8–1.0 s/rot; beam pitch: 0.5625–0.9375) and reconstruction parameters were predefined for each type of CT scanner (see Appendix). Beam pitch is defined as the ratio of table feed per rotation to the collimation, where collimation is the product of slice-thickness and the number of slices in each rotation. Beam pitch was kept under 1.0 except for one CT scanner

(Somatom Plus 4 Volume Zoom). Field of View (FOV) was defined as 350 mm to cover both hip regions. In-plane spatial resolution of 0.625–0.652 mm and reconstructed slice thickness of 0.500–0.625 mm was adjusted according to CT scanner type (see Appendix). The CT values were converted to bone mineral scale by using a solid reference phantom, Ku-0059436 B-MAS200 (Fujirebio Inc., Tokyo, Japan), containing hydroxyapatite (HA) at 0, 50, 100, 150, and 200 mg/cm3. For all of the CT data, a constant threshold value of 350 mg/cm3 was used to define the cortical bone. The MDCT scanners used in this study originally included four Asteion 4 scanners, one Aquilion 4 scanner,

and three Aquilion 16 scanners (Toshiba Medical Systems Corporation,Tochigi, Japan); one LightSpeed Ultra_8 scanner, and one LightSpeed Plus_4 scanner (GE-Yokogawa Medical,Tokyo,Japan); and one Somatom Plus 4 Volume Zoom scanner (Siemens, AG, Berlin and Munich, Germany). In two institutions, CT scanners were changed during the trial period (from Aquilion 16 to Aquilion 64, and from LightSpeed Plus_4 to LightSpeed Ultra_16); therefore, the pairs of CT data in 26 patients were obtained Selleckchem Epacadostat using different CT scanners. However, because the results of all patients did not differ from results excluding the 26 patients (data not shown), the results of all patients are presented in this article. Good linear correlations between the

CT values and HA concentrations were demonstrated (r = 0.996–0.999; p < 0.0002–0.05) in all CT scanners. Differences in CT values according to X-ray energy were corrected by using the reference phantom to convert CT values to HA equivalent values. However, it was necessary to confirm the longitudinal stability of the CT values of the threshold value used to define the cortical bone. For the rod containing 200 mg/cm3 HA equivalent, which was used as the threshold value to define the cortical region, there Casein kinase 1 was less than 0.01% difference between the baseline CT value and CT value at 144 weeks. The subjects were scanned in the supine position, with the reference phantom beneath the patient and placed so as to cover a region from the top of the acetabulum to 5 cm below the bottom of the lesser trochanter in each hip joint (average slice number was 298). Bolus bags were placed between the subject and the CT calibration phantom. Both feet were fixed using a custom-made adjuster for hip DXA, which kept the subject’s knees flat and the toes pointed inward.