To better understand whether altered gait was due to non-specific sickness behavior or anxiety, either of which could affect gait velocity, we conducted an open field test and found no changes in overall activity, rearing, or time spent in the center of the field between experimental groups. The presence of gait abnormalities in our model suggests that the increased mechanical sensitivity detected with Von Frey testing was due to inflammation of deeper tissues as opposed to simply the skin. Indeed, shortened stride length similar to that observed in our study was reported in a rodent model of “myofascial” inflammation produced by injection of paraformaldehyde into the multifidus muscles. Although it is not possible to rule out sensitization of sensory afferents within muscle in the current study, we were able to LDK378 1032900-25-6 document clear evidence of carrageenan-induced inflammation within paravertebral connective tissue that improved with stretch. With the stretching technique used in this study, the animal was encouraged to hold a position of stretch that was slightly beyond its usual range of motion. This technique is relevant to active stretch therapeutic interventions that involve slow and gentle, but non-habitual, body movements. Because our animal model involves a stretch of the whole, conscious animal, anti-inflammatory effects induced by stretching could involve central as well as local mechanisms. Centrally mediated effects of stretch could include stimulation of the hypothalamic-pituitaryadrenal axis and systemic cortisol secretion with direct antiinflammatory effect on tissues. Another possibility is that stress during tissue stretch could have activated descending pain inhibitory pathways with inhibition of neurogenic inflammation via reduced secretion of neuropeptides into the tissue. Future studies will attempt to differentiate between central and/or peripheral mechanisms that may underlie the improvements observed with stretch. A potentially important difference between our stretching method and stretching methods used in humans is stress caused by the added restraint imposed by holding the animal by the tail. Further experiments comparing stretching in conscious vs. anesthetized rats will be necessary to further explore this issue. Alternatively, or in addition, it is possible that stretching could have had a direct anti-inflammatory effect on the peripheral connective tissues of the low back. The direct response of cells and tissues to mechanical forces varies greatly depending on the manner in which the force is applied. In cultured fibroblasts, repetitive or high amplitude cyclic stretch can lead to the production of proinflammatory cytokines and apoptosis, whereas brief static stretch has been reported to decrease pro-inflammatory cytokines IL-3 and IL-6. Anti-inflammatory outcomes also were found in in vitro studies where low amplitude mechanical input was applied to chondrocytes and fibroblasts. Other potentially relevant peripheral mechanisms involve the complex relationship between TGFbfibrosis and inflammation. While repetitive or high amplitude mechanical input generally increases TGFbbrief static tissue stretch attenuated the increase in both soluble TGFb-1 and type-1 procollagen following tissue injury. It is therefore plausible that a number of potentially interrelated local and systemic mechanisms may have contributed to the reduction in tissue inflammation observed in our in vivo.