Cardiopulmonary Imaging Original Research

Transcription

1 Cardiopulmonary Imaging Original Research Bischoff et al. Versus Helical Mode for Coronary CTA Cardiopulmonary Imaging Original Research Bernhard Bischoff 1 Franziska Hein 1 Tanja Meyer 1 Markus Krebs 1 Martin Hadamitzky 1 Stefan Martinoff 2 Albert Schömig 1 Jörg Hausleiter 1 Bischoff B, Hein F, Meyer T, et al. Keywords: coronary artery disease, CT, prospective triggering, sequential scanning DOI: /AJR Received August 25, 2009; accepted after revision December 5, Klinik für Herz und Kreislauferkrankungen, Deutsches Herzzentrum Munchen, Lazarettstrasse 36, Munich, Bavaria 80636, Germany. Address correspondence to J. Hausleiter (hausleiter@dhm.mhn.de). 2 Institut für Radiologie und Nuklearmedizin, Deutsches Herzzentrum München, Klinik an der TU München, Munich, Germany. AJR 2010; 194: X/10/ American Roentgen Ray Society Comparison of and Helical Scanning for Radiation Dose and Image Quality: Results of the Prospective Multicenter Study on Radiation Dose Estimates of Cardiac CT Angiography (PROTECTION) I Study OBJECTIVE. Concerns have been raised about the radiation exposure of coronary CT angiography (CTA). Recently, a prospective ECG-triggered sequential coronary CTA technique was developed to reduce exposure to ionizing radiation. The purpose of this analysis was to determine the impact of a sequential scanning technique on image quality and radiation dose in a prespecified subgroup analysis of the Prospective Multicenter Study on Radiation Dose Estimates of Cardiac CT Angiography I (PROTECTION I) Study when compared with a standard helical scanning technique. MATERIALS AND METHODS. This analysis comprises MDCT coronary angiography studies of 47 international study sites in which the image quality was assessed by an experienced coronary CTA investigator using a 4-point score (1 = nondiagnostic, 4 = excellent image quality). Image quality was analyzed in all patients studied with the sequential scanning mode (n = 99) and in randomly selected patients of the population studied with the helical acquisition mode (n = 586). Radiation dose estimates were derived from the dose length product (DLP) and a conversion coefficient for the chest (0.014 msv mgy 1 cm 1 ). RESULTS. Although the sequential scanning mode significantly reduced radiation dose estimates by 68% from 11.2 msv for the helical mode to 3.6 msv for the sequential mode (p < 0.001), the median diagnostic image quality scores were comparable in both groups. The median diagnostic score for both scanning modes was 3.5 (interquartile range: sequential vs helical mode, vs , respectively; p = 0.62). CONCLUSION. The results of the PROTECTION I Study suggest that the prospective ECG-triggered sequential coronary CTA technique significantly reduces radiation dose without impairing image quality when compared with the standard retrospective helical data acquisition in patients with a low and stable heart rate. With improvements in the temporal and spatial resolution of CT, coronary CT angiography (CTA) has become a common diagnostic technique in clinical practice, particularly for the examination of patients with an intermediate pretest probability for obstructive coronary artery disease [1 3]. Nevertheless, there remains concern regarding the exposure to ionizing radiation and its potential hazards. The international Prospective Multicenter Study on Radiation Dose Estimates of Cardiac CT Angiography I (PROTECTION I) Study analyzed the extent of coronary CTA radiation dose estimates and the impact of different strategies to reduce dose in clinical practice [4]. The study showed that effective radiation exposure usually adds up to doses between 8 and 18 msv when performing 64-MDCT coro- nary angiography depending on the CT system, the scanning technique, and patientdependent factors [4]. For this reason, several dose-saving scanning techniques and algorithms have been developed. Usually, coronary CTA is performed using the retrospective ECG-gated helical scanning technique [1]. When compared with this helical image acquisition, the prospective ECG-triggered sequential scanning mode yields a considerable reduction of radiation dose. Originally, the prospective scanning algorithm was predominantly used for unenhanced calcium scoring. Recently, this scanning technique has been reintroduced into contrast-enhanced coronary CTA [5]. When applying the prospective ECG-triggered sequential scanning algorithm, image acquisition is performed only during a short predefined time period in the R-R interval, AJR:194, June

