Implantation of artificial materials elicits a local inflammatory response. In this study a chamber model technique, allowing sampling of the inflammatory exudate for further analysis in vitro, was used. Male Sprague Dawley rats were injected daily with two different anti-inflammatory drugs, betamethasone and indomethacin, and the local cellular response was compared with a control group. The retrieved exudate was evaluated with respect to the number of leucocytes, cell viability, differential counts and serum-opsonized zymosan stimulated chemiluminescence (CL). In all groups the majority of cells were polymorphonuclear granulocytes (PMNGs). Betamethasone and high-dose indomethacin (1.92 mg kg-1 body weight day-l) treatment caused a marked reduction in the number of accumulated leucocytes 6 days after implantation. A substantial inhibition of the CL response was observed 6 days after treatment with betamethasone (4.23 mg kg -1 body weight day -1). An increased CL responsiveness was observed after 24 h with low-dose indomethacin (0.03 mg kg -1 body weight day -1) and after 6 days with high-dose indomethacin (1.92 mg kg -1 body weight day -1) treatment. In summary, depending on the anti-inflammatory drug treatment, dose and time after implant surgery, either an inhibition or stimulation of leucocyte accumulation and activation was observed. This study shows the possibilities of sampling the inflammatory exudate adjacent to a biomaterial implanted in vivo. This chamber model may be useful for the analysis of the inflammatory reaction around an implanted biomaterial during pharmacological treatment.