Original Research
Cardiopulmonary Imaging
June 20, 2014

Incremental Value of Pharmacological Stress Cardiac Dual-Energy CT Over Coronary CT Angiography Alone for the Assessment of Coronary Artery Disease in a High-Risk Population

Abstract

OBJECTIVE. The purpose of this article is to prospectively determine the value of stress dual-energy CT (DECT) myocardial perfusion imaging to coronary CT angiography (CTA) for the assessment of coronary artery disease (CAD) in a high-risk population.
SUBJECTS AND METHODS. We prospectively enrolled 29 consecutive patients who were referred for cardiac SPECT examinations for known or suspected CAD to also undergo pharmacologic stress cardiac DECT. In 25 patients, cardiac catheterization was available as the reference standard for morphologically significant stenosis. The performance of coronary CTA alone, DECT myocardial perfusion alone, and the combination of both was assessed by calculating sensitivity, specificity, and AUC values.
RESULTS. For morphologically significant stenosis, coronary CTA alone and myocardial DECT assessment alone had 95% sensitivity and 50% specificity. The combined approach yielded 100% sensitivity and 33% specificity if either was positive and 90% sensitivity and 67% specificity if both were positive. The AUC value was highest (0.78) if both were positive. For hemodynamically significant lesions, coronary CTA alone had 91% sensitivity and 38% specificity, and DECT alone had 95% sensitivity and 75% specificity. The combined approach yielded 100% sensitivity and 38% specificity if either was positive and 86% sensitivity and 75% specificity if both were positive. AUC values were highest for DECT alone (0.85) and the “both positive” evaluation (0.80).
CONCLUSION. The combined analysis of coronary CTA and DECT myocardial perfusion reduces the number of false-positives in a high-risk population for CAD and outperforms the purely anatomic test of coronary CTA alone for the detection of morphologically and hemodynamically significant CAD.
Coronary CT angiography (CTA) is the primary noninvasive imaging technique for morphologic assessment of the coronary arteries and has gained widespread clinical acceptance, particularly in patients with a low-to-intermediate risk of coronary artery disease (CAD), because of its excellent negative predictive value [1]. CT can also be used to simultaneously assess the status of the myocardial blood supply, thus offering a comprehensive evaluation of coronary heart disease [2].
One of the described avenues for assessment of the myocardial blood supply consists in dual-energy CT (DECT) acquisitions during first-pass arterial enhancement. Although this approach does not truly represent “perfusion,” but rather a snapshot of the myocardial iodine distribution during imaging, for simplicity this method shall be referred to as “perfusion” imaging for the purpose of this contribution. Myocardial perfusion DECT can be performed using dual-source CT [3] or single-source CT with rapid kilo-voltage switching [4]. It has been found that DECT has advantages over single-energy CT for performing first-pass myocardial perfusion imaging [5]. Specifically, DECT reduces beam-hardening artifacts and directly visualizes myocardial iodine content. Thus, perfusion defects and late enhancement are often more easily recognized on DECT [6]. A limited number of studies have compared DECT myocardial perfusion imaging to SPECT or cardiac MRI [3, 610]. These studies have found good accuracy of DECT for myocardial perfusion defects.
The performance of coronary CTA is affected by the pretest risk of CAD. In particular, the specificity of coronary CTA can be limited by extensive coronary calcification, which can cause artifacts that may lead to false-positive findings. Accordingly, current guidelines consider coronary CTA inappropriate in high-risk populations [11]. With such patients, the combined approach of coronary CTA and myocardial CT perfusion assessment could be useful to reduce the number of patients who are needlessly referred to catheter angiography because of false-positive coronary CTA findings.
Therefore, the purpose of this prospective investigation was to determine the value of adding adenosine stress DECT myocardial perfusion imaging to coronary CTA for CAD assessment in a high-risk population using SPECT and catheter angiography as the reference standards.

