The purpose of this study was to examine if local thermo-applications affect central nervous reactions. In a crossover study, six normal, healthy volunteers at first received cold packs (Cryogel, 8-12 degrees C; Pino GmbH, Hamburg, Germany) and afterwards hot packs (Parafango, 50-60 degrees C; Pino GmbH), and another six volunteers started with the hot packs and had the cold packs later; both groups administered the hot and cold packs to their thighs. Before, during, and after treatment, cerebral blood flow velocity (CBFV) in the middle cerebri-artery (MCA) was measured continuously by transcranial Doppler sonography, whereas cerebral respiratory chain enzyme cytochrome aa3 (cCytaa3) and cerebral oxygen saturation (cHbO2) were measured by transcranial near infrared spectroscopy in frontal brain tissue. Furthermore, CO2 end-tidal and arterial blood pressure (noninvasive) were also measured. Six other volunteers received only one treatment; therefore, 15 measurements with cold and 15 measurements with hot packs were performed. During application of cold packs, a decrease of cHbO2 of 10.5% (P < 0.001) and cCytaa3 of 6.7% (P < 0.001) was found, whereas the CBFV(MCA) increased significantly (3.9%; P < 0.001) between preliminary and post-stimulus periods. When cold packs were removed, a significant increase of the cHbO2 (16.9%; P < 0.001) and cCytaa3 (9.7%; P < 0.001) was measured. With these values, cHbO2 and cCytaa3 showed an overshooting counterreaction beyond the initial level. When applying the hot packs, a contrary course of the parameters was found. cCytaa3 showed a significant increase of 9.3% (P < 0.001) at the end of the stimulus phase and a decrease of 1.9% (P = 0.02) during the post-stimulus period. The correlating increase of cHbO2 was significant at 13.7% (P < 0.005). At the end of the post-stimulus phase, a significant decrease of cHbO2 at 1.9% (P = 0.004) was recorded. With Parafango applications, a significant decrease of CBFV(MCA) at 6.9% (P < 0.001) was measured at the end of the stimulus in comparison with the preliminary phase. Crossover analysis showed no significant period effects and intraindividual changes between period and treatment. Therefore, both treatments can be compared within the individual using paired t test. Local cold and warm stimuli influence the cerebral hemodynamics and cerebral metabolism. Cerebral hemodynamics (CBFV(MCA) in comparison with cerebral metabolism (cCytaa3, cHbO2) show opposite reactions under thermo-stress. Of special interest is the overshooting counter-regulation of cerebral metabolism after cold stimulation. These effects may open new thermotherapeutic aspects in central nervous system diseases.