To estimate the accuracy of quantitative computed tomography (QCT) as a method to measure bone mineral density (BMD) in the vicinity of a titanium prosthesis, we investigated the effects of (1) titanium prosthesis, (2) offset of the longitudinal axis of the bone to be examined from that of the gantry of the CT scanner, (3) size of the field of view (FOV) and (4) the combination of these effects on CT based measurements of mineral density of cortical and cancellous bone specimens. 14 bovine cortical bone parallelepipeds and 14 bovine cancellous bone parallelepipeds were used in this investigation. The bone specimens were scanned with and without a titanium prosthesis, when centered in the gantry of the CT scanner and offset from the axis of the gantry of the CT scanner at a distance of 14 cm. Image data were then reconstructed separately with a FOV of 10 cm and 30 cm. All BMD values taken from CT images obtained under different scanning condition were compared with the BMD values of the corresponding bone parallelepiped obtained under standard condition (centered in the gantry of the CT scanner, 10 cm FOV, without titanium prosthesis). When centered in the gantry of the CT scanner, the mean relative difference of BMD measurements caused by the presence of the titanium prosthesis was less than 1% for both cortical bone and cancellous bone. Size of the FOV had a negligible effect on BMD measurements. Offset at 14 cm, however, caused a significant difference in BMD measurements (p<0.001). It was concluded that titanium prosthesis did not interfere with BMD measurements of cortical and cancellous bone when both the specimen and prosthesis were centered in the gantry of the CT scanner. However, the effect on BMD measurements of offset at 14 cm combined with the presence of a titanium prosthesis in bone was significant.