I read with interest the study of Salvarani and colleagues [1], which suggests that an infectious agent could be involved in giant cell arteritis. At the French Annual Meeting of Internal Medicine in June 1994, we presented preliminary data from a study that tried to identify a causative virus in 174 elderly patients (mean age, 76.5 ± 10.1 years) who had an unexplained inflammatory process and had temporal biopsy for presumed giant cell arteritis [2]. All arterial biopsy specimens were used to detect cytomegalovirus DNA by a previously described in situ hybridization technique [3]. We compared the results with the final clinical diagnosis and viral serologic status. The patients were classified into the following groups: the giant cell arteritis group (n = 38), the polymyalgia rheumatica group (n = 20), the systemic inflammatory disease group (n = 27), and the control group, which consisted primarily of patients with cancer and infectious disease (n = 89). No statistically significant differences were seen among the groups for age, the ratio of men to women, seasonal repartition, and inflammatory biological variables. Only the presence of cytomegalovirus DNA was statistically significant; the presence of this DNA was higher in the giant cell arteritis group than in the other three groups: 42% compared with 10.5% in the polymyalgia rheumatica group, 26% in the systemic inflammatory disease group, and 21% in the control group (P = 0.03). The prevalence of positive cytomegalovirus DNA detection by in situ hybridization was calculated according to a monthly record kept throughout the 3-year study. In the control group, we noted a substantial decrease in the number of positive test results throughout the year, from nearly 50% in January and February to less than 25% in the remaining 10 months (P = 0.03). The monthly pattern for cytomegalovirus DNA detection in artery walls could be interpreted as a seasonal preferential spread of this virus. This supports the idea that cytomegalovirus could be involved in inflammatory artery injuries (such as giant cell arteritis) in genetically predisposed or immunocompromised host.