Abstract

Article Series: Dual Energy CT – Scientific Evidence and Clinical Application (3/7) – Thoracic Imaging

posted by Thorsten R. C. Johnson, M.D. | Dec 6, 2010

This article is part of the seven-article series on “Dual Energy CT – Scientific Evidence and Clinical Application”. T. Johnson, MD, talks about Dual Energy applications in thoracic imaging comprising lung perfusion, lung ventilation, pulmonary nodules, and myocardial perfusion.

Thoracic Imaging

Lung Perfusion

Figure 2 Color coded superimposition of lung perfusion in axial and coronal plane showing a sub-segmental perfusion defect.

One of the most important applications of Dual Energy CT is the assessment of lung perfusion [29]. Actually, the technique is based on a static dataset representing the iodine distribution in the lungs, so there is no dynamic information. However, provided that the acquisition is obtained at a sensible delay after the arrival of the contrast material, the iodine map should closely match perfusion. Considering that dynamic CT acquisitions require considerable dose, this technique based on a single Dual Energy scan can provide an equivalent clinical information at comparatively low dose [2].
Initial animal studies had shown that the sensitivity for pulmonary embolism can be increased [30], also in correlation to scintigraphy [31]. Several studies in patients confirmed that the technique does identify perfusion defects due to pulmonary embolism [32-35], also in good agreement with scintigraphy [11]. The technique also works in children with appropriate protocol alterations [36].
Several studies reported problems in differentiating true perfusion defects from beam hardening artifacts which can occur in the presence of dense contrast material in the central veins and due to respiratory or cardiac motion [37-38]. However, optimized multi-phase protocols can help to reduce these artifacts [39]. Apart from washing the contrast material out of the central veins with a saline chaser bolus, it is important to apply a rather long delay in order to allow the iodine to pass from the pulmonary arteries into the parenchyma. Also, true perfusion defects have a morphology which corresponds to pulmonary segmental anatomy and can therefore be distinguished from artifacts, most easily in coronal reconstructions [40]. Apart from the additional functional aspect of the perfusion information, it can also increase the sensitivity for pulmonary embolism by identifying small sub-segmental embolism which is not visible as a clot in the small artery but still detected as small peripheral perfusion defect [41].

Additionally, it is possible to highlight vessels which contain less iodine than others with a specific algorithm. Although this technique has only a moderate specificity in clinical application, it can be used as a tool to increase conspicuity and negative prediction [42]. Some authors also advocate the use of the low energy dataset to increase iodine contrast and use less contrast material. However, this will be limited in large patients, and considering dose efficiency and image noise it would be advisable to apply only a low-energy spectrum if saving contrast material is the main aim [43].

Meanwhile, several clinical studies have looked into diagnostic benefits in specific pulmonary diseases. One aspect being investigated is the quantification of perfusion defects to assess the global severity and clinical relevance of pulmonary embolism, assuming that the limitation of blood oxygenation is of higher clinical importance than the clots in the vessels. Perfusion defect scoring systems have been introduced for this task and seem to correlate well with other clinical parameters like right ventricular diameter ratio [44].
Pulmonary hypertension is another field in which Dual Energy CT offers specific diagnostic benefits. If there are no changes in the pulmonary parenchyma like emphysema or fibrosis, the etiology of pulmonary hypertension mostly remains unclear or is classified as idiopathic in CT. With Dual Energy CT, an inhomogeneous perfusion with multiple peripheral defects can reveal previous or chronic recurrent pulmonary embolism as cause of pulmonary hypertension [45]. Also, Dual Energy perfusion imaging can help to assess the severity emphysema by quantifying the total perfusion [46] or to differentiate ground glass opacities of vascular and bronchioalveolar origin [47].

Lung Ventilation

Figure 3 Volume rendered image of xenon enhanced lung ventilation.

While pulmonary perfusion imaging with iodinated contrast material meanwhile represents a routine application of Dual Energy CT, ventilation imaging requires the application of xenon gas as contrast material and is therefore rather intricate [48]. Considering that only moderate concentrations have to be breathed for a few seconds, the safety profile should be even better than reported for brain perfusion imaging. Still, close monitoring seems mandatory at the present state of knowledge.

