Combined Cardiac CT and MRI for the Comprehensive Workup of Hemodynamically Relevant Coronary Stenoses

Transcription

1 Cardiopulmonary Imaging Original Research Cardiopulmonary Imaging Original Research Olivio F. Donati 1 Hans Scheffel 1 Paul Stolzmann 1 Stephan Baumüller 1 André Plass 2 Sebastian Leschka 1 Hatem Alkadhi 1,3 Donati OF, Scheffel H, Stolzmann P, et al. Keywords: catheter coronary angiography, coronary angiography, CT, MRI, myocardial perfusion DOI: /AJR Received June 24, 2009; accepted after revision September 22, Supported by the National Center of Competence in Research, Computer-Aided and Imaging-Guided Medical Interventions of the Swiss National Science Foundation. 1 Institute of Diagnostic Radiology, University Hospital Zurich, Zurich, Switzerland. 2 Clinic for Cardiovascular Surgery, University Hospital Zurich, Zurich, Switzerland. 3 Present address: Cardiac MR PET CT Program, Massachusetts General Hospital and Harvard Medical School, 100 Charles River Plaza, Boston, MA Address correspondence to H. Alkadhi (halkadhi@partners.org). AJR 2010; 194: X/10/ American Roentgen Ray Society Combined Cardiac CT and MRI for the Comprehensive Workup of Hemodynamically Relevant Coronary Stenoses OBJECTIVE. The purpose of our study was to prospectively evaluate the accuracy of a comprehensive assessment of coronary artery disease (CAD) with prospectively ECG-gated coronary CT angiography (CTA) and perfusion cardiac MRI for the detection of hemodynamically relevant coronary stenoses. SUBJECTS AND METHODS. Forty-seven consecutive patients underwent k-space and time broad-use linear acquisition speed-up technique accelerated perfusion cardiac MRI at 1.5 T and dual-source. Catheter coronary angiography (CA),, and perfusion cardiac MRI were all performed within a median time interval of 7.5 days. Detection of hemodynamically relevant stenoses by the combination of plus perfusion cardiac MRI was compared with the combination of CA plus perfusion cardiac MRI, the latter serving as the standard of reference. RESULTS. CA identified stenoses in 75 of 141 coronary arteries (53.2%) in 33 of 47 patients (70.2%). Cardiac MRI revealed perfusion defects in 30 of 47 patients (63.8%). Image quality of was diagnostic in 635 of 638 segments (99.5%). Coronary CTA revealed stenoses greater than 50% in 76 of 141 coronary arteries (53.9%) of 33 of 47 patients (70.2%). Sensitivity, specificity, negative and positive predictive value, and accuracy of and perfusion cardiac MRI versus CA and perfusion cardiac MRI for the detection of hemodynamically relevant stenoses were 96.7%, 100%, 94.4%, 100%, and 97.9%, respectively. CONCLUSION. The combination of and perfusion cardiac MRI shows diagnostic performance comparable to that of CA and perfusion cardiac MRI. Preliminary data suggest that may replace CA in the diagnosis of hemodynamically relevant CAD. A comprehensive assessment of coronary artery disease (CAD) should include both information on coronary artery morphology and myocardial function, the latter indicating the hemodynamic relevance of a coronary artery lesion [1 3]. In clinical practice, coronary artery morphology with identification of the target coronary lesion for revascularization procedures usually is performed by conventional coronary angiography (CA). Assessment of myocardial function (i.e., perfusion and viability imaging) commonly is performed using either nuclear tests, stress echocardiography, or cardiac MRI, all showing similar performance characteristics [4 6]. Recently, coronary CT angiography (CTA) has been introduced as a noninvasive tool allowing the visualization of coronary artery morphology, challenging the reference stan- dard technique of CA [7, 8]. Furthermore, a dose-reducing technique for (i.e., prospective ECG-gating) has been developed that is characterized by high accuracy for diagnosing coronary stenoses with a very low radiation dose [9 13]. Nonetheless, to our knowledge, no study to date has evaluated whether prospectively ECG-gated could replace CA for the comprehensive imaging workup of CAD with regard to the detection of the hemodynamic relevance of coronary stenoses. The purpose of this study was to prospectively evaluate a combined assessment of CAD with prospectively ECG-gated coronary CTA plus perfusion cardiac MRI for the detection of hemodynamically relevant coronary stenoses in comparison with CA plus perfusion cardiac MRI, the latter combination serving as the reference standard. 920 AJR:194, April 2010

