THE ROLE OF DUAL SOURCE CT IN ASSESSMENT OF MYOCARDIAL PERFUSION

Abstract

Coronary heart disease (CHD) showed recent increase in its incidence all over the world and also carries a serious health risk with increased morbidity and mortality. CHD leads to myocardial injury due to deficient blood supply as compared to myocardial needs. (Sampson et al., 2007)

The gold standard in evaluation of coronary arteries and detection of coronary artery stenosis is the conventional coronary angiography, yet its invasive technique limited its routine use. (Crossman, 2004)

With technological advancement, regarding multi-detector computed tomography technology, especially Flash DS-CT, which allows isotropic imaging in very fast image acquisition with remarkably increased temporal, spatial and contrast resolution, MIP, MPR and cMPR reconstructions as well as volume rendering with very high imaging quality and so the accuracy of detecting coronary artery stenosis has improved. Also it allowed combined assessment of coronary artery luminal integrity and myocardial perfusion status (cCTA and CT myocardial perfusion) at the same scan with the same amount of contrast material and radiation dose. (Crossman, 2004)

Flash DS-CT has many advantages over conventional coronary angiography which includes its noninvasive nature (Abdulla et al., 2007) and its ability to assess both coronary arteries and myocardial perfusion status at the same time (Feuchtner et al., 2011) and so Flash DS-CT has the potential to be used in patients with suspected or known coronary artery diseases. (Abdulla et al., 2007)

So, we conducted a study to elucidate the role of DSCT imaging in myocardial perfusion and coronary CTA assessment in a correlative and comparative manner in a single image acquisition considering the coronary CTA as the gold standard.

Our study, in contrast to previous studies, was based on the use of one of the most recent and advanced dual source systems using 2 tubes x 128 detector rows with 256 effective detector rows (Somatom Definition Flash; Siemens Healthcare, Forchheim, Germany).

In our study, we interpreted and analyzed the imaging data of 35 patients who had performed cardiac DSCT scan at ASUSH and met the inclusion criteria. The DSCT scanner was, unfortunately, operated in the single energy mode owing to unavailability of the dedicated post processing dual energy based perfusion software (Heart PBV cardiac CT perfusion analysis software), so the machine was operated in the single energy mode and we used the available basic perfusion software.

Two image sets were automatically created. The first of them is coronary CTA (cCTA) images for assessment of the coronary arteries for atherosclerotic changes and grading of the stenotic segments if present. The most valuable images were the row data axial images as well as the reconstructed images as volume rendered images and curved multi-planner reformats (cMPR) which enable grading of stenosis severity in a quantitative manner through semi-automated vessel analysis tool that reconstruct cross sectional images perpendicular to the long axis of the artery helping determine luminal diameters and cross sectional area that enable stenosis percentage calculation.

The second image set was the cardiac function set which depends on row images reconstruction into cardiac short axis, vertical long axis and horizontal long axis views. These views were evaluated, subjectively, in the gray scale mode with window width and level narrowing to enhance the attenuation difference between the hypo-perfused and normally perfused areas with a note is made that the attenuation difference depends on the difference of the amount of the perfused blood into the myocardium carrying the iodine based contrast material in different myocardial segments and areas. Also, these views were evaluated, semi-objectively, using the first pass and delayed enhancement tools which create colour coded and overlaid images that highlighted the hypo-perfused areas in different colour or deep colour shade. Finally, objective evaluation was made by drawing ROIs in the hypo-perfused and normally perfused areas creating numerical data that allowed analysis and comparison to be made.

We used the coronary CT angiography (cCTA) as the final reference standard for validating myocardial blood-pool deficits on DSCT.

The second, cardiac function, image set also, automatically created evaluation charts for left ventricular wall motion, thickness and thickening as well as it automatically calculated ejection fraction, myocardial mass, EDV, ESV, stroke volume and cardiac output.

Regional left ventricular wall motion assessment was done using the reconstructed short axis view images during systole and diastole as well as cine loop and 17 segments chart for any hypo- or dis-kinetic movements.

Statistical analysis of the collected results and data from our initial study using DSCT cardiac imaging during rest showed good correlation between DSCT myocardial perfusion and coronary CTA findings with overall 83% sensitivity, 87% specificity, 86% accuracy, 70% Youden’s index with 74 % positive predictive value and 92 % negative predictive value.