Subjects and Methods

Study Population

This study was institutional review board approved and HIPAA compliant. Written informed consent was obtained from all patients after a thorough explanation of the study purpose and the involved risks.
We prospectively enrolled 30 consecutive patients who were clinically referred for cardiac SPECT examination for known or suspected CAD. Patients were considered not eligible for the study if they were a pregnant or nursing woman, had received cardiac-related surgical intervention within the 30 days before the SPECT examination, were unwilling to comply with the protocol requirements, or suffered from any of the following: active wheezing, asthma, second or third degree heart blocks (without functioning pacemaker), acute psychiatric disorder, substance abuse, claustrophobia, impaired renal function (creatinine level, > 1.5 mg/dL), or any unstable condition.
Of the 30 patients who were enrolled, one was excluded from the study population because of extensive motion artifacts that occurred during the image acquisition. Thus, the final population consisted of 29 patients (23 men and six women) with an average (± SD) age of 60 ± 11 years and an average body mass index (kg/m2) of 32 ± 11 (Table 1). The prevalence of cardiac risk factors was very high (Table 1). Twenty-six patients (90%) had a history of angina, and 16 patients (55%) had experienced myocardial infarction before the examination. A significant portion of patients had been previously treated with percutaneous coronary intervention (n = 15; 52%) or coronary artery bypass grafting (n = 6; 21%) or both. The average heart rate at baseline was 68 ± 15 beats/min. Twenty-five of 29 patients (86%) had a cardiac catheterization performed within 1 month of the CT examination, which were available for comparison.
TABLE 1: Patient Demographic Characteristics
CharacteristicValue
No. of patients29
Age (y), mean ± SD60 ± 11
Female sex6 (21)
Body mass index (kg/m2), mean ± SD32 ± 11
Baseline heart rate (beats/min), mean ± SD68 ± 15
Cardiovascular risk factors 
 Diabetes9 (31)
 Smoking12 (41)
 Dyslipidemia25 (86)
 Hypertension25 (86)
 Family history of coronary artery disease18 (62)
Medical history 
 Angina26 (90)
 Myocardial infarction16 (55)
 Percutaneous coronary intervention15 (52)
 Coronary artery bypass grafting6 (21)

Note—Except where noted otherwise, data are number (%) of patients.

CT Image Acquisition

All patients were examined on a second-generation dual-source CT system (Definition Flash, Siemens Healthcare) in DECT mode. DECT studies were acquired during peak pharmacologic stress, at rest, and 6 minutes after the last contrast agent injection with retrospective ECG-gating, ECG-dependent tube current modulation, and the following scan parameters: 2 × 64 × 1.5-mm (stress, delayed) or 0.6-mm (rest) detector collimation with z-flying focal spot technique, 280-ms gantry rotation time, and heart rate adaptive pitch of 0.2–0.43. Automated tube current modulation (CareDose 4D, Siemens Healthcare) was used. One tube of the dual-source CT system was operated with 140 reference mAs per rotation at 140 kVp using an additional tin filter, and the second tube was operated with 165 reference mAs per rotation at 100 kVp. Data were acquired in a craniocaudal direction. A bolus-triggering technique was used with a threshold of 70 HU in the thoracic aorta and an additional delay of 2 seconds. Stress acquisition was performed after a 2- to 5-minute infusion of adenosine (140 μg/kg/min; Adenoscan, Astellas) using an automatic perfusion pump or a single injection of 0.4 mg of regadenoson (Lexiscan, Astellas) under physician supervision, using a second IV line placed in the contralateral antecubital vein. Stress acquisition was started when an adequate hemodynamic response to adenosine or regadenoson (heart rate increase > 10 beats/min or systolic blood pressure decrease > 10 mm Hg) was observed. A blood pressure cuff was placed on the arm of the contrast agent injection, to avoid interference with the administration of adenosine, in combination with a continuous ECG to monitor the patient heart rate and the response to adenosine or regadenoson administration. The attending physicians had aminophylline for adenosine receptor antagonism and nitroglycerine for persistent chest pain readily available, and a fully equipped resuscitation trolley with defibrillator was easily accessible.
For rest DECT examinations, patients with resting heart rates equal to or greater than 65 beats/min and those who did not readily return to their resting heart rate of less than 65 beats/min after completion of the stress portion received up to 15 mg of IV metoprolol tartrate (Lopressor, Novartis) before their rest examination.
Stress and rest examinations were enhanced with 80 mL of iopamidol (370 mg I/mL Isovue, Bracco) for each study followed by 50 mL of saline. Flow rate was 6 mL/s, whenever possible. Late-enhancement acquisition was performed 6 minutes after stress acquisition without application of additional contrast material.

