Feasibility Of 640-slice Volume CT Resting Myocardial Perfusion Imaging Combined With CT Coronary Angiography To Assess Myocardial Ischemia In Patients With Chronic Coronary Stenosis

BackgroundCoronary heart disease (CAD) is the most common type of heart disease. It also is the leading cause of death, especially in middle-aged and aged people. According to the Global status report on noncommunicable diseases 2014 released by World Health Organization (WHO), cardiovascular diseases (CVD) has become the leading causes of NCD deaths in 2012. The burden of disease remains high, in spite of death rates attributable to CVD declined in some high-income countries. And the increasing incidence of CVD is mainly in middle- and low-income countries, which our country was categorized into by World Bank Group (WBG). In 2012, the mortality from CAD in China is 131.64 per 100000. CAD happens when atherosclerosis plaque on arteries inner walls and making them hardened and narrowed. As it grows, blood that can flow through the arteries is declined. Consequently, they can’t satisfy the demand of heart muscle for blood and oxygen. A heart attack occurs if a blood clot suddenly severely reduces or completely cuts off the coronary blood supply, which is also called acute cardiovascular syndrome(CAS), causing damage to the heart muscle. Over time, CAD can also lead to heart failure and arrhythmias. Thus a timely and effective intervention, based on comprehensive understanding of changes in both coronary morphological features and myocardial hemodynamics, is critical for decision-making regarding further treatment. Coronary artery imaging along with cardiac perfusion is an important tool in assessing the degree of CAD. For the last decades, computed tomography coronary angiography (CCTA) has been used as a non-invasive alternative imaging of coronary. Morphological abnormalities of coronary can be evaluated more precisely by CCTA than by CAG. CT perfusion (CTP) as an adjunct to a conventional or morphologic CT examination, can routinely offer functional information.Objective1. To evaluate the accuracy of 640-slice volume CT coronary angiography (VCT-CCTA) in detecting and grading stenosis of coronary arteries, compared with conventional coronary angiography (CAG).2. To explore the feasibility of VCT resting myocardial perfusion imaging (VCT-CTP) in evaluating myocardial ischemia of chronic coronary heart disease.3. To evaluate the relationship between the transmural perfusion ratio (TPR) of left ventricular myocardium and the degree of coronary artery stenosisMaterials and Methods1. Accuracy of VCT-CCTA in detecting and grading coronary artery stenosis-comparison with CAG1.1 General informationClinical suspected or conformed CAD patients who underwent VCT-CCTA, followed by CAG in less than 2 weeks between June 2014 and February were include. Between the examinations, there were no obvious changes in the clinical condition and no interventions. The following candidates were excluded:(1)with congenital heart disease (2)previous coronary artery bypass graft surgery or coronary stent implantation.(3)contraindications to iodinated contrast material(known allergy, renal failure, cardiac failure et al).(4) pregnancy (5) high heart rate (HR) or arrhythmia.(6) claustrophobia. In total,50 patients(M36, F14) were enrolled, age from 35 to 75(57.8 ± 10.36). This study protocol was approved by institutional review board of Southern Medical University, and all patients gave written informed consent before the study.1.2 Imaging acquisitionCoronary heart disease patients have to beta blockers 24h prior to the examination. Record the heart rate, basic blood pressure, weight and height of subject, calculated the body mass index (BMI) before data acquisition. To patients with a prescan heart rate exceeding 70 beats per minute (bmp), metoprolol (20-25mg) were administered orally 1h before the CCTA to maintain the heart rate within 60-70 bmp.All the CT data were acquired by 640-slice dynamic volume CT scanner (Aquilion