Time dependency of bone density estimation from computed tomography with intravenous contrast agent administration

Our study has demonstrated that in contrast-enhanced multi-detector computed tomography (MDCT)-based bone density measurements, the scan delay time after contrast agent administration is a statistically significant variable for the derivation of quantitative computed tomography (QCT)-equivalent bone mineral density (BMD) values. Earlier investigators have proposed to derive QCT-equivalent BMD values from contrast-enhanced MDCT scans by using a merely density-based conversion equation. The purpose of this study was to investigate whether the scan delay after intravenous (IV) contrast agent administration might affect BMD values derived in this way. A retrospective data analysis was performed on 198 subjects who underwent standardized biphasic MDCT. Average densities values (in Hounsfield units) of lumbar vertebral bodies 1 to 3 (L1-L3) were compared between phases I and II of the biphasic MDCT scan. Furthermore, QCT-equivalent BMD (BMDQCT) values were calculated using a previously published conversion equation. Paired t-test analysis revealed that IV contrast agent administration leads to a statistically significant increase (8.6 %; p < 0.0001) in overall density of L1-L3 from phases I to II. Moreover, comparison of BMDQCT values between phases I and II reveals a change from osteoporotic to osteopenic in 4.5 % of the study population and from osteopenic to normal for 11.1 % of the subjects. Furthermore, it was revealed that the density increase from phases I to II shows a weak, yet statistically significant (p < 0.001) age dependency. Our study demonstrates that the use of a mere density-based conversion equation for deriving BMDQCT from MDCT scans ignores time dependency as an important variable. Furthermore, our results indicate that the actual age-dependent BMD itself might be another relevant variable that needs to be included in a MDCT-to-QCT conversion equation.