2 Bischoff et al. requiring a low and stable heart rate. Small single-center studies have indicated that in comparison with retrospective ECG-gated helical scanning, sequential scanning allows a significant reduction of the radiation dose without impairing image quality in adequately selected patients [6, 7]. The aim of this analysis was to compare the sequential scanning mode and the standard helical image acquisition in a large prospective multicenter and multivendor study with respect to image quality and radiation dose. Material and Methods Study Protocol The methods of the observational PROTEC- TION I Study have been described in detail elsewhere [4]. In brief, 50 international study sites provided image data and scanning protocols of 1,965 consecutive patients undergoing coronary CTA during 1 month. All data were collected and analyzed in a central coronary CTA core laboratory. All patients underwent coronary CTA for visualization of coronary arteries or bypass grafts, combined examination of the coronary and pulmonary arteries in patients with chest pain, or visualization of the cardiac anatomy before or after electrophysiologic procedures. The current analysis comprised only the examinations performed for visualization of the coronary arteries with a 64-MDCT system (n = 1,544 studies; sequential coronary CTA: n = 99; helical coronary CTA: n = 1,445). Of these patients, we included all patients scanned with a sequential scanning technique (n = 99, 100%) and a large group of randomly selected patients examined with helical CT (n = 586, 41%) in the current analysis. In these patients, image quality grading was performed. Because of the large number of patients, it was not practicable to obtain image quality in all patients examined with helical CT; therefore, a subgroup of these patients was randomly selected for image quality grading. The scanning protocol including the selection of the scanning mode was at the discretion of the performing physician. The study was approved by the ethics committee, and all patients gave written informed consent as required at the individual study sites. Image Quality To obtain objective indexes of image quality, the image noise, signal intensity, signal-to-noise ratio, and contrast-to-noise ratio were obtained for all coronary CTA examinations. The image noise was defined as the averaged SDs of the CT attenuation values (in Hounsfield units) inside two large regions of interest (ROIs) in the proximal segments of the left and right coronary arteries. The signal intensity was derived from the mean attenuation values measured in the same two ROIs. The signal-to-noise ratio was calculated as the mean CT attenuation values of the left and right coronary arteries divided by image noise. The contrast-to-noise ratio was defined as the difference between the mean CT attenuation values of the proximal coronary arteries and the mean density of the left ventricular wall, which was divided by image noise. An experienced coronary CTA core laboratory reader, who was blinded to coronary CTA acquisition details, determined image quality on a per-vessel basis. Image quality was determined by a score describing image quality of the four main coronary arteries left main, left anterior descending, left circumflex, and right coronary arteries based on a 4-point grading system. A score of 1 was defined as nondiagnostic and meant that impaired image quality precluded appropriate evaluation of the coronary arteries due to severe motion artifacts, extensive coronary calcifications, severe image noise, or insufficient contrast. A score of 2, defined as adequate, indicated that image quality was reduced because of artifacts due to motion, image noise, or low contrast attenuation but that image quality was still sufficient to rule out significant stenosis. A score TABLE 1: Patient and Scanning Characteristics Patient characteristics Characteristics of 3 was defined as good and was assigned when artifacts, caused by motion, image noise, coronary calcifications, or low contrast, did not interfere with assessment concerning the presence of luminal stenosis and the presence of calcified and noncalcified coronary atherosclerotic plaque. A score of 4, defined as excellent, indicated a complete absence of motion artifacts, strong attenuation of vessel lumen, and clear delineation of vessel walls and ability to assess luminal stenosis as well as plaque characteristics. All coronary CTA examinations with at least one coronary artery with nondiagnostic image quality (image quality score = 1) were rated nondiagnostic. Estimation of Radiation Dose The collected parameters relevant to radiation dose included the volume