Quantified dual energy computed tomography perfusion imaging using myocardial iodine concentration: Validation using CT derived myocardial blood flow and invasive fractional flow reserve in a porcine model.

BACKGROUND: Myocardial CT perfusion imaging with dual energy (DE-CTP) can produce myocardial iodine perfusion maps. This study evaluated the accuracy of first pass myocardial iodine concentration in DE-CTP compared to CT derived dynamic myocardial blood flow (MBF) to determine regional myocardial ischemia in an animal model of coronary stenosis using invasive Fractional Flow Reserve (FFR). METHODS: Seven anaesthetised pigs (mean weight 51 ±â€¯4 kg) had a graded coronary artery stenosis produced in six vessels (plus one control animal) using a methacrylate plug with FFR recorded in the target artery (ischemia = FFR<0.80). During adenosine vasodilation, dynamic myocardial CTP and DE-CTP imaging was performed. Using vendor supplied applications, matching regions of interest (ROIs) were drawn in myocardial segments supplied by the target coronary artery to compare the two techniques. RESULTS: FFR correlated strongly to MBF (r = 0.81) and modestly to myocardial iodine concentration (r = 0.65) and myocardial CT attenuation (r = 0.62) (p < 0.0001 each). The correlation to FFR was stronger using relative ratios (absolute value/reference value of normal segments) than absolute values for MBF (r = 0.86), myocardial iodine concentration (r = 0.80) and CT number (r = 0.79) (p < 0.0001 each). Comparing normal and ischaemic territories there were significant differences in MBF (96 ±â€¯14 vs. 27 ±â€¯18 ml/100 ml of tissue/min, p < 0.0001), myocardial iodine concentration (3.5 ±â€¯1 vs. 1.0 ±â€¯0.7 mg/ml, p < 0.0001) and myocardial CT number (89 ±â€¯9 vs. 73 ±â€¯14 HU, p = 0.002). Myocardial iodine concentration had 91% sensitivity and 98% specificity for detecting FFR <0.8. CONCLUSION: Quantified myocardial iodine content from first pass DE-CTP correlates with CT derived myocardial blood flow and FFR and accurately discriminates ischemic territories in a porcine model. The accuracy and utility of myocardial iodine content in DE-CTP warrants further investigation in a clinical population with FFR as a reference standard.

Development of a Porcine Model of Coronary Stenosis Using Fully Percutaneous Techniques Suitable For Performing Cardiac Computed Tomography, CT-Perfusion Imaging and Fractional Flow Reserve.

BACKGROUND: The aim of this study was to develop and describe percutaneous coronary angiographic techniques to create a porcine model of acute coronary stenosis with methacrylate plugs that can by assessed using fractional flow reserve (FFR), invasive coronary angiography and coronary computed tomographic (CT) perfusion imaging without introducing artefacts associated with surgical models. METHODS: Following animal care and institutional approval and using percutaneous coronary catheterisation techniques within an animal laboratory we introduced precision drilled methacrylate plugs into one of the three main coronary arteries of 10 experimental female pigs. Coronary pressure wire measurements were performed across the experimental stenosis for the calculation of FFR. Invasive coronary angiograms were obtained in stenosed arteries. Animals were transported to a dual source CT scanner (Siemens Healthcare, Forcheim, Germany) and CT perfusion imaging was performed. RESULTS: Ten (10) pigs were investigated with seven data sets obtained. Three (3) pigs expired prior to CT imaging secondary to pneumothorax, high grade coronary stenosis with induced cardiac arrhythmia and iatrogenic air embolism. Graded coronary stenosis was produced in six pigs in the LAD (2), LCX (2) and RCA (2) territories and one animal served as a control. Fractional flow reserve ranged from 0.21 to 0.91. Myocardial blood flow derived from dynamic CT perfusion imaging ranged from 3.5 to 136.7ml/100ml of tissue/minute. No artefacts from the deployment of the methacrylate plug, nor the plug itself, were identified. CONCLUSIONS: Fully percutaneous preparation of a pig model of acute coronary stenosis is feasible and provides subjects for imaging that are free of surgically induced artefact. This technique is substantially less expensive than surgically induced coronary stenosis and can be performed using standard catheterisation techniques with mobile imaging equipment. The technique is extendable to produce multivessel acute coronary stenosis and can be used for multimodality imaging.