This article is a review of our experimental results regarding the physiological statuses and roles of chemical mediators in tourniquet shock, and a novel phenomenon, modulation reflex, that is commonly observed in this shock model is discussed. In a rabbit with a tourniquet applied to a hind limb for 24 hrs, blood pressure (BP) gradually falls after release of the tourniquet, but the decline in BP stops when a tourniquet is again applied to the hind limb, indicating that shock mediators are attributed to the hind limb. The levels of dipeptides (anserine and carnosine) and lysosomes in blood samples as well as the levels of leukotrienes (LTD4 and LTE4) in blood and muscle samples from rabbits in tourniquet shock were elevated. However, injection of a large amount of a dipeptide into an ear vein of a rabbit did not reduce BP, suggesting that both peptides may not be directly related with reduction in BP of rabbits in tourniquet shock. Injection of a platelet-activating factor (PAF) antagonist into an ear vein resulted in slight elevation of BP and the elevated level was maintained for about 1 to 4 hrs during the period of decline in BP in tourniquet shock. As for interleukin-6 (IL-6), IL-6-deficient mice at young ages have a significantly greater blood volume than do wild-type mice without concomitant changes in body composition. Therefore, the role for IL-6 in the regulation of peripheral circulation may be to elevate, not reduce BP. In mice in tourniquet shock, superoxide (O2-) production is observed in skeletal muscle cells and these cells correspond to mitochondria-rich cells. However, RT-PCR of muscle samples showed no significant nitric oxide synthase (NOS) mRNA expression after tourniquet release. Pretreatment with NOS inhibitors before tourniquet release reduced O2- production in the skeletal muscle. These results indicate that O2- produced in muscle subjected to ischemia/repefusion may be involved in shock. As for changes in mRNA expression patterns of pro-inflammatory cytokines and nerve growth factors in blood samples from rats in tourniquet shock, up-regulation of M-CSF mRNA began at 2 h after tourniquet application and was short-lived. The level of ATF-3 mRNA had increased at 1 h and NGF mRNA gradually increased and reached a significantly high level at 4 h after tourniquet application. These results indicate that the transient mRNA expressions probably trigger secondary events that may be beneficial to wound repair and regeneration. In the early stage of tourniquet shock, the levels of IL-6 mRNA in the liver and kidneys of rats increased progressively and significantly, and the levels of iNOS mRNA in the kidneys increased. These findings suggest that that humoral and/or cellular mediators produced locally in the hind limb are responsible for remote organ injuries. Thus, these mediators, interacting each other, may contribute to the progress of shock. We have also found a novel phenomenon in tourniquet shock using rabbits. When a tourniquet is applied to the upper hind limb of a rabbit for 24 hrs, and pressure is applied to the femoral medial area immediately after tourniquet release, a reflex of decrease in blood pressure and decrease in heart rate, which last for a short period, is usually observed. This reflex is mediated through the ipsilateral femoral nerves, central nervous system and vagal nerves. Since the modulation reflex may be due to peripheral nerve injury, we investigated morphological and molecular changes in sciatic nerves and dorsal root ganglion (DRG) neurons in rats after tourniquet application. At 4 hr after tourniquet application, light microscopic examination showed only degeneration of the tourniquet segment in the sciatic nerve but no morphological changes in the DRG, while electron microscopic examination revealed mitochondrial swelling in some DRG neurons on the tourniquet-applied side and calcium deposition in these swollen mitochondria. These findings suggest that peripheral nerve injury induced a large amount of calcium influx into neuronal cell somas and that excess amounts of calcium-influx into neurons resulted in mitochondial swelling. Results of mRNA level analyses showed NGF mRNA expression followed by NGF protein expression in Schwann cells of the ipsilateral DRGs at 4 h after tourniquet application but not in the contralateral or control DRGs. Similarly, significantly high nNOS and iNOS mRNA levels were observed in the ipsilateral DRGs at 4 h, and expressions of nNOS and iNOS proteins were detected in the ganglion of the ipsilateral DRG. In addition, the TNF-alpha mRNA levels were significantly increased in the ipsilateral DRGs at 1 h after tourniquet application, indicating that TNF-alpha was activated in the early stage of nerve injury and then induced iNOS mRNA expression. Large amounts of nitric oxide (NO) produced by iNOS might result in damage to the host cells, and an overdose of NO might induce apoptosis and eliminate damaged cells during the early stage of nerve injury.