Various cytokines have been identified at sites of chronic inflammation such as periodontitis. Cytokines are synthesized in response to bacteria and their products, inducing and maintaining an inflammatory response in the periodontium. The purpose of the present study was to investigate the involvement of interleukin-1 beta (IL-1 beta), IL-8, and IL-10 and RANTES (regulated on activation, normally T cell expressed and secreted) and the cell populations associated with the immune response in destructive periodontitis, as well as the effect of periodontal therapy on cytokine levels in gingival crevicular fluid (GCF). Data were obtained from 12 patients with moderate to advanced periodontitis and 6 healthy controls. Patients presenting at least 2 sites with > or =2 mm clinical attachment loss were included in the destructive periodontitis group. After monitoring for 4 months, only 6 patients showed destructive periodontitis and GCF samples and soft tissues biopsies were collected from these patients. GCF samples and biopsies were collected both from active (12 CGF samples and 6 biopsies) and inactive (12 CGF samples and 6 biopsies) sites. The comparison with healthy controls was carried out by collecting GCF samples from 6 healthy volunteers (12 samples) and biopsies during the surgical removal of wisdom teeth. In periodontal patients, clinical data and GCF samples were obtained prior to periodontal treatment (72 samples) and 2 months after periodontal therapy (72 samples). GCF was collected using a paper strip; eluted and enzyme-linked immunoabsorbent assays (ELISA) were performed to determine cytokine levels. The inflammatory infiltrate was analyzed by immunohistochemistry of gingival biopsy samples with monoclonal antibodies against CD3, CD8, CD4, CD11c, and CD19 antigens. Cellular components of the inflammatory infiltrate include B and T lymphocytes and monocyte/macrophages. Active sites contained a higher number of B lymphocytes and macrophages. IL-8 and IL-1 beta and RANTES in GCF were detected in the majority of sites from periodontal patients (100%, 94% and 87%, respectively); IL-10 was found in only 43%. IL-8 was the only cytokine detected in the GCF (75%) of the control group. Moreover, IL-1 beta levels were significantly higher in active sites versus inactive sites (P <0.05). IL-8 and IL-10 and RANTES were increased in active sites; however, differences were not significant (P>0.05). A positive correlation between the IL-8 and RANTES (r = 0.677, P<0.05) was observed in periodontitis patients. Periodontal therapy reduced the total amount of IL-1 beta, IL-8, and IL-10 and RANTES. Data showed a weak correlation between the clinical parameters and the total amount of cytokines in periodontitis. These data suggest that the amount of crevicular IL-1 beta, IL-8, and IL-10 and RANTES is associated with periodontal status. Removal of the bacterial plaque reduces the antigenic stimuli and consequently could modulate the chemokines present in GCF. We propose that the dynamic interactions between cytokines, their production rates, and their quantity could represent factors controlling the induction, perpetuation, and collapse of the cytokine network present in the periodontal disease.