CT first-pass myocardial perfusion imaging is feasible using a simple semi-quantitative analysis which provides reasonable estimates of MBF.

The purpose of this study was to compare the signal intensity (SI) and image characteristics of 1st-pass (FP) CT imaging vs. MR FP in an animal model of resting myocardial blood flow (MBF). Pigs (n=5) were scanned at rest using FP MRI perfusion imaging. Gd-DTPA was injected IV at 5cc/sec at 0.04mmol/kg. FP MR was acquired at 3T with 3 axial slices 3mm thick per R-R interval during bolus injection using a saturation recovery gradient echo sequence. ECG gated FP CT was performed immediately thereafter on a 64-slice scanner in axial mode with 10 2.5mm thick slices during bolus injection (5cc/sec) of 40% dilution of iopamidol 370. MBF was measured at rest using fluorescent microspheres between studies. SI was measured for baseline and peak myocardial enhancement phases. Contrast enhancement (CE)= (peak-baseline) SI/baseline SI X 100. Rest MBF was 0.60±0.15 ml/min/g. Temporal resolution was similar by technique but more slices were imaged by CT per R-R interval (10 vs 3) and spatial resolution was superior for FP CT. Image characteristics are shown in the table . Baseline (top) and peak enhancement (bottom) images for FP CT and MRI are shown in the figure . Results were similar when undiluted contrast was used for each modality. Despite similar absolute change in SI, CE during FP CT is markedly reduced compared to FP MR due to a more than 20-fold higher baseline SI of the myocardium.

The accurate assessment of myocardial blood flow (MBF) is a potential adjunct to the anatomy of CT coronary angiography. To compare quantitative parameters from 1st-pass CT (FP CT) imaging with absolute measures of MBF in an animal model of altered MBF. A pig model of intracoronary adenosine (n=9) was used during FP CT. This produces a zone with hyperemic MBF and a control zone within a slice. Fluorescent microspheres (Mcsp) were injected into the left atrium with to determine absolute MBF concurrent with CT imaging. Pigs were placed in a 64-slice (Philips) CT with acquisition performed during IC adenosine. A 40% dilution of Iopamidol 370 (1ml/Kg) was injected IV at 5ml/sec. CT acquisition was ECG gated over 40 cardiac phases with the following parameters: 180° axial mode (pitch=0), field of view=250mmsq, 512×512 matrix, slice thickness=2.5 mm × 10 slices, temporal resolution=210ms, 120KV, 495ma. Mcsp were injected immediately following CT imaging. The heart was sectioned into 2.5mm slices to match the CT images and segmented. Time intensity curves (TIC) were generated from CT in adenosine and control zones based on Mcsp values. Mcsp coronary flow reserve (CFR) = hyperemic/control MBF, and CT CFR was derived from hyperemic/control area under curve from baseline corrected TIC (Klocke, Circ. 2001). MBF control=0.65±0.25, MBF adenosine=2.6±0.7 ml/min/g (p<0.0001). CFR=4.5±1.05, CT CFR=4.4±1.5 (p=NS). There was a significant (r=0.93, p<0.0001) correlation between CFR and CT CFR (figure ). FP CT myocardial perfusion imaging is feasible at clinically relevant radiation dosimetry and provides reasonable estimates of CFR during hyperemia.

First-pass (FP) MR myocardial perfusion (perf) imaging can quantify absolute myocardial blood flow (MBF) but images are of low signal at conventional field strength. Purpose: To determine the accuracy of absolute FP MBF measures at 3 Tesla and compare these measures with FP 1.5T using MBF by microspheres (mcsp) as the gold-standard. Methods: A pig model was used to alter MBF in a coronary artery during FP MRI (intracoronary adenosine followed by ischemia). This produces an active zone with a range of MBF and a control zone. Mcsp were injected into the left atrium with concurrent reference sampling. FP MR perf imaging was performed at 1.5T (n=8) or 3.0T (n=7) using a saturation-recovery gradient echo sequence in short-axis slices (SAX) during a bolus injection of 0.025mmol/kg gadolinium-DTPA. Fermi function deconvolution was performed on active and control ROI from SAX slices with an arterial input function from the LV cavity. These MR values of MBF were matched to mcsp values obtained from SAX slices at pathology. Results: Occlusion MBF was 0.22±.26 ml/min/g, adenosine MBF was 2.11±1.13 ml/min/g and control zone MBF was 0.70±0.22ml/min/g. The correlation of MR FP MRI with mcsp is shown in Figure 1 : 3T) r=0.95, p<0.0001, 1.5T, r=0.94, P<0.0001. The 95% confidence limits were slightly narrower at 3T (3T=0.41ml/min/g, 1.5T=0.55ml/min/g, p=NS. FP MRI characteristics were better at 3T (noise, SNR, contrast enhancement all superior at 3T). In zones where MBF<0.50ml/min/g, the correlation with mcsp was closer at 3T (r=0.55 at 1.5T, r=0.85 at 3T). Conclusion: Absolute MBF by FP perfusion imaging is accurate at both 1.5 and 3T. Signal quality is better at 3T which may confer a benefit in low MBF zones.