Fibromyalgia (FM) is characterised by chronic widespread pain and allodynia (pain from stimuli which are not normally painful with pain that may occur other than in the area stimulated) of more than 3 months duration. The current hypothesis of the aetiology of FM includes inflammatory and neuroendocrine disorders. The biophysiology of this syndrome, however; remains still widely elusive, and there are no formally approved therapies. Non-pharmacological interventions in FM patients include habitual exercise programs which improve physical function and quality of life of patients and may even reduce pain. However the mechanisms through which exercise benefits FM symptoms needs to be elucidated. In this article we firstly review the main topics and characteristics of the FM syndrome, while focusing our attention on the inflammatory hypothesis of FM, as well as on the beneficial effects of habitual exercise as a co-therapy for FM patients. In this context, the latest developments in research on anti-inflammatory effects of exercise are also reviewed and discussed. To find out what is known about the connection between benefits of exercise for FM and anti-inflammatory effects of exercise, we carried out a PubMed search using the term "fibromyalgia" and "exercise" together with "inflammation", and no more than ten published articles were found (six of them reviews), which are also discussed. In the second part of the article we present a pilot investigation on a group of 14 female FM patients with a diagnosis of FM by a rheumatologist. They took part in a pool-aquatic program in warm water over a period of fourth months (three weekly 60-min sessions). Circulating inflammatory (IL-1beta, IL-2, IFNgamma, TNFalpha, IL-8, IL-6, IL-4, IL-10 and CRP) and neuroendocrine (NA and cortisol) markers were determined. FM patients showed higher circulating levels of IL-8, IFNgamma and CRP as well as cortisol and NA than age-matched healthy control women. After the exercise program, a significant decrease in IL-8, IFNgamma, and CRP were found, in parallel with a decrease in circulating concentrations of cortisol and increased levels of NA. The results confirm an elevated "inflammatory status" in the FM syndrome and strengthen the hypothesis that the benefits of exercise in FM patients are mediated, at least in part, by its anti-inflammatory effects. A better regulation of the cytokine-HPA axis feedback may be also involved.