Quantitative computed tomography of spine - Comparison of three-dimensional and two-dimensional imaging approaches in clinical practice

Quantitative computed tomography (QCT), a widely utilized technique for det ermining bone mineral density (BMD) at various skeletal sites, may yield le ss precise measurements if there are operator measurement errors and other technical variables. Two-dimensional (2D) and three-dimensional (3D) QCT sc ans of the lumbar spine of 21 women were compared in order to investigate t he effects of potential operator variability on the precision of BMD measur ements and to examine resulting differences of these imaging approaches in clinical practice. No significant difference was found (p > 0.05) in precis ion between the 2D and 3D QCT BMD measurements owing to operator measuremen t errors on the CT scans. The variability in BMD values within numerous sma ll regions of interest (ROIs) (similar to 75 mm(2)) of cancellous bone with in a single vertebra was 10.1%, larger than the 2D or 3D BMD variability me asured in typical regions (similar to 250 mm(2)) by an order of magnitude. 3D BMD values in this population, which represented a wide range of clinica l values, were found to be significantly greater than 2D BMD values by an a verage of 5.6% (p = 0.00024) relative to the 2D QCT values. Our findings su ggest technical measurement error does not have a significant effect on pre cision of BMD measurements obtained with either QCT method. Several factors , however, including the incorporation of focal regions of higher density b one mass within the 3D QCT ROI may account for the higher BMD values compar ed with those for 2D QCT.