CT Image Reconstruction

Merged DECT images were reconstructed with equal contributions from the 140-kVp dataset and the 100-kVp dataset at 1.5-mm reconstruction thickness, using a dedicated DECT kernel (D30f). Both systolic and diastolic series were reconstructed. Iodine distribution maps were generated using the “heart PBV” postprocessing algorithm of the DECT application on a workstation (Multimodality Workplace, version VE36A, Siemens Healthcare). The resulting color-coded iodine maps were then superimposed onto gray-scale multiplanar reformats of the virtual unenhanced datasets of the myocardium in short- and long-axis views of the left ventricle (Fig. 1).
Fig. 1A —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
A, Pharmacologic stress dual-energy myocardial iodine distribution maps in four-chamber (A), long-axis (B), and short-axis (C) views show homogeneous iodine distribution consistent with normal myocardial perfusion.
Fig. 1B —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
B, Pharmacologic stress dual-energy myocardial iodine distribution maps in four-chamber (A), long-axis (B), and short-axis (C) views show homogeneous iodine distribution consistent with normal myocardial perfusion.
Fig. 1C —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
C, Pharmacologic stress dual-energy myocardial iodine distribution maps in four-chamber (A), long-axis (B), and short-axis (C) views show homogeneous iodine distribution consistent with normal myocardial perfusion.
Fig. 1D —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
D, Stress SPECT images in four-chamber (D), long-axis (E), and short-axis (F) views also show normal myocardial perfusion.
Fig. 1E —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
E, Stress SPECT images in four-chamber (D), long-axis (E), and short-axis (F) views also show normal myocardial perfusion.
Fig. 1F —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
F, Stress SPECT images in four-chamber (D), long-axis (E), and short-axis (F) views also show normal myocardial perfusion.
Fig. 1G —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
G, Curved multiplanar reconstructions of coronary CT angiography show patent stents (arrowheads) in left anterior descending artery (G) and circumflex branch (H) and some eccentric calcifications without significant stenosis.
Fig. 1H —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
H, Curved multiplanar reconstructions of coronary CT angiography show patent stents (arrowheads) in left anterior descending artery (G) and circumflex branch (H) and some eccentric calcifications without significant stenosis.
Fig. 1I —62-year-old man with history of coronary revascularization and recent onset of chest discomfort.
I, Right coronary artery shows no evidence of coronary artery disease. Catheter angiography was not performed in this patient.

SPECT Examination

SPECT was performed after IV administration of tetrofosmin (99mTc). In summary, nuclear myocardial perfusion imaging was acquired with a single-day rest and stress protocol. Radionuclide activities of 370 MBq and 1110 MBq were administered IV at rest and stress, respectively. Ergometric stress testing was performed with the use of the Bruce treadmill protocol. Pharmacologic stress testing with adenosine or regadenoson was performed if there were contraindications for ergometric testing or in the event of an inadequate exercise stress test. A triple-head camera system (Vertex 601, collimator VXGP, Philips Healthcare) with attenuation correction was used for ECG-gated data acquisition. Emission data were reconstructed using a gaussian smoothing filter with scatter correction, followed by iterative reconstruction (FLASH3D, Siemens Healthcare). Short-axis, horizontal long-axis, and vertical long-axis images were automatically generated. In cases when the software failed to detect left ventricular contours, manual contours correction was performed by an experienced technologist. Also, if there was significant patient motion noted during review of raw projections, motion correction software was applied.

Analysis of SPECT Images

To establish a reference standard for hemodynamically significant stenosis, SPECT examinations were interpreted for perfusion defects by two experienced readers (one nuclear medicine physician and one cardiologist) in consensus. Images were analyzed on a dedicated console using commercially available software (Syngo VA60A, Siemens Healthcare). Evidence of ischemia was recorded by visual comparison of rest and stress SPECT perfusion scans. Perfusion defects were visually rated as reversible (only present at stress), fixed (present in both rest and stress images), and mixed (more pronounced at stress, partly reversible at rest). Readers were blinded to the results of DECT and catheter angiography examinations.

Analysis of Cardiac Catheterization Images

To establish a reference standard for morphologically significant coronary artery stenosis, angiograms obtained from these cardiac catheterizations were interpreted for the presence of significant coronary artery stenosis in consensus by two experienced interventional cardiologists. Catheter angiograms were viewed on a dedicated workstation (Vue Cardio PACS, Carestream Health). A significant coronary stenosis was defined as 50% or more luminal narrowing.

