Regarding orthopaedic implant loosening it has been hypothesized that particle-activated macrophages release interleukin-1 (IL-1) and tumor necrosis factor-alpha (TNF-alpha). This in turn stimulates osteoblasts to release interleukin-6 (IL-6) and prostaglandin E(2) (PGE(2)). These mediators recruit and activate osteoclasts and may therefore lead to bone resorption and loss of implant fixation. In this study we compared the ability of different materials to induce the release of IL-6 and PGE(2) from primary isolated, human osteoblasts without preceding activation by macrophages. We tested stainless steel, cobalt-chromium alloy (CoCrMo), commercially pure titanium (cpTi), Ti-6Al-7Nb and Ti-6Al-4V processed in the same manner as corresponding clinical implants. After 12 and 24h the cells had actively secreted IL-6 and PGE(2). There were no clear differences among the implant materials or with the plastic control. The amount of factors the cells released in our study compare well with the findings of other authors who investigated osteoblasts on plastic. In comparison with the literature these amounts are lower than secretion levels of osteoblasts stimulated with implant particles, IL-1 or TNF-alpha. Moreover, other authors found that osteoclasts require higher concentrations of PGE(2) to become activated than the concentrations measured in our experiments. Therefore, the amount of PGE(2) released from the osteoblasts in our study is probably not sufficient to induce osteolytic activity. Because of contradictory statements in the literature it is unclear if the measured IL-6 concentrations promote osteolytic activity. Differences in material composition does not significantly influence the release of these factors if the materials have similar surface roughnesses.