Although transcutaneous electrical nerve stimulation (TENS) is used extensively in inflammatory joint conditions such as arthritis, the underlying mechanisms are unclear. This study aims to demonstrate an opiate-mediated activation of descending inhibitory pathways from the rostral ventral medulla (RVM) in the antihyperalgesia produced by low- (4 Hz) or high-frequency (100 Hz) TENS. Paw withdrawal latency to radiant heat, as an index of secondary hyperalgesia, was recorded before and after knee joint inflammation (induced by intra-articular injection of 3% kaolin and carrageenan) and after TENS/no TENS coadministered with naloxone (20 microg/1 microl), naltrindole (5 microg/1 microl), or vehicle (1 microl) microinjected into the RVM. The selectivity of naloxone and naltrindole doses was tested against the mu-opioid receptor agonist [D-Ala2,N-Me-Phe4,Gly-ol5]-enkephalin (DAMGO) (20 ng, 1 microl) and the delta2-opioid receptor agonist deltorphin (5 microg, 1 microl) in the RVM. Naloxone microinjection into the RVM blocks the antihyperalgesia produced by low frequency (p < 0.001), but not that produced by high-frequency TENS (p > 0.05). In contrast, naltrindole injection into the RVM blocks the antihyperalgesia produced by high-frequency (p < 0.05), but not low-frequency (p > 0.05) TENS. The analgesia produced by DAMGO and deltorphin is selectively blocked by naloxone (p < 0.05) and naltrindole (p < 0.05), respectively. Thus, the dose of naloxone and naltrindole used in the current study blocks mu- and delta-opioid receptors, respectively. Hence, low-frequency and high-frequency TENS produces antihyperalgesia by activation of mu- and delta-opioid receptors, respectively, in the RVM.