CT Image Analysis

Analysis of coronary CT angiography—Coronary CTA datasets were evaluated by two readers with extensive experience in cardiac CT for the presence of significant coronary artery stenosis defined as luminal narrowing greater than 50%. Clinical coronary CTA analysis of coronary artery morphology for stenosis detection and grading was performed visually using dedicated software (Syngo.via, Siemens Healthcare) on the basis of a combination of transverse sections and curved multiplanar reformats of the target vessels.
Visual analysis of dual-energy CT myocardial iodine maps—From each set of DECT raw data, one image series was reconstructed only on the basis of the 80-kV x-ray spectrum and another only on the basis of the 140-kV x-ray spectrum on the workstation. The myocardial iodine distribution was analyzed according to the unique x-ray absorption characteristics of this element at different kilovoltage levels. The resulting color-coded iodine maps were then superimposed onto gray-scale multiplanar reformats of the myocardium in short- and long-axis views of the left ventricle. The resulting images were displayed using the “Heart Perfused Blood Volume” application of the DECT image postprocessing software.
Two experienced readers in consensus analyzed all DECT studies for myocardial iodine defects using the American Heart Association 17 segmental model. Perfusion assessment was performed using both the stress and rest images. Readers were blinded to the results of SPECT and catheter angiography examinations.
Myocardial perfusion defects on DECT iodine maps were defined as circumscribed areas of decreased or absent iodine content within the left ventricular wall, relative to the remainder of the myocardium. Rest and stress iodine distribution maps were viewed side-by-side and were assessed for the presence of perfusion defects. If present, perfusion defects were visually rated as reversible (only present at stress), fixed (present in both rest and stress images), and mixed (more pronounced at stress, partly reversible at rest). The beam-hardening artifacts in the dual-energy iodine distribution maps in our study were overall mild and could readily be distinguished from true perfusion defects because of their typical location adjacent to dense contrast material, streaklike appearance, and lack of correlation with coronary territories. In addition, the delayed acquisition images for each patient were viewed in short- and long-axis views of the left ventricle and were assessed for the presence of late iodine enhancement of the left ventricular wall. Late enhancement was recorded as present or absent for each patient.

Estimation of Radiation Dose

The CT dose index and dose-length product were recorded from patient protocols. Effective radiation dose was calculated using a standard conversion factor of 0.0145 for adult chest CT [12].

Statistical Analysis

Catheter angiography was used as the reference standard for the presence of a morphologically significant coronary stenosis, which was defined as more than 50% luminal narrowing. SPECT examination was used as the reference standard for the detection of myocardial perfusion defects when determining hemodynamic significance of lesions. Against these reference standards, the diagnostic performance of coronary CTA alone, DECT myocardial perfusion assessment alone, and the combined assessment (both positive and either positive) were evaluated. For each comparison, we calculated per-patient sensitivity and specificity. ROC curve statistics were performed, and the AUC and the pertinent p value were calculated. A p value less than 0.05 was considered statistically significant. All statistical analyses were conducted using statistical software (SPSS Statistics version 21, IBM).

Results

Prevalence of Perfusion Defects on SPECT and Dual-Energy CT

In 21 of 29 patients (72%), a myocardial perfusion defect was identified on SPECT. Perfusion defects were classified as fixed in 14 patients, reversible in five patients, and mixed in two patients. On DECT, perfusion defects were detected in 22 patients. DECT classified 16 perfusion defects as fixed, three as reversible, and three as mixed. In 11 of the 16 patients with fixed perfusion defects, late iodine enhancement was detected at DECT.

Prevalence of Significant Coronary Artery Stenosis on Coronary CT Angiography and Catheterization

Of the 25 patients with cardiac catheterization, 19 were found to have at least one significant (> 50% diameter narrowing) stenosis at invasive angiography. Coronary CTA revealed at least one significant stenosis in 23 of 29 patients.

