Thirty 8-week-old male Wistar rats were randomly allocated into three groups of 10 rats each. A 5-mm defect in the left parietal bone was made in each rat. In the defects of the first group of rats, no implant was used (control group). In the second group, polyorthoester membranes were placed in the defects without active substance. In the third group, polyorthoester membranes were placed with insulinlike growth factor I. The rats were sacrificed 6 weeks postoperatively. Bone formation in the defects was quantified by computer-assisted measurements of the area of the residual defect on radiographs. Host-tissue response was evaluated by light microscopy. The area of residual bone defect was greatest in the control group, less for the defects with polyorthoester membrane without active substance, and least for the defects with polyorthoester membranes with the growth factor. During histologic evaluation, no inflammation was seen, and only traces of the polyorthoester were detected in the defects with polyorthoester membrane with or without the growth factor.