Interactions between the input from cutaneous and nociceptive muscle afferents in polysynaptic reflex pathways were investigated in man. Interaction was tested by evoking reflexes before, during, and after a period of muscle pain induced by intramuscular injection of hypertonic saline. Muscle pain was induced either in the ankle flexor (tibialis anterior, TA) or in the extensor (soleus, SOL) muscles by injection of 1 mi hypertonic saline. Electrical skin stimulation (1.1 X initial reflex threshold) at the dorsum of the foot over the tarsal joint was used to elicit cutaneo-muscular polysynaptic reflexes in the knee flexor semitendinosus (ST). The injected hypertonic saline evoked a robust muscle pain (the subjects made a continuous score of the muscle pain on a 0-10 cm VAS scale, and the mean VAS area was 1229 +/- 251 cm X s and lasting 390 +/- 30 s). In five of 12 subjects, the infusion of hypertonic saline into TA evoked referred pain to the dorsal aspect of the ankle. A significant inhibition (17 +/- 8.2%, P < 0.05) of the ST-reflex by pain in SOL was observed. Pain in TA facilitated (92 +/- 36%, P < 0.05) the short-latency part (50-70 ms post stimulation) of the reflex. The muscle pain did not modulate the perceived sensory intensity of the electrical stimuli. The findings indicate an interaction of input from thin muscle afferents and cutaneous group A-fibre afferents in polysynaptic segmental reflex pathways, which seems to depend on the location of the muscle pain. (C) 2000 International Association for the Study of Pain. Published by Elsevier Science B.V.