ONE, Toshiba Medical Systems, Otawara, Japan) with the following scanning parameters:detector width of 160 mm and 320 detector rows(collimation=320 X 0.5 mm); minimum gantry rotation time of 350 ms; tube voltage of 100 kVp, tube current of 300-400mAs(depending on BMI), slice thickness/intervel=0.5/0.25mm, field of view (FOV) 180-220mm. Coronary calcium scanning was performed first, followed by prospective electrocardiogram (ECG)-gated volume scan, ranging from the tracheal bifurcation to heart under the diaphragm about 1.5cm, with full dose window of 60-80% of the cardiac cycle.Through the median vein in the forearm, a dual-head power injector was used to injected a volume of 50-60 mL of contrast media (320mgI/ml) at a flow rate of 5.0 mL/s, followed by 20 mL of 50% contrast/saline. Start the CCTA scan when the density of thoracic aortic proximal reaches a150-180Hu.1.3 Image processing1.3.1 Choose the best phase to reconstructed with a slice thickness/interval of 0.5/0.25 mm. Transferred the reconstructed image data to a workstation (Toshiba Vitreal) for postprocessing. Combine axial images with maximum intensity projections (MIP), multi planar reformation (MPR), volume rendering (VR) and curved planner reformation (CPR) to evaluate the morphological features of coronary as comprehensively as possible.1.4 Image analyse1.4.1 Coronary artery sectionAccording to the standardized coronary segmentation recommend by American Heart Association (AHA), coronary were divided into 16 segments. The right coronary artery (RCA) including the proximal (S1), the middle (S2), the distal (S3) segment and the posterior descending artery (PDA, S4), posterior artery of left ventricle (PLA, S16). The left coronary artery (LCA) consists of left main coronary artery (LM, S5), left anterior descending artery (LAD) and left circumflex artery (LCX). LAD consists of the proximal (S6), the middle (S7), the distal (S8) segment and 2 diagonal branches (D1=S9, D2=S10). LCX is divided into the proximal (S11), the distal (S13) segment and 2 obtuse marginal branches (OM1=S12, OM2=S14), posterior artery of left ventricle from (L-PLA, S15).1.4.2 Image quality evaluationAccording to guidelines for the interpretation and reporting of CCTA from Society of Cardiovascular Computed Tomography (SCCT), Criteria for DVCT-CCTA images was as follows:5 excellent:all segments was clear displayed continuously without any motion artifacts.4 good:vascular continuous, with only minor artifact at the flexure of RCA (junction of S1 and S2 or of S2 and S3).3 moderate:with minor artifact, segmental wall was slightly blurred.2 general:vascular with moderate motion artifact, segmental wall and surrounding structures were fuzzy.1 point poor:vascular displayed discontinuously, with obvious "stairstep artifacts" that can’t be evaluated.1.4.3 Quantitative Stenosis GradingQuantification ranges used for quantitative stenosis grading in the study:0 Normal:Absence of plaque and no luminal stenosis1 Mild:Plaque with< 50% stenosis2 Moderate:50%-75% stenosis3 Severe:≥75% stenosisAll VCT-CCTA image was assessed by two radiologists experienced in the evaluation of CTA, who didn’t know patient’s any clinical information. When it comes to controversy, then negotiated and finally decided it.1.6 CAGData acquired by INTEGRIS Allura 12"&15" Monoplane DSA (PHILIPS, Philips Medical Systems Nederland B.V., The Netherlands). Seldinger technique was applied in arterial sheath placement of femoral or radial artery. Injected contrast agent (370mgI/ml) respectively 4-6ml to the LCA and RCA. Images of LCA including 4-6 standard posture, RCA including 2-3 were taken.1.7 Statistical analysisThe SPSS 20.0 software was applied in this study. Marginal Homogeneity Test was used to evaluate the difference between VCT-CCTA and CAG in diagnosis of different segments. Kappa test was used to evaluate the consistency between VCT-CCTA and CAG in quantitatively grading stenosis. Served CAG diagnosis as the standard of reference, we respectively calculated the