Since there are numerous studies on CMV seroprevalence in various groups in Turkey, the number of population based, age-stratified cross-sectional studies which include epidemiological characteristics of the virus are limited. The aim of the study was to investigate the age-stratified seroprevalence and epidemiological characteristics of CMV infection in Antalya (a province located in Mediterranean region of Turkey). Study group was selected by cluster sampling method. The sample size was calculated as 360 subjects (151 male, 209 female; age range: 1-49 years, mean age: 22.5 +/- 14.4 years), with an expected prevalence rate of 80%, at a confidence level of 95% and a sample error less than 5%. With the thought of the presence of maternal antibodies, 0-1 year age group was not included to the study. Serum samples have been screened for CMV-IgG, and those given negative results were also searched for CMV-IgM by a commercial microELISA (Radim, Italy) test. The overall seroprevalence of CMV-IgG was found as 93.6% (337/360) in Antalya municipality and IgM positivity was not detected in CMV-IgG negative sera. An increase in the seroprevalence rates was observed with age (p < 0.001), and the rate was found quite high (93.3%) for the first year of life. The seropositivities in the age groups of 1-6, 7-14 and 14-49 years were detected as 82.1%, 92% and 97.8%, respectively. The seroprevalence rate of 82.1% before the age of seven has rised to 96.8% after that age, and being > or =7 years old was found statistically significant in terms of CMV infection (p < 0.001, OR: 6.635). Ages one and seven were found to be the critical ages for CMV infection in our region. CMV seropositivity was 97.4% in woman at childbearing age (15-49 years). Gender, marital status, education, living area, residence, income, history of sexually transmitted diseases, surgery, blood transfusion and day care attendance did not contribute independently to the seroepidemiology of CMV (p > 0.01). In addition, the data of this study were evaluated and discussed together with the results obtained from the other Turkish studies, as far as accessible. In conclusion, since CMV seroepidemiology in Turkey differs as the socioeconomic changes occur, the changes in CMV serostatus and dire consequences of high seroprevalence rates on public health should be evaluated with prospective, population based studies in further years.