SEASONAL PATTERN IN THE ONSET OF POLYMYALGIA RHEUMATICA (PMR)

Background: The frequent acute infectious-like onset of Polymyalgia Rheumatica (PMR) and Giant-Cell Arteritis (GCA), often associated with PMR, has stimulated interest in searching for environmental agents involved in their pathogenesis, and a close temporal relation between the peaks of incidence of these diseases and those of different infections has been reported (1,2). Although some epidemiological reports suggest that infections may trigger PMR and GCA, direct evidences are lacking and a seasonal pattern in their onset is still debated (2-6). Objectives: To retrospectively analyse seasonal pattern in the onset of symptoms of PMR in a large group of patients from 2 Italian centres and its relationship to age and gender utilising for the first time non linear Fourier analysis. Methods: Over a 10 year period (January 1992-December 2002) 201 patients with PMR (mean age 72.6 years; 126 females and 75 males) were retrospectively included in the study. In 21 patients, PMR was associated with GCA. Basing on the onset of clinical symptoms, each patient were assigned to a month of the year and to the corresponding season. A chi square test for goodness of fit and a single cosinor analysis were used to evaluate non-uniformity and rhythmicity of PMR during monthly periods. Results: The chi square test showed that frequency of onset of PMR was significantly increased during the winter (P<0.005). Single cosinor analysis identified significant circannual variations in the occurrence of PMR (P<0.001), with a peak in January and a nadir between June and July. By subgroup analysis, a seasonal variation with a chronobiological pattern similar to the whole group was identified also in male and female patients, in patients above and under 75 years and in patients with PMR without GCA. No significant seasonal variation was observed in the small group of patients with GCA associated with PMR. Conclusion: PMR shows a significant circannual variation in its occurrence, with a peak in the winter and in January both in the whole group of patients and in all subgroups, except for PMR/GCA patients. The circannual pattern of PMR could suggest a role for infectious agents or for the seasonal changes of immune function in the pathogenesis of the disease. Further studies are needed to elucidate the mechanisms involved in mediating the effects of season on PMR onset.