Purpose of the study The authors report their experience with the use of a biphasic macroporous calcium phosphate bone substitute. Materials and methods In 23 cases (22 patients) a biphasic macroporous calcium phosphate ceramic was used to fill a pathological bone defect. The ceramic used in this study was a macroporous (400 to 600 mu) component consisting of 60 per cent hydroxyapatite and 40 per cent beta-tricalcium phosphate. It was in the form of granules (2 to 3 mm), sticks (20x5x5 or 10x5x5 mm) or custom made blocks. In 6 cases, the ceramic was used alone; in 12 cases with autologous bone marrow and in 5 cases with autologous cancellous bone grafts. In 14 cases, the bone defect was due to conservative treatment of a benign tumor, in 3 cases due to aseptic post-traumatic non union, in 3 cases due to wide resection for malignant tumors of the pelvis and in 3 cases following osteotomy. Post operative assessment was made from clinical, radiographic and histological findings. Results 2 patients died 6 and 8 months post operatively and 2 were lost to follow up at 2 and 5 months with both having good clinical and radiographic results when last seen. For the remaining 19 cases, the average follow-up was 20 months (from 6 to 62 months). No local, regional or general deleterious effects were noted. Radiologically the bone ceramic junction healed in all cases except 2 within 3 months. In these last two cases, healing required 6 and 7 months. No radiolucent line appeared around the ceramic. No stress fractures occurred in the substitute. Histologically, 3 biopsies showed new bone formation throughout the ceramic with apposition of a well differentiated lamellar bone directly apposed to the ceramic. Discussion Animal experimentations have proven the interest of similar ceramics : macroporosity enhances bone rehabitation and the biphasic characteristics associate the advantages of slow resorption of hydroxyapatite and more easily resorbed beta-calcium phosphate. No deleterious clinical, radiographical or histological effects were observed, confirming the biocompatibility of this substitute. Despite the poor mechanical properties of this macroporous ceramic before implantation, good clinical and radiographic results suggest improvement of these properties in the composite new formed bone-ceramic after implantation. Conclusion We believe that macroporous biphasic ceramic is a good substitute for use in bone defects when good primary mechanical stability and contact with the host bone are present. Further clinical and experimental studies are necessary to determine the limits of such a substitute in terms of volume and to control its mechanical properties following implantation.