The Reliability Of Volume Computed Tomographic Digital Subtraction Angiography In Cerebral Vessels

PART ONE:EXPERIMENTAL STUDY OF THE INFLUENTIAL FACTORS OF COMPUTED TOMOGRAPHIC ANGIOGRAPHY ON VASCULAR CALIBER MEASUREMENTSObjective:To investigate the influence of scanning parameters,manual and automatic vessel analysis on the accuracy of vascular calibre measurements in vessel phantoms for CT angiography. To establish the calibration criteria of opacity and window width on three dimension reconstruction of CTA.Methods:9 vascular models were filled with contrast material of nine different densities,to measure the diameter by vernier calipe as their classic calibre. CTA were performed to 9 models,data were obtained with different scanning parameters, including current modification of 150 mA,200 mA,250 mA and 300 mA,voltage modification of 140 kV,120 kV,100 kV and 80 kV. Vascular calibre and CT number were measured on 4 vascular models. Vascular calibre was measured by different sizes of field of view,opacity in VR,and window width in MIP on current of 300 mA and voltage of 120kV. Correlations among them were analyzed. The last,we evaluated the difference between the diameter measured by the automated vessel analysis software and the true inner diameter of the vascular model. 30 clinical data of window width and window level of MIP reconstruction for VCTDSA were compared with experimental results.Results:(1)Standard opacity of volume rendering was negatively correlated with vessel calibre in vascular models(r = - 0.847,P = 0.000).(2)Maximum intensity projection window width was positively correlated with vessel calibre in vascular models(r =0.983,P = 0.000).(3)There was no correlation between field of view and vessel caliber measurement in vascular models(r = -0.042,P = 0.381).(4)There was prominent widening of window width in MIP reconstruction , and the calibres were exaggerated as noted on CT angiography.(5) With the increase of kV, CT numbers and vessel calibre decreased non-linearly.There was no significant difference in CT numbers and vessel calibre by current modification.(6) Automated vessel analysis overestimated the inner diameter of the vascular model,maximum error was 2.7mm. With the increase of CT numbers,the more accurate diameter of the vascular model was.Conclusion:The accuracy of the calibre measurements by vascular model for CT angiography was affected by opacity in VR reconstruction, window width in MIP reconstruction and kV. In order to ensure the reliability and accuracy of the vascular diameter measurement,it is necessary to adjust these parameters according to intravascular density of the contrast material. It is important to apply the correction of vessel analysis on cerebrovascular CTA analysis and diagnosis, otherwise, vessel diameter measurements by automatic vessel analysis will be overestimated.PART TWO:COMPARISON BETWEEN VCTDSA AND ROUTINE CTA ON CALIBER MEASUREMENTS OF CEREBRAL VESSELSObjective:To evaluate the reliability of volume computed tomographic digital subtraction angiography (VCTDSA) on caliber measurements of cerebral vessels to compare with routine CTA .Methods:From December 2008 to March 2009,VCTDSA datas of 60 patients without cerebrovascular disease in our hospital were analyzed. The circle of Willis on three dimension volume rendering of VCTDSA and routine CT angiography were displayed with region view. The first measurements on regulative opacity by routine CTA.In the subsequent measurements,manual opacity corrections by CT number of cerebral vessels in VCTDSA were applied.Vascular calibre was measured on bilateral ACA,MCA and ICA by VCTDSA and routine CT angiography.Results:Between routine CTA and VCTDSA , there were significant differences in caliber measurements of cerebral vessels on regulative opacity by routine CTA (P<0.05). Calibre of VCTDSA group was smaller than the routine CTA group. But,there were no significant differences in caliber measurements of cerebral vessels on manual opacity corrections by CT number of cerebral vessels in VCTDSA (P>0.05). Calibre of VCTDSA group were equal to the routine CTA group. CT number of cerebral vessel decreased about 65 HU after subtraction.Conclusion:For manual opacity corrections,calibre of VCTDSA were proved to consistent with routine CTA,and caliber measurements of cerebral vessels were reliability and accuracy. Calibration of VCTDSA can be achieved in criteria of living morphology. VCTDSA would be wide application in the future,it can be prefer to the non-invasive imaging on cerebrovascular diseases.