Bedeutung des Volume Rendering in der Diagnostik der Koronargefäße mittels Mehrschichtspiral-CT
Dedden, Katrin Margarete; Schmitz-Rode, Thomas (Thesis advisor)
Aachen : Publikationsserver der RWTH Aachen University (2005)
Dissertation / PhD Thesis
Aachen, Techn. Hochsch., Diss., 2005
The intention of this study is on the one hand to compare the measurement precision in determining vessel diameters of the 3D-volume rendering techique (3D-VRT) to quantitative coronary angiography (QCA) and on the other hand to challenge the diagnostic value of 3D-VRT. For 53 patients (42m, 11f), retrospectively ECG-gated MSCT of the heart was performed as well as quantitative X-ray coronary angiography. The standardized examination protocol included a collimation of 4x1mm,a tube voltage of 120kV with an effective tube current of 400 mAs and a tube rotation time of 500ms. An effective slice thickness of 1.25mm and a reconstruction increment of 0.6mm were chosen for image reconstruction. During examination 120ml of a non-ionic contrast material were administered by injection in a cubital vein with a flow-rate of 2.5ml/s. Using the Adaptive-Cardio-Volume algorithm, 3D-VR image series were calculated every 10% (30-80%) of the RR interval. For further analysis the images were transferred to an external workstation. Using a 3D-VR technique (512x512 matrix), 4770 coronary artery segments (AHA-classification) were evaluated for their visibility to classify them in a 4-point grading scale. In a second step, the diameter of each coronary artery of 50 patients in the individually best reconstruction was measured at the origin as well as 1cm, 3cm and 5cm distally. For each measurement from the 3D-VRT image the minimum, maximum and mean diameters were determined and compared with QCA- diameters using the Bland-Altman method. Finally, coronary artery stenosis were detected using both techniques. Further more, the diameter of 34 stenoses, detected on QCA, was measured as well as the vessel diameter 1cm proximal and distal of the stenosis. The measurement results of the 3D-VRT were compared to QCA results. 1851 (38.8%) of 4770 coronary artery segments were assessable (score 2-3), 1566 (32.84%) were scored with 1 point, meaning visible but not assessable and 1353 (28.36%) were not visible. Image reconstruction at 70% RR-interval shows for 13 coronary artery segments the best visibility with 51.95% assessable segments (min: 15.09%, max: 92.45%). The proximal segments were more often assessable when compared to the distal segments. Spearman-rank- order-correlation between visibility score and average heart rate or range of heart rate of each patient does not reveal any dependency (min: -0.3661, max: -0.5536 or min: 0.0156, max: 0.1320). Using the Bland-Altman method the difference between vessel diameter measured in 3D-VRT and QCA was evaluated in 581 measuring points (97 for quantification of coronary artery stenoses). The comparison of the maximum diameter of 3D-VRT and QCA shows the lowest differences (min: -0.007mm, max: -0.447mm) with a mean variance of 0.2552mm (min: -0.007, max: 2.184mm). Pearson correlation between diameter difference and average heart rate does not reveal any correlation (min: 0.0022, max: 0.2495). Coronary artery stenoses were detected with a sensitivity of 85% and a specificity of 98.26%. When compared to QCA, the ability of 3D-VRT to quantitatively assess coronary artery diameters and coronary artery stenoses is insufficient for clinical purposes. The good visibility and assessment of proximal vessel segments give reason to expect that current problems can be overcome in the near future.