Spinal cord stimulation (SCS) was an outgrowth of the well-known gate control theory presented by Melzack and Wall in 1965. Although the method has been used to treat chronic severe pain for more than three decades, very little was known about the physiological and biochemical mechanisms behind the beneficial effects until recently. We now know that SCS activates several different mechanisms to treat different types of pain such as neuropathic and ischemic. In general, these mechanisms seem most dependent on activation of only a few segments of the spinal cord. However, both animal studies and human observations have indicated that supraspinal circuits may contribute as well. In the treatment of neuropathic pain, intermittent SCS may give several hours of pain relief after cessation of the stimulation. This protracted effect indicates long-lasting modulation of neural activity involving changes in the local transmitter systems in the dorsal horns. In ischemic pain, animal experiments demonstrate that inhibition of afferent activity in the spinothalamic tracts, long-term suppression of sympathetic activity, and antidromic effects on peripheral reflex circuits may take part in the pain alleviation. Moderate SCS intensities seem to evoke sympathetic inhibition, but higher stimulation intensities may induce antidromically mediated release of vasoactive substances, eg, the calcitonin gene-related peptide (CGRP), resulting in peripheral vasodilation. The anti-ischemic effect of SCS in angina pectoris due to intermittent coronary ischemia probably occurs because application of SCS appears to result in a redistribution of cardiac blood supply, as well as a decrease in tissue oxygen demand. Recent studies indicate that SCS modulates the activity of cardiac intrinsic neurons thereby restricting the arrythmogenic consequences of intermittent local coronary ischemia. The present state of knowledge is briefly reviewed and recent research directions outlined.