After hip prosthetic replacement, a progressive enlargement in the radiolucent area has often been observed around the implant, leading to loosening of the prosthesis. The purpose of this study was to investigate the mechanism of the radiolucent area formation. Radiolucent areas can be classified into either linear type or the erosive type, and these two types were compared histologically and biochemically. Interface membranes were obtained from patients at the time of surgery for revision of either cemented THA or cementless bipolar endprosthetic replacement. Histological specimens were stained by H.E., tartrate-resistant acid phosphate, and by the immunohistochemical reagents anti-macrophage antibody (CD 68), anti-T-lymphocyte (CD 3, CD 4, CD 8, CD 43), anti-interleukin-1 beta polyclonal antibody, anti-interleukin-6 polyclonal antibody, and anti-tumor necrosis factor-alpha polyclonal antibody. Biochemically, interleukin-1 beta, IL-6, IL-8, TNF-alpha were assayed by ELISA in the supernatant of homogenized samples and in organ culture media. Prostaglandin E2 was assayed by radioimmunoassay. The interfaces of the erosive type contained more debris (cement, high density polyethylene and metal), macrophages and multinucleated giant cells than the linear type. The interfaces of the linear type showed mainly fibrosis and necrosis. The levels of IL-6 and IL-8 in the homogenates and culture media from the erosive type were significantly higher than those from the linear type. We concluded that the bone resorption around the implant after hip prosthetic replacement occurred by two different pathways. One pathway involved the stimulation of macrophages by various debris and micromovement to form foreign body granulomas, which produced cytokines, prostaglandin E2 and metalloproteinase to resorb bone. The erosive type would arise from this pathway. The other possible mechanism involved a biomechanically unstable implant which caused bone necrosis probably by mechanical stress. The linear type may arise from this pathway.