CT dose index (CTDI vol ) and dose length product (DLP), both of which were obtained from the CT scanning protocol of each coronary CTA study. The DLP was the primary study outcome parameter. For estimation of the effective dose, the product of the DLP and an organ weighting factor for the chest as the investigated anatomic region (k = msv (mgy cm) 1 averaged between male and female models) was calculated as proposed by the European Helical Mode No. of patients Mode Height, m 1.70 ( ) 1.72 ( ) Weight, kg 76.0 ( ) 75.0 ( ) Body mass index, kg/m ( ) 25.8 ( ) Heart rate, bpm 61 (55 69) 56 (52 61) < Sinus rhythm, no. (%) 568 (96.9) 98 (99.0) Scanning characteristics Scan length, mm 129 ( ) 139 ( ) < Tube voltage, kv 120 ( ) 120 ( ) 0.08 CT scanner < GE Healthcare 64-MDCT, a no. (%) 96 (16.4) 87 (87.9) Philips Healthcare 64-MDCT, b no. (%) 28 (4.8) 11 (11.1) Siemens Healthcare single-source 64-MDCT, c no. (%) 96 (16.4) 0 (0) Siemens Healthcare dual-source 64-MDCT, d no. (%) 334 (57.0) 1 (1.0) Toshiba 64-MDCT, e no. (%) 32 (5.5) 0 (0) Note Unless noted otherwise, data are presented as median (interquartile range). a LightSpeed VCT. b Brilliance 64. c Somatom Sensation. d Somatom Definition. e Aquilon 64. p 1496 AJR:194, June 2010

3 Versus Helical Mode for Coronary CTA Working Group for Guidelines on Quality Criteria in CT [8]. Statistical Analysis For selection of patients for the helical CT group, a random selection was performed using SPSS software (version , SPSS). To allow a sufficient size of the control group, approximately 40% of all helical coronary CTAs were selected, resulting in a control group of 40.5% of all patients examined with helical CT. All parameters of the sequential scanning protocol were compared with the standard helical image acquisition. Continuous variables were expressed as medians (interquartile range [IQR]) and were compared using the Wilcoxon s rank sum test. Categoric variables were expressed as frequencies or percentages. The chi-square test was used to test differences in the frequency of categoric variables. A p value of < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (version ). Results A total of 685 patients undergoing 64- MDCT coronary angiography at 47 study sites were included in this analysis. Prospective ECG-triggered sequential image acquisition was performed in 99 patients, whereas 586 patients were examined using retrospective ECG-gated helical scanning. image acquisition was used in five of the 47 study sites. Table 1 summarizes patient and scanning characteristics, respectively, of both groups. The frequency of sinus rhythm, height, weight, and body mass index were comparable in both groups. Heart rate was significantly lower in patients scanned with the sequential mode than those examined with the standard helical mode (56 vs 61 beats per minute [bpm], respectively; p < 0.001). Furthermore, there was a significant difference regarding scanning length between both groups (139 and 129 mm for sequential and helical scanning mode, respectively; p < 0.001). When retrospective ECG-gated image acquisition was used, the median CTDI vol added up to 49.8 mgy (IQR, mgy). In contrast, in patients examined with the sequential scanning technique, the median CT- DI vol was significantly reduced to 18.4 mgy (IQR, mgy) (p < 0.001). Furthermore, sequential image acquisition led to a significantly reduced DLP when compared with standard helical scanning (median DLP [IQR], 259 mgy cm [ mgy cm] vs 801 mgy cm [563 1,091 mgy cm], respectively; p < 0.001), resulting in a 68% TABLE 2: Quantitative Image Quality Parameters Image Quality Parameters reduction of estimated radiation dose. This difference translates to a median estimated radiation dose of 3.6 msv (IQR, msv) versus 11.2 msv (IQR, msv) for sequential versus helical scanning technique, respectively. Table 2 displays the quantitative image quality data including image noise, signalto-noise ratio, and contrast-to-noise ratio derived on scanning mode. With sequential scanning, image noise and signal intensity significantly increased by 9% and 6%, respectively. Consequently, the derived signaland contrast-to-noise ratios did not differ significantly between both groups. Despite the reduction in radiation dose, there was no significant difference between groups regarding the image quality score (median score [IQR] for sequential vs helical mode, 3.50 [ ] vs 3.50 [ ], respectively; p = 0.622). Figure 1 displays the 68% reduction of radiation dose and comparable image quality in both groups. Whereas the image quality achieved using sequential coronary CTA was rated diagnostic in 