[Endogenous analgesic mechanism: new concepts from functional neuroanatomy, neurophysiology, neurobiology and chaos research.].

Modern concepts of pain therapy involve neuronal mechanisms of endogenous analgesia. Recent animal experiments have provided new insights into the anatomy, physiology and neurobiology of endogenous antinociception. We have shown that antinociception can be maximally activated by disinhibition-and not by direct electrical or chemical excitation-in the midbrain periaqueductal grey matter. This disinhibition is a likely mechanism of opioid analgesia. 'Purely analgesic' stimulation produces a very distinct pattern of activated neurons within the periaqueductal grey matter and other areas of the brain stem, as revealed by the expression of the nuclear c-FOS protein as a cellular marker of activated neurons. In addition to the classic segmental and supraspinal, descending inhibition, a third principle of endogenous antinociception exists: the propriospinal, intersegmental inhibition of nociceptive spinal dorsal horn neurons. Propriospinal antinociception partly mediates the descending inhibition from the brain stem and can be activated by conditioning heterosegmental noxious stimuli, thus possibly contributing to analgesia by counterirritation. In addition to the fast synaptic transmission mediated by classic neurotransmitters, the extrasynaptic transmission of chemical signals such as neuropeptides may play an important role for long-term effects following intense noxious stimulation. The controlled superfusion of the dorsal cord with neuropeptides produces a similar distribution in the spinal cord to that of endogenously released neuropeptide. We have shown that extrasynaptic neuropeptides such as substance P may increase the excitability of nociceptive spinal dorsal horn neurons and may induce the expression of 'immediate-early genes' in dorsal horn neurons in vivo. Changes in gene expression following extrasynpatic spread of neuropeptides in the spinal cord may be involved in chronic pain syndromes after massive peripheral trauma. This hypothesis has led to the concept of pre-emptive analgesia. The available evidence suggests that the known systems of endogenous antinociception do not affect the endogenous release of neuropeptides in the spinal cord. All previous concepts of endogenous antinociception are based on changes in dischargerates of nociceptive neurons. Background activity in the absence of noxious stimulation was considered to be purely stochastic 'noise'. By the use of modern tools for the analysis of nonlinear dynamics in point processes we have, however, shown that background activity of most nociceptive spinal dorsal horn neurons is highly deterministic with a low degree of freedom. The high order in the discharges of these neurons is maintained, at least in part, by tonically active descending systems. Thus, the spinal shock syndrome seen in some species after acute spinalisation may result from the loss of order in spinal neuronal discharges normally provided by the brain. The use of modern methods in studies of the functional neuroanatomy, neurophysiology and neurobiology of endogenous antinociception may help in the achievement of better application of results from basic sciences to clinically relevant pain problems.