasts into the injury site (Zhang et al., 1996, 2004). Furthermore, in mice, there is extensive invasion of fibroblasts even without damage to the dura and the fibroblasts participate in the formation of a tissue matrix that is supportive for regeneration of at least some types of CNS axons. Following penetrating injuries, the potential contribution of fibroblasts (positive or negative) must be considered in evaluating experimental interventions to promote repair and functional recovery
By virtue of the means by which compression injuries occur, there is a large amount of variability in the severities of spinal cord injuries. However, when initial compression studies are performed, it is important to be able to study a large population of animals that have the exact same initial injury characteristics before the experimental therapeutic intervention. Protocols have been developed to help minimize the variability in injury from animal to animal. Three impactors are widely accepted as standard methods for the delivery of contusion injuries to rodents: the Ohio State University (OSU) impactor, the Infinite Horizons device, and the Multicenter Animal Spinal Cord Injury Study (MASCIS) impactor (Bresnahan et al., 1987; Noyes, 1987; Kwo et al., 1989; Gruner, 1992; Young, 2002).
Although it is important to test therapeutic interventions in animals before they can become established treatments in the clinic, genetic differences between animal species can potentially result in different responses to spinal cord injuries or treatments. For example, in response to injury, humans and rats develop a cavity in the spinal cord, but this does not occur in mice (although the precise cellular and molecular bases for this are not yet