Nociceptive stimuli are modulated at the dorsal horn of the spinal cord. This modulation is performed by various systems working independently complementarily, additively or supra-additively. Non-opioid analgesics relieve pain without a motor blockade. In contrast to spinal opioids a reduced risk of respiratory depression is expected. In the therapy of chronic pain non-opioid analgesics may be an alternative, given alone or in combination with an opioid. Clinically relevant dosages for antinociception mediated by the alphaadrenoceptoragonistclonidine are >/=150 mug epidurally. Clonidine is effective in reducing acute and chronic pain. In combination with opioids the action of the opioids is intensified. Clonidine intensifies and prolongs the action of local anesthetics. If opioid tolerance occurs, epidural clonidine alone or in combination with an opioid has good antinociceptive action.Midazolam, a water-soluble benzodiazepine, was injected spinally for the reduction of pain for various indications (postoperative, malignancy, chronic back pain, spinal spasticity). Spinal benzodiazepine should not be injected into the spine in patients until it has been proven that there are no neurotoxic effects. Intrathecally injectedbaclofen is a well-known means of reducing spinal spasticity. Used in this way, it may have a secondary analgesic effect. No significant direct analgesic effect has so far been demonstrated. Spinalcalcitonin often leads to insufficient pain relief when given alone. Combination with an opioid may reduce the dosage of the opioid. Nausea and vomiting are frequent side effects of spinal calcitonin. Intrathecalsomatostatin produces antinociception. However, in animal studies neurotoxic action has been observed. Administration in man has not yet been proved to be safe. Spinalketamine has procluted controversial results in clinical studies, and has not yet been excluded that the substance is not neurotoxic.Lysine acetylsalicylic acid (L-ASA) has been given intrathecally for the therapy of severe cancer pain and chronic back pain. In most patients good analgesia was observed up to 2 months after a single injection. If neurotoxity can be excluded, L-ASA may be an alternative in the therapy of cancer pain before neurodestructive therapy is done.