Although the frequency-dependent antinociceptive mechanisms of electroacupuncture (EA) have been well demonstrated, the anti-inflammatory mechanisms that underlie the suppressive effects induced by different frequencies of EA stimulation on peripheral inflammation are largely unknown. We have previously reported that EA stimulation can activate the sympathetic nervous system (SNS) and that this activation is responsible for the EA-induced suppression of zymosan-induced leukocyte migration. The present study was designed to evaluate the differential effect of low (1Hz, LF EA) versus high (120Hz, HF EA) frequency EA stimulation on SNS activation and ultimately on carrageenan-induced inflammation. Immediately after carrageenan injection, we applied either LF EA or HF EA bilaterally to the Zusanli (ST36) acupoints. To evaluate the anti-inflammatory effect of EA (EA-AI), paw volume and myeloperoxidase (MPO) activity, a marker of infiltrated leukocytes, were measured and the paw withdrawal latency to noxious heat stimulation was also assessed. Both LF EA and HF EA significantly suppressed the carrageenan-induced paw edema and MPO activity. Moreover, thermal hyperalgesia was strongly attenuated in both the LF EA and HF EA groups. Adrenalectomy significantly diminished HF EA-AI without affecting LF EA-AI. Pretreatment with the corticosterone receptor antagonist, RU-486 did not affect either LF EA- or HF EA-AI. On the other hand, administration of 6-hydroxydopamine (a neurotoxin for peripheral sympathetic nerve endings) selectively blocked LF EA-AI. Propranolol (a beta-adrenoceptor antagonist) completely abolished both LF EA- and HF EA-AI. The results of this study suggest that the suppressive effects of LF EA on carrageenan-induced paw inflammation are mediated by sympathetic post-ganglionic neurons, while the suppressive effects of HF EA are mediated by the sympatho-adrenal medullary axis.