An increasing number of papers deal with immunological factors in headache syndromes such as migraine and cluster headache. The aim of this review is to give an overview of the factors that have been measured and to assess their reliability and relevance for the pathogenesis of these headaches. Most of the studies are handicapped by methodological problems, especially the different classifications of headaches, the lack of adequate controls and methodological problems with the measurement of certain immune parameters. Nevertheless, immunological abnormalities have been shown to be reproducible. These are the increased number of autoantibodies in migraine and the augmented number of deficient NK cells and monocytes in cluster headache. Furthermore, some cytokine levels (IL-2) have been shown to be decreased in migraine and tension-type headache, while others (IL-1, TNFalpha) seem to be elevated. Cluster headache seems to be associated with persisting viral infections. Studies of immunoglobulins and immuncomplex levels and analyses of immunomediators such as prostaglandins and histamine still yield contradictory results. Although the immunological changes have been shown to be valid, their pathogenesis in these headaches is unclear. With the increasing recognition of the existence of a neuroimmunologic network, alterations in each system should always be considered to be associated with changes in an other. Acute or chronic pain seems to trigger immunological abnormalities.