It has been suggested that epidural anaesthesia may increase blood flow in free flaps on the lower extremity. The objective of the present study was to test this hypothesis in 21 patients undergoing reconstructive surgery of the lower extremity with free muscle (n=8), fasciocutaneous (n=6) or musculocutaneous (n=7) flaps. Microcirculatory blood flow was measured continuously with a multichannel laser Doppler flowmetry, both in muscle and skin of the free flap as well as in the intact skin and muscle on the same extremity. After completion of surgery, general anaesthesia was continued and the epidural block was induced by an injection of 2% lignocaine-hydrochloride into a pre-operatively inserted and tested epidural catheter. The epidural block caused no change in microcirculatory flow in the intact skin and muscle, however, it resulted in a marked decrease in microcirculatory blood flow in all the free flaps studied (20–30%; P < 0.05). The epidural block also caused a significant decrease in mean arterial blood pressure, from 85 (±2.8) mmHg to 68 (±2.8) mmHg (P < 0.01). It was concluded that epidural anaesthesia may decrease microcirculatory blood flow in free flaps on the lower extremity by diverting flow away from the flap to normal intact tissues (a steal phenomenon).