Diagnostic Value of Dual-Energy CT for the Detection of Morphologically Significant Stenosis

First, we assessed the diagnostic value of DECT for detecting morphologically significant stenosis, defined as greater than 50% luminal narrowing on catheterization (n = 25; Table 2). The presence of a significant stenosis on coronary CTA predicted the presence of a significant stenosis on catheterization with 95% sensitivity and 50% specificity. A perfusion defect on DECT predicted the presence of a significant stenosis on catheter with 95% sensitivity and 50% specificity. If either coronary CTA or DECT perfusion maps, or both, were positive (“either positive”), this predicted the presence of a significant stenosis on catheter with 100% sensitivity and 33% specificity. If both coronary CTA and DECT perfusion maps were positive (“both positive”), this predicted hemodynamically significant stenosis with 90% sensitivity and 67% specificity. The AUC was higher for “both positive” (0.78) than for coronary CTA alone (0.72), perfusion defects on DECT myocardial iodine maps alone (0.72), and “either positive” (0.67). On ROC analysis, the diagnostic value for the prediction of perfusion defects on SPECT was only significant for “both positive” (p = 0.04), but not for coronary CTA alone (p = 0.11), perfusion defects on DECT myocardial perfusion maps alone (p = 0.11), and “either positive” (p = 0.23).
TABLE 2: Per-Patient Test Characteristics of Dual-Energy CT (DECT) for Detection of Morphologically Significant Stenosis
Evaluation ProtocolSensitivity (%)Specificity (%)AUCp
Coronary CT angiography positive (at least one stenosis > 50%)95500.720.11
Perfusion defect on DECT myocardial iodine maps95500.720.11
Either positive100330.670.23
Both positive90670.780.04

Diagnostic Value of Dual-Energy CT for the Detection of Hemodynamically Significant Stenosis

Next, we assessed the diagnostic value of DECT for the detection of hemodynamically significant stenosis, defined as a perfusion defect on SPECT (n = 29; Table 3). The presence of significant stenosis on coronary CTA predicted the presence of a perfusion defect on SPECT with 91% sensitivity and 38% specificity. A perfusion defect on DECT predicted the presence of a perfusion defect on SPECT with 95% sensitivity and 75% specificity. If either coronary CTA or DECT perfusion maps or both were positive (“either positive”), this predicted the presence of a perfusion defect on SPECT with 100% sensitivity and 38% specificity. If both coronary CTA and DECT perfusion maps were positive (“both positive”) (Fig. 2), this predicted hemodynamically significant stenosis with 86% sensitivity and 75% specificity. The AUCs were significantly higher for perfusion defects on DECT myocardial iodine maps (0.85) and for “both positive” (0.80) than for coronary CTA alone (0.64) and “either positive” (0.69). On ROC analysis, the diagnostic value for the prediction of perfusion defects on SPECT was only significant for perfusion defects on DECT myocardial perfusion maps (p < 0.01) and “both positive” (p = 0.01), but not for coronary CTA alone (p = 0.25) or “either positive” (p = 0.12).
TABLE 3: Per-Patient Test Characteristics of Dual-Energy CT for Detection of Hemodynamically Significant Stenosis
Evaluation ProtocolSensitivity (%)Specificity (%)AUCp
Coronary CT angiography positive (at least one stenosis > 50%)91380.640.25
Perfusion defect on myocardial iodine maps95750.85< 0.01
Either positive100380.690.12
Both positive86750.800.01
Fig. 2A —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
A, Pharmacologic stress dual-energy CT myocardial iodine distribution maps in four-chamber (A), long-axis (B), and short-axis (C) views show circumscribed area of decreased iodine content (arrows) in apicoanterior left ventricular myocardium, consistent with perfusion defect.
Fig. 2B —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
B, Pharmacologic stress dual-energy CT myocardial iodine distribution maps in four-chamber (A), long-axis (B), and short-axis (C) views show circumscribed area of decreased iodine content (arrows) in apicoanterior left ventricular myocardium, consistent with perfusion defect.
Fig. 2C —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
C, Pharmacologic stress dual-energy CT myocardial iodine distribution maps in four-chamber (A), long-axis (B), and short-axis (C) views show circumscribed area of decreased iodine content (arrows) in apicoanterior left ventricular myocardium, consistent with perfusion defect.
Fig. 2D —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
D, Stress SPECT images in four-chamber (D), long-axis (E), and short-axis (F) views confirm area of ischemia (arrows) in apicoanterior wall of left ventricle.
Fig. 2E —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
E, Stress SPECT images in four-chamber (D), long-axis (E), and short-axis (F) views confirm area of ischemia (arrows) in apicoanterior wall of left ventricle.
Fig. 2F —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
F, Stress SPECT images in four-chamber (D), long-axis (E), and short-axis (F) views confirm area of ischemia (arrows) in apicoanterior wall of left ventricle.
Fig. 2G —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
G, Catheter angiography in right anterior oblique view (G), coronary CT angiography in curved multiplanar reconstruction (H), and volume-rendered image (I) show critical stenosis of proximal portion of left anterior descending artery (arrows), which was subsequently treated with stent.
Fig. 2H —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
H, Catheter angiography in right anterior oblique view (G), coronary CT angiography in curved multiplanar reconstruction (H), and volume-rendered image (I) show critical stenosis of proximal portion of left anterior descending artery (arrows), which was subsequently treated with stent.
Fig. 2I —68-year-old man with diabetes who experienced shortness of breath after mild exercise.
I, Catheter angiography in right anterior oblique view (G), coronary CT angiography in curved multiplanar reconstruction (H), and volume-rendered image (I) show critical stenosis of proximal portion of left anterior descending artery (arrows), which was subsequently treated with stent.

