Quantitative myocardial perfusion measurement using CT Perfusion: a validation study in a porcine model of reperfused acute myocardial infarction

Aaron So, Jiang Hsieh, Jian-Ying Li, Jennifer Hadway, Hua-Fu Kong, Ting-Yim Lee

We validated a CT perfusion technique with beam hardening (BH) correction for quantitative measurement of myocardial blood flow (MBF). Acute myocardial infarction (AMI) was created in four pigs by occluding the distal LAD for 1 h followed by reperfusion. MBF was measured from dynamic contrast enhanced CT (DCE-CT) scanning of the heart, with correction of cardiac motion and BH, before ischemic insult and on day 7, 10 and 14 post. On day 14 post, radiolabeled microspheres were injected to measure MBF and the results were compared with those measured by CT perfusion. Excised hearts were stained with 2,3,5-triphenyltetrazolium chloride (TTC) to determine the relationship between MBF measured by CT Perfusion and myocardial viability. MBF measured by CT perfusion was strongly correlated with that by microspheres over a wide range of MBF values (R = 0.81, from 25 to 225 ml min(-1) 100 g(-1)). While MBF in the LAD territory decreased significantly from 98.4 ± 2.5 ml min(-1) 100 g(-1) at baseline to 32.2 ± 9.1 ml min(-1) 100 g(-1), P < 0.05 at day 7 and to 49.4 ± 9.3 ml min(-1) 100 g(-1), P < 0.05 at day 14, the decrease in remote myocardium (LCx territory) from baseline (103.9 ± 1.9 ml min(-1) 100 g(-1)) was minimal throughout the study (90.6 ± 5.1 ml min(-1) 100 g(-1) on day 14 post, P > 0.05). TTC staining confirmed incomplete infarction in the LAD territory and no infarction in the LCx territory. Microvascular obstruction in infarcted tissue resulted in no-reflow and hence persistently low MBF in the reperfused LAD territory which contained a mixture of viable and non-viable tissue. CT perfusion measurement of MBF was accurate and correlated well with histology and microspheres measurements.