sensitivity、specificity and accuracy of VCT-CCTA in diagnosis of coronary with stenosis.2. Feasibility of resting VCT-CTP to assess myocardial ischemia in patients with chronic coronary stenosis2.1 clinical informationVCT-CTP images of 176 subjects were analysied. They all underwent coronary angiography with VCT in radiology department between November 2014 and February 2015. The inclusion and exclusion criteria was the same as part1. Besides, candidates that present "left-dominant" also be excluded, since the feeding artery of the myocardial segments analysed in this study is different from those "right-dominant" people. In total, there are 106 male and 70 female enrolled, age from 35-75, average 54.92±11.46. All participants were divided into four groups (Normal, mild stenosis, moderate and severe stenosis) according to their CCTA results.2.2 Imaging acquisition was the same as part 12.3 Image processing2.3.1 VCT-CCTA image processing and analysis were the same as part 1.2.3.2 resting CTP image processingReconstruct every 5 phase resting CTP image from 60% to 80% of the cardiac cycle with 1.0mm slice thickness and 1.0mm overlap. Transferred the total 6 data package to Toshiba Vitreal workstation, applied the Cardial Functional CT software package to draw the outline of endocardium and epicardium automatically in the shot axis images. Adjust the apical of left ventricle (LV) to the same line with the long axis of LV. Checked and manual adjusted outline of endocardium and epicardium of each phase images in the left ventricular to make sure the calculation is accuracy (papillary muscles and interventricular septum must be the calculation). After these, the software picked out the end-diastolic (ES), "end-systolic (ED)" phase and calculated the functional index of LV, such as end-diastolic volume (EDV), end-systolic volume (ESV), stroke volume (SV) and ejection fraction (EF). However, the full dose window of our CT protocol didn’t cover the whole systolic phase, ES phase and EDV, ESV, SV, EF were inaccurate. And we use the ED phase image but not the other index in the following analysis.Input the ED phase images into the clinical cardiac myocardial perfusion software package. After checked and manual adjusted outline of epicardium and endocardium, mean transmural perfusion ratio (TPR) of each myocardial segment of LV was calculated (TPR= subendocardium/subepicardiurn).17-segment system:LV wall were divided into 4 parts (basal, mid-cavity, apical and apex) to the long axis of the heart,17-segments, according to recommendation from AHA, considering cardiac anatomy (the territories of RCA, LAD and LCX) and clinical needs (accurate intra- and cross-modality comparisons for both research and clinical patient management). The 17 segments are S1-basal anterior, S2-basal anteroseptal, S3-basal inferoseptal, S4-basal lateral, S5-basal inferolateral, S6-basal anterolateral, S7-mid anterior, S8-mid anteroseptal, S9-mid inferoseptal, S10-mid lateral, S11-mid inferolateral, S12-mid anterolateral, S13-apical, S14-apical septal, S15-apical inferior, S16-apical lateral, S17-apex. In this system, S3,4,9,10, and 15 are assigned to RCA distribution when it is "right-dominant". S1,2,7,8,13,14, and 17 are assigned to the LAD.S5,6,11,12, and 16 are assigned to LCX.2.4 Statistical analysisThe statistical software SPSS 20.0 package was applied. If P<0.05, considering the correlation or difference was statistical significant. Differences among groups were evaluated by one-way ANOVA (depending on P value of the Homogeneity test, if.P>0.05, by LSD test, otherwise, by Dunnett’s T3 test). Relationship between TPR values and degree of coronary stenosis were analyzed by Spearman partial correlation analysis.Results1. Accuracy of VCT-CCTA in detecting and grading coronary artery stenosis -- comparison with CAG1.1 The total number of Normal, Mild(<50%), Moderate(50%-75%) and Severe(≥75%) stenosis diagnosed by VCT-CCTA was 469,119,80 and 53, by CAG 485,107,75 and 54 respectively.