The mechanisms underlying local and referred muscle pain are poorly understood. The aim of this experiment was to determine to which degree referred pain is dependent on peripheral or central mechanisms. This was studied by blocking sensory input from the referred pain area. Intramuscular electrical stimulation in the right anterior tibial muscle induced local muscle pain and referred pain in 12 subjects. Nerve blocks were applied on the lower limb between the local and the referred pain area for 60 min. Three different studies were performed: (1) no nerve block (2) compression block, and (3) compression block combined with intravenous regional anaesthesia. The referred pain intensity was assessed every 5 min. To monitor the blockade of myelinated and unmyelinated nerve fibres, touch, pinprick, position sense and heat and heat-pain detection thresholds were assessed on the dorsal side of the foot every 5 min. A significant reduction in referred pain intensity (40.5%) compared with the control experiment was found after 40 min with the compression ischemia nerve block until release of the tourniquet. In the combined nerve block experiment, a significant reduction of referred pain intensity (38.5%) was seen after 20 min and until the release of the tourniquet. There was no significant difference in the referred pain intensity between the two types of blocks. The present findings suggest that both peripheral and central mechanisms play a role in referred pain and that the myelinated fibres mediate the peripheral component.