Eur Radiol DOI 10.1007/s00330-015-4161-x
COMPUTED TOMOGRAPHY
Coronary CT angiography in obese patients using 3rd generation dual-source CT: effect of body mass index on image quality Stefanie Mangold 1,2 & Julian L. Wichmann 1,3 & U. Joseph Schoepf 1,4 & Sheldon E. Litwin 4 & Christian Canstein 5 & Akos Varga-Szemes 1 & Giuseppe Muscogiuri 1,6 & Stephen R. Fuller 1 & Andrew C. Stubenrauch 1 & Konstantin Nikolaou 2 & Carlo N. De Cecco 1
Received: 4 May 2015 / Revised: 23 September 2015 / Accepted: 7 December 2015 # European Society of Radiology 2015
Abstract Objectives To evaluate the image quality of coronary CT angiography (CCTA) in obese patients using a 3rd generation, dual-source CT scanner. Methods We retrospectively evaluated 102 overweight and obese patients who had undergone CCTA. Studies were performed with 3rd generation dual-source CT, prospectively ECG-triggered acquisition at 120 kV, and automated tube current modulation. Advanced modeled iterative reconstruction was used. Patients were divided into three BMI groups: 1)25– 29.9 kg/m2; 2)30–39.9 kg/m2; 3) ≥ 40 kg/m2. Vascular attenuation in the coronary arteries was measured. Contrast-to-noise ratio (CNR) was calculated. Image quality was subjectively evaluated using five-point scales. Electronic supplementary material The online version of this article (doi:10.1007/s00330-015-4161-x) contains supplementary material, which is available to authorized users. * U. Joseph Schoepf
[email protected]
Results Image quality was considered diagnostic in 97.6 % of examinations. CNR was consistently adequate in all groups but decreased for groups 2 and 3 in comparison to group 1 as well as for group 3 compared to group 2 (p = 0.001, respectively). Subjective image quality was significantly higher in group 1 compared to group 3 (attenuation proximal: 4.8 ± 0.4 vs. 4.4 ± 0.6, p = 0.011; attenuation distal: 4.5 ± 0.7 vs. 4.0 ± 0.8, p = 0.019; noise: 4.7 ± 0.6 vs. 3.8 ± 0.7, p < 0.001). The mean effective dose was 9.5 ± 3.9 mSv for group 1, 11.4 ± 4.7 mSv for group 2 and 14.0 ± 6.4 mSv for group 3. Conclusion Diagnostic image quality can be routinely obtained at CCTA in obese patients with 3rd generation DSCT at 120 kV. Key Points • Diagnostic CCTA can be routinely performed in obese patients with 3rdgeneration DSCT. • 120-kV tube voltage allows diagnostic image quality in patients with BMI > 40 kg/m2. • 80-ml contrast medium can be administered without significant decline in vascular attenuation. Keywords Coronary computed tomography angiography . Body mass index . Dual-source . Tube current . Obesity
1
Division of Cardiovascular Imaging, Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, 25 Courtenay Drive, Charleston, SC 29425-2260, USA
2
Department of Diagnostic and Interventional Radiology, Eberhard-Karls University Tuebingen, Tuebingen, Germany
Introduction
3
Department of Diagnostic and Interventional Radiology, University Hospital Frankfurt, Frankfurt, Germany
4
Division of Cardiology, Department of Medicine, Medical University of South Carolina, Charleston, SC, USA
5
Siemens Medical Solutions, Malvern, PA, USA
6
Department of Radiological Sciences, Oncology and Pathology, University of Rome BSapienza^, Rome, Italy
Coronary computed tomography angiography (CCTA) is a well-established and robust modality for the noninvasive diagnostic work-up of suspected coronary artery disease (CAD). For multi-detector CT systems, sensitivities ranging from 73–100 % and specificities from 86–100 % for diagnosing obstructive CAD in non-obese patients have been reported [1–5]. However, obese patients remain a
Eur Radiol
challenge for CCTA with a reported potential for reduced diagnostic image quality or non-diagnostic studies. Only a limited number of prior investigations have evaluated the effect of body mass index (BMI) on CCTA [4, 6–8], and most of these included a limited number of patients [4, 6, 7, 9] or restricted inclusion criteria to patients with a BMI below 30 or 35 kg/m2 [6, 8, 10, 11]. Raff et al. reported significantly decreased diagnostic performance for CCTA in obese patients [4]. Similarly, Alkadhi et al. noted a decline in specificity and positive predictive value for CCTA in obese patients [12]. Thus, considerable interest remains in improving image quality and diagnostic performance of CCTA in obese patients, especially as this population is rapidly increasing throughout the world. With prior CT systems, one of the main solutions for improving the image quality in obese patients was increasing the tube voltage, e.g., to 140 kV [7, 13]. Although this improves photon flux, the use of higher tube voltage results in lower attenuation and, thus, lower signal of iodinated contrast media, and substantially increased radiation dose at CCTA. The recently introduced 3rd generation dual-source CT (DSCT) system provides technical developments that may be able to enhance image quality in obese patients without increasing the tube voltage, including advanced x-ray tubes with substantially increased power, higher spatial resolution due to a smaller focal spot, improved temporal resolution due to shorter gantry times and the use of dual x-ray sources and detectors, along with a fully integrated circuit detector system for noise reduction [14]. Furthermore, iterative reconstruction algorithms have been shown to substantially reduce image noise and thereby improve image quality [11, 15–20]. Taken together, wet hypothesize that these promising features may serve to provide adequate image quality of CCTA in morbidly obese patients compared to patients with normal body habitus while keeping radiation dose reasonable and thus enable the noninvasive assessment of coronary arteries in a broader patient population. The aim of this study was, therefore, to evaluate the objective (vessel attenuation and contrast) as well as subjective image quality of cardiac CTA at a standard tube voltage of 120 kV in overweight and obese patients with 3rd generation DSCT and to assess the effect of BMI on image quality.