Effective Radiation Dose of Dual-Energy CT Examination

Mean CT dose indexes were 24, 29, and 20 mGy, and mean dose-length products were 398, 452, and 310 mGy × cm for rest, stress, and delayed DECT, respectively. The corresponding average effective radiation dose equivalent from rest, stress, and delayed DECT were 5.8, 6.6, and 4.5 mSv, respectively. The mean radiation dose from all three acquisitions was 16.9 mSv.

Discussion

In high-risk patients, blooming artifacts from severe coronary calcifications can impair the visualization of the coronary lumen, thus reducing the diagnostic accuracy of coronary CTA for significant stenosis. The resulting reduction in coronary CTA specificity frequently leads to further functional testing or to direct referral to coronary catheterization. In our study, we found a per-patient specificity of 50% and 38% for coronary CTA, compared with catheterization and SPECT, respectively. This is in line with a recent meta-analysis analyzing coronary CTA performance in patients with an elevated calcium score, which showed a significant decrease in specificity from 88.2% to 50.6% when the calcium score range moved from 101–400 to 401–1000 [13]. This result was expected, because it is generally accepted and also has been our experience that coronary CTA performs better in patients with a low-to-intermediate likelihood of having CAD than in high-risk patients. Functional imaging, on the other hand, is less likely to suffer in patients with significant CAD burden. Thus, SPECT is superior for assessing the hemodynamic relevance of coronary plaques in high-risk patients, overcoming this intrinsic limitation of anatomic diagnostic techniques.
In our study, adding myocardial perfusion assessment to coronary CTA significantly increased the specificity for coronary artery stenosis. Where significant stenosis on coronary CTA was associated with a perfusion defect on DECT myocardial perfusion maps, the specificity was 67% and 75% compared with catheterization and SPECT, respectively. On ROC analysis, the “both positive” approach had the highest discriminatory power for the detection of morphologically significant coronary artery stenosis using catheterization as the reference standard, outperforming both coronary CTA-only and DECT-only myocardial perfusion assessment. With SPECT as the reference standard for hemodynamically significant stenosis, both the “both positive” approach and the DECT myocardial perfusion maps alone had a similarly high discriminatory power. This reflects that coronary CTA, as an anatomic method, is more similar to coronary catheterization, and that DECT perfusion, as a functional test, shows a stronger correlation with SPECT.
In the first clinical study on adenosine-stress DECT myocardial perfusion imaging, Ruzsics et al. [3] reported a per-segment sensitivity of 92% and 93% specificity. Wang et al. [8] reported 68% sensitivity and 93% specificity. Using adenosine-stress MRI as the reference standard, Ko et al. [7] found 89% sensitivity and 76% specificity of rest and stress DECT for the detection of reversible perfusion defects. Our results are likely influenced by a high prevalence of CAD in the selected population; in fact, the majority of patients in our study had a known history of CAD or myocardial infarction or both. Also, we chose to perform a per-patient analysis rather than a per-segment analysis, because this more directly relates to clinical implications of the diagnostic test, whether or not a patient will be referred for catheter angiography or nuclear stress testing.
The true value of DECT myocardial perfusion imaging can only be appreciated by not considering it as a stand-alone test, such as SPECT, but as the functional complement to an anatomic evaluation of the coronary arteries, which should be applied in cases where coronary CTA is unable to discriminate flow-limiting stenosis from nonobstructive disease. A preliminary evaluation of the incremental value of DECT has recently analyzed the combination of rest-only DECT myocardial perfusion and coronary CTA in a low-risk population, showing a sensitivity of 90% and specificity of 86% for identifying 50% or greater coronary stenosis [8]. However, rest-only DECT myocardial perfusion fails to unveil the full potential of the technique, because reversible perfusion defects are detected only with the addition of a stress phase. Moreover, the true clinical value of the perfusion technique is likely to be limited in a population with a low prevalence of CAD. Thus, the value of DECT perfusion can be assessed only using the complete rest and stress protocol in a high-risk population with a considerable number of false-positive coronary CTAs.