93% Median Value (Interquartile Range) Helical Mode Mode Image noise (HU) 20 (17 24) 22 (19 27) Signal intensity (HU) 365 ( ) 388 ( ) Signal-to-noise ratio 17.7 ( ) 18.2 ( ) Contrast-to-noise ratio 13.0 ( ) 13.3 ( ) Dose Length Product (mgy cm) 2,000 1,000 0 Helical p < A Image Quality Score (Grade) Helical p < of the patients, only 89% of the patients examined with conventional helical CT had diagnostic examinations (p = 0.245) Figure 2 displays the image quality of two coronary CTA examinations acquired with either the sequential or helical scanning mode. Discussion In recent years, coronary CTA has evolved as a useful noninvasive imaging technique with a very high diagnostic accuracy for the detection of obstructive coronary artery disease [9 11]. In addition, coronary CTA has been shown to have a prognostic impact in the evaluation of patients with chest pain symptoms [12, 13]. However, the exposure to ionizing radiation associated with coronary CTA has raised concerns. Consequently, several radiation dose saving techniques have been developed for coronary CTA to obtain diagnostic coronary CTA images with the lowest possible radiation dose. The international PROTECTION I Study, which is a prospective multicenter survey study that focuses on radiation dose of coronary CTA p Fig. 1 Box-and-whisker plots. A and B, Box-and-whisker plots show dose length product (A) and image quality score (B) for coronary CT angiography examinations using either helical or sequential scanning mode. Horizontal line in each box shows median and top and bottom lines of boxes show interquartile range (IQR). Whiskers show lowest value still within 1.5 IQR of lower or upper quartile. B AJR:194, June

4 Bischoff et al. Fig year-old man who underwent coronary CT angiography for visualization of coronary arrteries (A C) and 67-year-old man who underwent coronary CT angiography for visualization of coronary arteries (D F). A F, Curved-planar maximum-intensity-projection CT angiograms of left descending artery (A and D), left circumflex artery (B and E), and right coronary artery (C and F) obtained using sequential scanning mode (A C) and helical scanning mode (D F) are shown for comparison with regard to image quality. in daily practice, revealed a median radiation dose of 12 msv (IQR, 8 18 msv) for coronary CTA [4]. Furthermore, in this study several independent predictors for the extent of coronary CTA radiation exposure have been identified. Among these factors, the prospective ECG-triggered sequential scanning protocol was a very strong independent predictor of a reduced exposure to ionizing radiation. However, the impact of the reduced dose on image quality and the level of diagnostic confidence are unknown. Therefore, the current study investigates the impact of a sequential image acquisition in coronary CTA on image quality and radiation dose in more detail in a prespecified subgroup analysis of the PROTECTION I Study. The current analysis shows a 68% reduction in coronary CTA radiation dose for the prospective ECG-triggered sequential scanning technique in patients with a low and stable heart rate when compared with the standard helical scanning technique. Taking the significantly longer scanning length in patients undergoing sequential scanning (139 and 129 mm for sequential and helical scanning mode, respectively) into account, the effect of applying the prospective ECG-triggered sequential scanning mode on radiation dose would be even higher. The differences in scanning length between the helical and sequential scanning mode may be explained by differences in scan range definition. Whereas scan range can be defined accurately to a millimeter in conventional helical CT, it must be a multiple of scan coverage less the overlapping zones in sequential scans. Therefore, in some cases the scan range must be larger in sequential scans than one would have defined it in conventional helical CT. Despite this tremendous reduction in radiation dose, image quality was comparable in both groups. Accordingly, the image quality score did not differ significantly between both scanning modes. In this analysis, the prospective ECG-triggered sequential scanning technique even showed a trend to a higher rate of diagnostic coronary CTA examinations than the conventional helical coronary CTA technique (93% vs 89%, respectively; p = 0.245). However, this analysis due to the study design provides no data about how many repeated scans were obtained when image quality was nondiagnostic. The relatively high rate of nondiagnostic image quality may be explained by the unselected study collective of the PROTECTION