Manuscript 1: Methodology Dependent Variation in Volumetric Bone Mineral Density Calculation throughout the Body
Introduction: Bone quality assessment using quantitative computed tomography (QCT) may provide a more in-depth and accurate assessment of osteoporosis and fracture risk than dual-energy x-ray absorptiometry (DXA). However, QCT methodologies utilizing single-scan calibration curves may not account for differential x-ray attenuation caused by the patient which may influence calculated volumetric bone mineral density (vBMD) and skew bone quality and fracture risk assessment.
Methodology: Clinical CT scans were conducted on 50 male post-mortem human subjects with phantom calibration rods throughout the scan. Height and weight were collected to determine subject BMI. Hounsfield units (HU) from skeletal volumes of interest (VOIs) were collected from the lumbar spine and left femoral neck, humerus, radius, tibia, and calcaneus. The femoral neck was segmented into trabecular (Tb), cortical (Ct) and Total (Tb and Ct) VOI’s, the lumbar spine and the calcaneus consisted of Tb and Total VOIs, and the humerus, radius, and tibia were assessed for Ct bone. HU from each VOI was converted to vBMD using both a general scan specific (Gen.) calibration curve constructed from phantom rods within the CT slices of the lumbar region and location specific (LS) calibration curves constructed from phantom rods in slices for each of the skeletal VOIs.
Results: Significant variation in vBMD calculated from Gen. and LS calibration curves was observed in the femoral neck, calcaneus, and tibia in all skeletal compartments
15
(p<0.01). However, no significant differences were observed in any of the lumbar spine, humerus, or radius VOIs (p>0.01). Additionally, BMI was not able to explain variation in vBMD values at any site (p>0.01).
Conclusions: Using a single calibration curve to calculate vBMD in other anatomical locations, may skew bone quality and differential fracture risk assessment within the tibia, femoral neck, and calcaneus. LS calibration curves may account for variable x-ray attenuation and result in more accurate evaluations of bone quality from vBMD, improving patient quality of life and potentially reduce the incidence of osteoporotic fracture.
Manuscript 2: Comparison of Phantomless and Phantom Based Volumetric Bone Mineral Density throughout the Human Body
Introduction: Fragility fractures and the associated healthcare cost are of an increasing concern. As current clinical standards using dual-energy x-ray absorptiometry (DXA) do not provide accurate assessments of bone strength and fracture risk, quantitative computed tomography (QCT) may provide more in-depth and accurate assessment of bone quality. However, traditional QCT requires the use of phantom calibration rods to calculate volumetric bone mineral density (vBMD), which are not always included in the scan. To reduce the need for repeat scans with phantom calibration rods, phantomless methodologies can use internal reference tissues to calibrate and calculate vBMD retrospectively. However, there is minimal research on phantomless methods outside of the lumbar spine and femur and inconsistencies with the known densities used for phantomless reference tissues.
Methods: Clinical CT scans were conducted on 60 male post-mortem human subjects (PMHS) spanning a large age range. 50 PMHS were used for the method development sample and 10 PMHS were used for the method validation sample. Hounsfield units (HU) were obtained from skeletal volumes of interest throughout the body including, the lumbar spine and left femoral neck, humerus, radius, and tibia. The femoral neck was further divided into trabecular (Tb), cortical (Ct), and Total (Tb and Ct) VOIs. The lumbar spine consisted of Tb and Total VOIs and the humerus, radius, and tibia only had Ct VOIs collected from the midshaft of each bone. Location specific (LS) phantomless calibration curves using muscle, fat, and air reference tissues and mediums and phantom
37
based calibration curves using INTable™ phantom rods were calculated. A general scan specific (Gen.) phantomless calibration curve was created from the region of the third lumbar spine. Phantomless HU reference values were collected in addition to the skeletal HU of each VOI. vBMD was then calculated using both the LS phantomless and phantom based calibration curves. Phantomless vBMD was also calculated using the Gen. calibration curve. Finally, phantomless vBMD was calculated using a series of multivariate regression model equations until it was optimized. Phantomless vBMD values were compared against traditional phantom based vBMD through paired t-test, univariate, and multivariate regressions.
Results: Phantomless vBMD calculated from phantomless LS and Gen. calibration curves have significantly different vBMD values in the tibia, and femoral neck sites with the exception of the femoral neck Tb VOI (p<0.05). No significant differences were found in the lumbar spine, humerus, or radius (p>0.05). Phantomless vBMD calculated from LS phantomless calibration curves demonstrated no significant differences from vBMD calculated from phantom based calibration curves with the largest percent difference being 0.77% (p>0.542). Linear regressions showed that phantomless vBMD calculated form calibration curves was able to significantly predict phantom based vBMD (p<0.015). Phantomless vBMD from Tb and Total VOIs demonstrated and higher R2 values than phantomless vBMD from Ct VOIs when regressed against phantom based vBMD. Multivariate regressions were performed to optimize the R2 value of each skeletal VOI for calculated phantomless vBMD. Results showed highest R2 values in the lumbar
38
spine at L2 Total VOI (81.31%) and in the femoral neck at Fem-Neck Tb VOI (76.62%). The humerus, radius, and tibia demonstrated optimized R2 values of 36.69%, 33.75%, and 18.25%, respectively. Paired t-tests form the validation sample demonstrated that phantomless vBMD calculated from the phantomless calibration curve was not significantly different than phantom based vBMD at any site (p>0.05) with the exception of the humerus and radius VOIs (p<0.05). Phantomless vBMD calculated from the optimized regression model equations revealed no significant differences between phantomless vBMD or phantom based vBMD at any site (p>0.05) with the exception of the humerus and tibia (p<0.05) and showed smaller percent differences than phantomless vBMD calculated from phantomless calibration curves.
Conclusion: Phantomless vBMD calibration methodologies need further study as there is potential to drastically increase the number of bone quality and fracture risk assessments without the need for additional radiation. Results from this study demonstrate that phantomless vBMD calculated from multivariate regression equations may increase the accuracy of phantomless vBMD to phantom based vBMD. Further research is also needed to explore other skeletal sites within the body including trabecular envelopes in the extremities. Comprehensively assessing the clinical utility of phantomless methodologies may improve patient quality of life and reduce fracture incidence by increasing the availability to more accurate assessments of bone quality and strength.