Materials and methods Patient population This retrospective study was approved by the local Institutional Review Board with a waiver of informed consent.
We identified 110 overweight (BMI ≥ 25–29.9 kg/m2, n = 24) and obese patients (BMI ≥ 30 kg/m2, n = 78) who had undergone CCTA (n = 88) or acute chest pain Btriple-rule-out^ CT (TRO-CT, n = 22) at 120 kV between June 2014 and December 2014. Studies were excluded in the presence of severe metal or motion artefacts, examinations in which the patient could not elevate the arms over the head, and faulty electrocardiogram (ECG) synchronization.
CT acquisition parameters All examinations were performed with a 3rd generation DSCT system (SOMATOM Force, Siemens Healthcare, Forchheim, Germany) equipped with a fully integrated circuit detector system (Stellar Infinity, Siemens) and two x-ray tubes (Vectron, Siemens) with increased power (120 kW each) enabling tube currents up to 1300 mAs. A prospectively ECG-triggered adaptive sequential mode acquisition protocol was used and images were acquired at 20 % of the nominal tube current between 30–90 % of the RR interval and full nominal tube current at ~40 % or 70 % of the cardiac cycle depending on the heart rate. The scan length was defined for CCTA examinations as the distance between the carina as the upper border and the cardiac apex as the lower border of the scan range. TRO-CT examinations also included the aortic arch and the proximal portions of the supraaortic arteries. All studies were performed with automated attenuation-based anatomical tube current modulation (CARE Dose4D, Siemens) and tube voltage was selected by the use of an automated, attenuationbased tube voltage selection functionality (CarekV, Siemens). Further scan parameters were as follows: adaptive detector collimation varying from 96–192 in steps of 8 × 0.6 mm, gantry rotation 0.25 s, pitch 1, 512 × 512 pixel matrix size, with the field of view for reconstruction adapted to the size of each patient. Vessel attenuation was achieved with the use of intravenously administered iodinated contrast agent (350 mgI/ mL iohexol, Omnipaque, General Electric, Chalfont St. Giles, UK) with a flow rate of 4.5 ml/s (iodine delivery rate [IDR], 1.6 gI/s), and a contrast volume calculated as (4 s delay + scan time) x flow rate, followed by a 50-ml saline chaser injected at the same flow rate. Scan initiation was determined by bolus tracking using a dedicated software application (CARE Bolus, Siemens). A region of interest (ROI) was placed within the descending aorta at the level of the carina and the scan started automatically 4 seconds after a threshold of 100 Hounsfield units (HU) was reached in the ROI.
Eur Radiol
Image reconstruction All studies were reconstructed with dedicated 3rd generation advanced modeled iterative reconstruction (ADMIRE, Siemens) with strength level of three using a medium sharp convolution kernel (Bv36), 0.6-mm section thickness and increment of 0.4 mm. The reconstruction phase providing the best image quality while minimizing motion artefacts (BestPhase, Siemens) was chosen for the analysis. Subjective image quality analysis For objective and subjective image analysis, dedicated postprocessing and evaluation software (syngo.via VA30, Siemens) was used. Assessment of subjective image quality was performed independently by two readers with 6 (S.M.) and 4 years (J.L.W.) of experience in cardiovascular imaging. The reviewers were blinded to the clinical indication for imaging, to the patient characteristics including body weight, and the clinical imaging report. Study interpretation was performed individually and window settings were freely adjustable. The contrast attenuation in the proximal and distal major coronary artery segments (left main [LM], left anterior descending [LAD], circumflex [CX], and right coronary artery [RCA]) and the severity of image noise were graded by using a modified scoring system described by Yuan et al. [21]. In this scoring system, vessel attenuation was defined as 1 = poor opacification, non-diagnostic, 2 = suboptimal opacification, low diagnostic confidence, 3 = acceptable opacification of the major coronary arteries, sufficient for diagnosis, 4 = good opacification of proximal and distal segments, 5 = excellent opacification of proximal and distal segments. The accompanying image noise was also rated using a five-point scale defined as 1 = major noise, non-diagnostic, 2 = major noise, suboptimal evaluation with low confidence, 3 = moderate noise but sufficient for diagnosis, 4 = minor, diagnosis not influenced and 5 = none perceivable. Image quality was classified as diagnostic if both vessel opacification and noise were rated as ≥3. In case of disagreement regarding diagnostic image quality (i.e., a score of ≥3 by one reviewer and