I Study and the strict image quality rating in the central core laboratory. In concordance with the results of the current analysis, Hein et al. [14] recently reported a 63% reduction of radiation dose for sequential image acquisition in comparison with standard helical image acquisition in patients examined using a dual-source CT system. Similar to our analysis, image quality did not differ significantly between both study groups. A radiation dose reduction of even 77% without deterioration of image quality was reported by Shuman et al. [7] who retrospectively analyzed coronary CTA studies of 100 matched patients either scanned with helical (n = 50) or sequential (n = 50) mode. Hirai et al. [6] performed both helical and sequential image acquisitions in 60 patients with a heart rate below 75 bpm. When compared with the helical scanning mode, the prospective ECG-triggered image acquisition led to a 79% reduction of calculated effective dose while maintaining image quality and the ability to assess for luminal stenosis. Despite its enormous potential for radiation dose reduction and high image quality, the prospective ECG-triggered sequential scanning mode has limitations. Using the sequential scanning mode, image acquisition usually is performed during a short predefined period during mid-to-late diastole, allowing only a small reconstruction window. In patients with higher heart rates and thus increased coronary artery motion, coronary CTA acquired during diastole often suffers from motion artifacts. Therefore, the prospective ECG-triggered sequential scanning mode is recommended for only patients with lower heart rates. In contrast, standard helical CT with ECG-dependent tube current modulation and with a wide time window of the full tube current also allows image reconstruction in late systole, where coronary artery motion in patients with high heart rates is least and thus less motion artifacts occur [15, 16]. In patients with lower heart rates, prospective ECG-triggered sequential scanning technique should be used more often. In the PRO- TECTION I Study, only 5% of all patients were examined using the sequential scanning technique. Applying a threshold of 63 bpm as suggested by Husmann et al. [17] below which sequential image acquisition can be performed with good image quality, 56% of the PROTECTION I Study population would have qualified for the sequential scanning 1498 AJR:194, June 2010

5 Versus Helical Mode for Coronary CTA technique. This points out the enormous potential of this dose-saving algorithm. There are two main reasons for the low rate of sequential scanning in the PROTECTION I Study: First, many CT systems used in the study did not allow sequential scanning during the study period. Second, many examiners may not have been familiar with the sequential scan algorithm and may have feared a potential loss of diagnostic image information. The sequential scanning technique can be combined with other dose-saving strategies. For example, an approach could be the combination of prospective ECG-triggered sequential image acquisition and a reduced tube voltage of 100 kv in nonobese patients with a slow and stable sinus rhythm. By combining these radiation dose reduction techniques, an estimated radiation dose of even below 2 msv can be obtained [18]. Limitations The PROTECTION I Study was an observational study and therefore the scanning protocol including the selection of the scanning mode was at the discretion of the performing physician. This nonrandomized study design creates a potential selection bias. The radiation dose associated with coronary calcium scoring, which is performed in some institutions before coronary CTA, was not obtained in the current study. Therefore, in this analysis only radiation dose associated with coronary CTA was taken into account. In the PROTECTION I Study, only patients with low heart rates were examined using the sequential scanning mode. Therefore, the diagnostic performance of ECGtriggered sequential image acquisition could not be evaluated in patients with higher heart rates. The mean heart rate in the group who underwent helical coronary CTA was significantly higher than that of patients examined with the sequential scanning technique. This difference in heart rate may have caused an image quality bias in favor of the sequential coronary CTA group. Besides heart rate, heart rate variability significantly influences image quality of the sequential scanning mode. Unfortunately, heart rate variability was not recorded in the PROTECTION I Study, so the effect of heart rate variability on image quality could not be