Deafferentation in animals as a model for the study of pain: an alternative hypothesis.

The notion that post-deafferentation autonomy is a pain response is unsupported by the results of studies with neurotoxins. The selective massive destruction of a fiber system considered essential to normal nociception--unmyelinated primary afferent axons--prior to deafferenting nerve lesions did not stop or even significantly impede post-denervation DI despite massive evidence from humans and animals that pain following nerve lesions originates in the periphery and is generated by abnormal discharges in the injured nerve. In addition, when a reduction in abnormal impulse discharges of both large and small injured sensory axons could be inferred following neonatal sympathectomy, DI was not reduced in incidence or severity. This latter observation (1) provides further support for a dissociation between DI and pain, since any contribution of myelinated primary afferent axons to painful pathology probably was substantially reduced by sympathectomy and (2) suggests that DI also may be unrelated to non-painful sensory pathology attributable to abnormal activity in the thick-diameter fiber population. These findings and an evaluation of other relevant observations suggest that DI may not be a manifestation of deafferentation pain and perhaps this animal model for the experimental study of pain should be discarded. An alternative view of DI, reconcilable with known properties of this behavior, is that it reflects a proclivity in some species and circumstances to shed a functionally-impaired insensate appendage.