>50%stenosis was more offen occurred at the proximal and middle (S6 and S7) segment of LAD.1.2 In analysing coronary morphological changes by coronary segmentsThe exist difference between VCT-CCTA and CAG wasn’t statistical significant (MH=1.000, <0.001,<0.001,1.414,1.414,1.000,1.000,<0.001,1.414,1.732, 1.000, <0.001,1.414,1.414,1.000,1.000; P=0.317,1.000,1.000,0.157,0.157,0.317, 0.317,1.000,0.157,0.083,0.317,1.000,0.157,0.157,0.317,0.317).The consistency between VCG-CCTA and CAG was statistical significant (Kappa=0.963,0.938,0.904,0.864,0.891,0.973,0.973,0.908,0.925,0.817,0.962, 0.884,0.922,0.766,0.762,0.913; P<0.001,<0.001,<0.001,<0.001,<0.001,<0.001, <0.001,<0.001,<0.001,<0.001,<0.001,<0.001,<0.001,<0.001,0.002,<0.001).1.3 The accuracy of VCT-CCTA—compared with CAG1.3.1 defined normal arteries as negative, arteries with stenosis as positive:the accuracy of VCT-CCTA in segment 1-16 was 98.00%,98.00%,98.00%,97.83%, 98.00%,100.00%,100.00%,98.00%,95.83%,93.33%,100.00%,97.61%,100.00%, 93.94%,93.33%,97.63% respectively; Youden index was 96.88%,96.15%,97.37%, 97.37%,97.44%,100.00%,100.00%,97.30%,93.55%,91.89%,100.00%,96.88%, 100.00%,93.10%,92.31%,97.22%.1.3.2 defined normal and mild stenosis as negative, moderate and severe as positive:the accuracy of VCT-CCTA in segment 1～16 was 100.00%,100.00%, 100.00%,97.83%,98.00%,98.00%,100.00%,100.00%,100.00%,100.00%,98.00%, 97.62%,98.00%,100.00%,100.00%,100.00%,100.00%,100.00%. Both true and false positive of segment 14 and 15 were 0. Youden index of segment 1-13,16 was 100.00%,100.00%,100.00%,97.67%,97.96%,96.00%,100.00%,100.00%, 100.00%,100.00%,97.50%,83.33%,97.73%,100.00% respectively.1.3.3 defined normal, mild and moderate stenosis as negative, severe as positive: the accuracy of VCT-CCTA in segment 1-16 was 100.00%,98.00%,98.00%, 100.00%,100.00%,100.00%,98.00%,98.00%,100.00%,100.00%,100.00%, 100.00%,98.00%,100.00%,100.00%,100.00%. Both true and false positive of segment 14 and 15 were 0. Youden index of segment 1-13,16 was 100.00%,66.67%, 66.67%,100.00%,100.00%,100.00%,97.50%,66.67%,100.00%,100.00%, 100.00%,100.00%,97.83%,100.00%.2. Feasibility of resting VCT-CTP to assess myocardial ischemia in patients with chronic coronary stenosis2.1 The TPR values in the territories of normal RCA, LAD, LCX group was 1.17 ±0.08,1.19 ±0.12 and 1.14 ±0.06, while respectively 1.07±0.10,1.10±0.13 and 1.06±0.05 in group with severe stenosis.2.2 RCA had some but not huge influence on LV myocardial perfusion. Among segments assigned to RCA, S9-mid inferoseptal was most vulnerable to coronary stenosis, and TPR was negative correlation with degree of RCA stenosis(r=-0.513, P<0.001). When RCA with moderate stenosis, there was statistical significance decline of TPR in both S3 and S9, with severe stenosis, S4-basal lateral and S10-mid lateral was suffering too.2.3 LAD stenosis had important influence on LV myocardial perfusion. Among segments assigned to LAD, TPR decline at mid-cavity and apical were earlier than at the basal part, anteroseptal zone earlier than inferolateral zone. S13-apical anterior response most sensitively to LAD stenosis (r=-0.568, P<0.001). The TPR of S13, S7-mid anterior, S8-mid anteroseptal and S14-apical septal were significant declined.2.4 Influence of LCX stenosis on LV myocardial perfusion is huge too. LCX blood flow has stronger effect on basal part than that on other segments assigned to LCX. In S5-basal inferolateral and S6-basal anterolateral, TPR was negative correlated with degree of LCX coronary (r=-628,-560; P< 0.001,<0.001).conclusion1. The diagnosis of VCT-CCTA in evaluating morphological changes of coronary segments was close to CAG, especially in coronaries with moderate or severe stenosis.2. Stenosis of RCA influenced more obviously on S9 than other segments.3. Stenosis of LAD influenced more obviously on S13, S7, S8 and S14 than other segments.4. LCX influenced more obviously on S5 and S6 than other segments5. VCT resting CTP may be helpful to provide informations of both coronary morphological changes and myocardial perfusion in patients with CAD.