assessed in the current analysis. Furthermore, invasive angiography results were not obtained routinely in all patients; for this reason, the diagnostic accuracy of the sequential scanning mode could not be investigated. Summary The results of this study show that, compared with the standard helical image acquisition, the prospective ECG-triggered sequential coronary CTA scanning technique significantly reduces radiation dose without impairing image quality in appropriately selected patients. References 1. Budoff MJ, Achenbach S, Blumenthal RS, et al. Assessment of coronary artery disease by cardiac computed tomography: a scientific statement from the American Heart Association Committee on Cardiovascular Imaging and Intervention, Council on Cardiovascular Radiology and Intervention, and Committee on Cardiac Imaging, Council on Clinical Cardiology. Circulation 2006; 114: Fox K, Garcia MA, Ardissino D, et al.; The Task Force on the Management of Stable Angina Pectoris of the European Society of Cardiology. Guidelines on the management of stable angina pectoris: executive summary the Task Force on the Management of Stable Angina Pectoris of the European Society of Cardiology. Eur Heart J 2006; 27: Hendel RC, Patel MR, Kramer CM, et al.; American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group; American College of Radiology; Society of Cardiovascular Computed Tomography; Society for Cardiovascular Magnetic Resonance; American Society of Nuclear Cardiology; North American Society for Cardiac Imaging; Society for Cardiovascular Angiography and Interventions; Society of Interventional Radiology. ACCF/ACR/SCCT/SCMR/ASNC/ NASCI/SCAI/SIR 2006 appropriateness criteria for cardiac computed tomography and cardiac magnetic resonance imaging: a report of the American College of Cardiology Foundation Quality Strategic Directions Committee Appropriateness Criteria Working Group, American College of Radiology, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, American Society of Nuclear Cardiology, North American Society for Cardiac Imaging, Society for Cardiovascular Angiography and Interventions, and Society of Interventional Radiology. J Am Coll Cardiol 2006; 48: Hausleiter J, Meyer T, Hermann F, et al. Estimated radiation dose associated with cardiac CT angiography. JAMA 2009; 301: Hsieh J, Londt J, Vass M, Li J, Tang X, Okerlund D. Step-and-shoot data acquisition and reconstruction for cardiac x-ray computed tomography. Med Phys 2006; 33: Hirai N, Horiguchi J, Fujioka C, et al. Prospective versus retrospective ECG-gated 64-detector coronary CT angiography: assessment of image quality, stenosis, and radiation dose. Radiology 2008; 248: Shuman WP, Branch KR, May JM, et al. Prospective versus retrospective ECG gating for 64-detector CT of the coronary arteries: comparison of image quality and patient radiation dose. Radiology 2008; 248: Menzel H, Schibilla H, Teunen D. European guidelines for quality criteria for computed tomography. European Commission, 2000: publication no. EUR EN 9. Dewey M, Rutsch W, Schnapauff D, Teige F, Hamm B. Coronary artery stenosis quantification using multislice computed tomography. Invest Radiol 2007; 42: Leber AW, Knez A, von Ziegler F, et al. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol 2005; 46: Raff GL, Gallagher MJ, O Neill WW, Goldstein JA. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 2005; 46: Hadamitzky M, Freißmuth B, Meyer T, et al. Prognostic value of coronary computed tomographic angiography for prediction of cardiac events in patients with suspected coronary artery disease. JACC Cardiovasc Imaging 2009; 2: Min JK, Shaw LJ, Devereux RB, et al. Prognostic value of multidetector coronary computed tomographic angiography for prediction of all-cause mortality. J Am Coll Cardiol 2007; 50: Hein F, Meyer, T, Hadamitzky M, et al. Prospective ECG-triggered sequential scan protocol for coronary dual-source CT angiography: initial experience. Int J Cardiovasc Imaging (in press) 15. Herzog C, Arning-Erb M, Zangos S, et al. Multidetector row CT coronary angiography: influence of reconstruction technique and heart rate on image quality. Radiology 2006; 238: Seifarth H, Wienbeck S, Püsken M, et al. Optimal systolic and diastolic reconstruction windows for coronary CT angiography using dual-source CT. AJR 2007; 189: Husmann L, Valenta I, Gaemperli O, et al. Feasibility of low-dose coronary CT angiography: first experience with prospective ECG-gating. Eur Heart J 2008; 29: Scheffel H, Alkadhi H, Leschka S, et al. Lowdose CT coronary angiography in the step-andshoot mode: diagnostic performance. Heart 2008; 94: AJR:194, June