A previous study has shown the superior cost-effectiveness of utilizing coronary CTA as a gatekeeper to catheter angiography in a population with a high prevalence of CAD compared with myocardial perfusion techniques [14]. It may be hypothesized that the combined anatomic and functional information offered by DECT could further increase the cost-effectiveness of the technique in high-risk populations.
In the first generation of dual-source CT scanners used for cardiac DECT, the dual-energy mode was limited by a decreased temporal resolution of 165 ms, instead of 83 ms, which could reduce the diagnostic performance in the detection of coronary stenosis. However, this historical disadvantage of cardiac DECT has been overcome in the second-generation dual-source CT, which offers the full temporal resolution of 75 ms also in dual-energy mode, thus not affecting the visualization of the coronary vasculature [15]. The rest DECT acquisition can be performed at the same radiation dose as a standard coronary CTA study [16]. For this reason, the rest DECT acquisition does not have disadvantages for the patient in comparison with a conventional coronary CTA study, while providing additional information on the myocardial blood pool distribution. The full cardiac DECT protocol with rest, stress, and delayed acquisition adds additional radiation exposure compared with a conventional coronary CTA. However, if used as an alternative to SPECT for the evaluation of myocardial perfusion, the comprehensive assessment of coronary anatomy and myocardial perfusion with DECT may actually decrease the total radiation exposure of the patient.
Cardiac DECT is a very recent technique, and the experience and evidence regarding the optimal interpretation of DECT myocardial iodine distribution maps are limited. The optimal diagnostic criteria of DECT myocardial iodine maps for the detection of fixed or reversible ischemia remain to be established. In particular, it has been reported that DECT rest acquisition alone can detect the presence of reversible perfusion defects [3, 6, 8], secondary to an intrinsic vasodilatory effect of the iodinated contrast medium [6]. If confirmed, this effect could cause a misclassification of perfusion defects by DECT compared with SPECT. Further investigations are needed to determine how this phenomenon is best understood and how it should be accounted for in the interpretation of cardiac DECT examinations.
The results of this study should be considered within the context of the study design and its limitations. First, the number of patients in this initial study was limited. In particular, the number of patients with a negative reference standard was very low in this high-risk population. Therefore, the specificity values calculated from our data are approximate estimations and should be confirmed in larger studies with representative populations. Also, catheter angiography was performed only if clinically indicated and was therefore only available for 25 patients. Two readers in consensus performed the visual analysis of DECT iodine distribution maps in our study. Therefore, we cannot assess the inter- and intrareader variability of this technique from our dataset.
In conclusion, the combined analysis of coronary CTA and DECT myocardial perfusion reduces the number of false-positives in a population at high-risk for CAD and outperforms the purely anatomic test of coronary CTA alone for the detection of morphologically and hemodynamically significant coronary artery stenoses.

Footnote

WEB
This is a web exclusive article.