Initial studies have confirmed that the visualization of xenon gas in the lungs as a surrogate for ventilation is technically feasible [49] and shows ventilation changes behind bronchial obstruction. Clinical studies in specific pulmonary diseases showed impaired regional ventilation in children with bronchiolitis obliterans [50], ventilation defects in asthma patients [51] and dynamic ventilation changes in chronic obstructive pulmonary disease [52].

The combination of xenon based ventilation and iodine based perfusion imaging offers the perspective of a comprehensive one-stop-shop assessment of pulmonary anatomy and morphology, high-resolution structure of the parenchyma as well as ventilation and perfusion as most important functional parameters [53].

Pulmonary nodules

In the evaluation of pulmonary nodules, the assessment of iodine enhancement and the detection of calcifications can be improved with Dual Energy CT [54]. Benefits of the Dual Energy technique in comparison to repeated single energy acquisitions are the reduced dose and the simultaneous acquisition, so that there are no problems concerning respiratory position or partial volume effects which have considerable impact in small nodules with the surrounding air.
It is essential to reliably identify calcifications and eliminate the corresponding voxels from the quantification of iodine enhancement, because the photo effect of calcium would be misinterpreted as iodine enhancement, i.e. a sign of benignancy would be quantified as degree of malignancy. A technical study on a single source CT system found variable results with influence of body size, anatomic region and nodule size [55]. However, a clinical study on a Dual Source CT system showed that the technique works well and a prediction of malignancy based on the assessment of calcifications and iodine enhancement is quite reliable [56].

Myocardial Perfusion

One of the most challenging applications of Dual Energy CT is the assessment of myocardial perfusion [57]. Due to the continuous rapid motion of the heart, the scan has to be gated or triggered like for coronary CT angiography. The aim is of course to assess the coronary arteries as well as myocardial perfusion in a single acquisition.
With the first generation Dual Source CT scanner, this requires giving up some temporal resolution, so the two tubes and detectors perform simultaneous gated acquisitions at different tube voltages. This is problematic because the main benefit of Dual Source CT in cardiac imaging, the high temporal resolution and the resulting robustness in variable and high heart rates [58-61], is lost. Also, the photon output at 80kVp is limited and hardly sufficient with half-scan segments in gated acquisitions, especially in large patients.
The second generation Dual Source CT system offers two major advantages to resolve these issues:

A combination of projection data from dual energy acquisitions is feasible so that the high temporal resolution of 75ms can be preserved.

The application of the tin filter on the high energy spectrum makes it possible to use 100kVp as lower energy, resulting in sufficient transmission and projection data even in large patients.

Initial studies investigating the dose of Dual Energy coronary CTA on the first generation system showed a reduced dose in comparison to single energy acquisitions and, expectedly, some more motion artifacts [62]. Initial studies investigating the diagnostic value revealed a similarly high diagnostic accuracy for myocardial perfusion with reference to SPECT as for coronary CTA with reference to conventional angiography [63-66]. Its seems that the sensitivity of the perfusion analysis is somewhat limited but specificity is superior to coronary CTA, so it may to some degree decrease the problem with false-positive findings in cardiac CT. Studies additionally applying adenosine stress showed a good accuracy in the detection of reversible perfusion defects with reference to stress perfusion MRI [67]. These studies were performed on the first generation system, while trials evaluating the performance of the second generation Dual Source CT with its added functionality for cardiac Dual Energy imaging are still lacking.

Although the assessment of myocardial perfusion may seem the most promising application in cardiac imaging, there are also studies investigating the quantification of coronary artery iodine content as a marker for ischemia, showing a good diagnostic accuracy with reference to SPECT [68]. Approaches using Dual Energy CT to visualize late myocardial enhancement as a marker for scars showed only a limited diagnostic value in comparison to MRI [69]. Attempts to improve stent imaging by correcting beam hardening showed no advantages as blooming artifacts remain the main problem [70]. Another application of Dual Energy CT of the heart is the quantification of iron in thalassemia patients. The results of an initial clinical study confirmed the feasibility and good correlation to T2* quantification in MRI [71].