Early investigations involving central nervous system (CNS) temperature lowering to protect against the detrimental effects of hypoxia and ischemia were based on the observation that hypothermia reduces brain metabolism and energy consumption. The protective effects of hypothermia have been demonstrated in numerous experimental models of cerebral ischemia and recently in models of brain trauma. These observations also led to the application of hypothermia, in the form of local spinal cord cooling (LSCC), in animal models of experimental spinal cord injury (SCI). Although some investigators have reported negative results in studies of LSCC following traumatic SCI, the majority of studies have noted beneficial effects. The favorable results in animal experimentation led to a limited number of cases where LSCC was used in the treatment of human SCI. However, results are difficult to interpret because (1) most investigators report only a small number of cases, (2) the studies lack a control population, (3) the time interval from injury to the application of cooling has been highly variable, and (4) several investigators combined drug treatments with LSCC. In these experiments, LSCC was achieved via perfusion with a cold solution or an epidural heat exchanger and the aim was to lower cord temperatures significantly (about 10 degrees C). The application of the technique itself is fraught with difficulties. It requires acute surgery in a traumatized patient, a wide multilevel laminectomy, and minimizing the time interval between injury and the application of spinal cord cooling. Recent studies in experimental brain ischemia strongly suggest that a drastic lowering of CNS temperature may be unnecessary to lessen the degree of tissue damage occurring following an ischemic brain injury.(ABSTRACT TRUNCATED AT 250 WORDS)