References

1.
von Ballmoos MW, Haring B, Juillerat P, Alkadhi H. Meta-analysis: diagnostic performance of low-radiation-dose coronary computed tomography angiography. Ann Intern Med 2011; 154:413–420
2.
Vliegenthart R, Henzler T, Moscariello A, et al. CT of coronary heart disease. Part 1. CT of myocardial infarction, ischemia, and viability. AJR 2012; 198:531–547
3.
Ruzsics B, Schwarz F, Schoepf UJ, et al. Comparison of dual-energy computed tomography of the heart with single photon emission computed tomography for assessment of coronary artery stenosis and of the myocardial blood supply. Am J Cardiol 2009; 104:318–326
4.
So A, Hsieh J, Imai Y, et al. Prospectively ECGtriggered rapid kV-switching dual-energy CT for quantitative imaging of myocardial perfusion. JACC Cardiovasc Imaging 2012; 5:829–836
5.
Arnoldi E, Lee YS, Ruzsics B, et al. CT detection of myocardial blood volume deficits: dual-energy CT compared with single-energy CT spectra. J Cardiovasc Comput Tomogr 2011; 5:421–429
6.
Vliegenthart R, Pelgrim GJ, Ebersberger U, Rowe GW, Oudkerk M, Schoepf UJ. Dual-energy CT of the heart. AJR 2012; 199:S54–S63
7.
Ko SM, Choi JW, Song MG, et al. Myocardial perfusion imaging using adenosine-induced stress dual-energy computed tomography of the heart: comparison with cardiac magnetic resonance imaging and conventional coronary angiography. Eur Radiol 2011; 21:26–35
8.
Wang R, Yu W, Wang Y, et al. Incremental value of dual-energy CT to coronary CT angiography for the detection of significant coronary stenosis: comparison with quantitative coronary angiography and single photon emission computed tomography. Int J Cardiovasc Imaging 2011; 27:647–656
9.
Weininger M, Schoepf UJ, Ramachandra A, et al. Adenosine-stress dynamic real-time myocardial perfusion CT and adenosine-stress first-pass dual-energy myocardial perfusion CT for the assessment of acute chest pain: initial results. Eur J Radiol 2012; 81:3703–3710
10.
Yoon YE, Choi JH, Kim JH, et al. Noninvasive diagnosis of ischemia-causing coronary stenosis using CT angiography: diagnostic value of transluminal attenuation gradient and fractional flow reserve computed from coronary CT angiography compared to invasively measured fractional flow reserve. JACC Cardiovasc Imaging 2012; 5:1088–1096
11.
Taylor AJ, Cerqueira M, Hodgson JM, et al. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 appropriate use criteria for cardiac computed tomography. Circulation 2010; 122:e525–e555
12.
Deak PD, Smal Y, Kalender WA. Multisection CT protocols: sex- and age-specific conversion factors used to determine effective dose from dose-length product. Radiology 2010; 257:158–166
13.
den Dekker MA, de Smet K, de Bock GH, Tio RA, Oudkerk M, Vliegenthart R. Diagnostic performance of coronary CT angiography for stenosis detection according to calcium score: systematic review and meta-analysis. Eur Radiol 2012; 22:2688–2698
14.
Min JK, Gilmore A, Budoff MJ, Berman DS, O'Day K. Cost-effectiveness of coronary CT angiography versus myocardial perfusion SPECT for evaluation of patients with chest pain and no known coronary artery disease. Radiology 2010; 254:801–808
15.
Nance JW Jr, Bastarrika G, Kang DK, et al. High-temporal resolution dual-energy computed tomography of the heart using a novel hybrid image reconstruction algorithm: initial experience. J Comput Assist Tomogr 2011; 35:119–125
16.
Kerl JM, Bauer RW, Maurer TB, et al. Dose levels at coronary CT angiography: a comparison of dual energy-, dual source- and 16-slice CT. Eur Radiol 2011; 21:530–537

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: W70 - W77
PubMed: 24951230

History

Submitted: August 13, 2013
Accepted: October 17, 2013

Keywords

  1. coronary CT angiography
  2. dual-energy CT
  3. high-risk population
  4. myocardial perfusion imaging
  5. SPECT

Authors

Affiliations

Carlo Nicola De Cecco
Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401.
Departments of Radiological Sciences, Oncology, and Pathology, University of Rome “Sapienza”–Polo Pontino, Latina, Italy.
Brett S. Harris
Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401.
U. Joseph Schoepf
Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401.
Division of Cardiology, Department of Medicine, Medical University of South Carolina, Charleston, SC.
Justin R. Silverman
Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401.
Cullen B. McWhite
South Carolina Clinical and Translational Research Institute, Medical University of South Carolina, Charleston, SC.
Aleksander W. Krazinski
Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401.
Richard R. Bayer
Division of Cardiology, Department of Medicine, Medical University of South Carolina, Charleston, SC.
Felix G. Meinel
Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401.
Institute for Clinical Radiology, Ludwig-Maximilians-University Hospital, Munich, Germany.

Notes

Address correspondence to U. J. Schoepf ([email protected]).

Funding Information

This study was supported by a research grant provided by Bracco Diagnostics. U. J. Schoepf is a consultant for or receives research support from Bayer, Bracco, GE Healthcare, and Siemens Healthcare.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

View Options

View options

PDF

View PDF

PDF Download

Download PDF

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media