2 Subjects and Methods Study Population We prospectively screened 65 consecutive patients with known or suspected CAD who underwent elective CA. Data for 41 of the patients in this study are from our earlier study [14]. The clinical decision to perform CA was based on the history or symptoms of the patient or on the results from exercise stress testing. Patients were excluded from this study if they had a previous coronary artery bypass graft (n = 5), impaired renal function (n = 1), a heart rate greater than 70 beats per minute (which is considered to be a contraindication for prospectively ECG-gated coronary CTA [11] (n = 4), contraindications for adenosine (second or third atrioventricular block, sick sinus syndrome, symptomatic bradycardia, severe asthma, or obstructive pulmonary disease) (n = 2) or contraindications to MRI (implanted electronic devices, metallic foreign bodies in the eye, severe claustrophobia, and others according to local regulations and manufacturer s recommendations (n = 6). Finally, a total of 47 patients (38 men and nine women; mean age, 64 ± 9 years) could be included in this study (Fig. 1). Patient characteristics are summarized in Table 1. CA,, and perfusion cardiac MRI examinations were all performed within a median time interval of 8 days. The study protocol was approved by the local institutional review board and all patients gave written informed consent before enrollment. Prospectively ECG-Gated All CT examinations were performed on a dualsource CT system (Somatom Definition, Siemens CAD at CA (n [patients] = 30) (n [arteries] = 50) No CAD at CA (n [patients] = 0) (n [arteries] = 16) CAD at cardiac MRI (n [patients] = 30) (n [arteries] = 66) Healthcare) using prospective ECG-gating. All patients received a single 2.5-mg dose of sublingual isosorbide dinitrate (Isoket, Schwarz Pharma). No additional β-blockers were given before CT. Depending on body weight, ml of contrast medium (iopromide, Ultravist 370, Bayer Schering Pharma) was administered at a flow rate of 5 6 ml/s, followed by 50 ml of a 20% contrast agent 80% saline solution mixture. The contrast agent was administered using a dualhead power injector (Stellant, Medrad) and was controlled by bolus-tracking using a region of interest in the ascending aorta (attenuation threshold, 120 HU). Data were acquired in the craniocaudal direction during mid inspiration using the following parameters: detector collimation, mm; slice acquisition, mm by means of a z-flying spot; and gantry rotation time, 0.33 second. Attenuation-based tube current modulation was used with a reference tube current time product set at 190 mas per rotation. The data acquisition window was set at 70% of the R-R interval; the temporal resolution was 83 milliseconds. Patients with a body mass index (BMI) 25 kg/m 2 were examined with a tube voltage of 120 kv; patients with a BMI < 25 kg/m 2, with 100 kv. Coronary CTA images were reconstructed in a monosegment mode with a slice thickness of 0.6 mm using a medium smooth-tissue convolution kernel (B30f). If the vessel segment was calcified, additional reconstructions were performed using a sharp-tissue convolution kernel (B45f) to compensate for blooming artifacts. All images were anonymized and transferred to Patients excluded because of previous coronary artery bypass graft (n = 5) CA, low-dose and cardiac MRI (n = 47) CAD at low-dose (n [patients] = 29) (n [arteries] = 51) No CAD at low-dose (n [patients] = 1) (n [arteries] = 15) Eligible patients (n = 65) CAD at low-dose (n [patients] = 4) (n [arteries] = 25) No CAD at low-dose (n [patients] = 13) (n [arteries] = 50) Patients excluded from low-dose : Heart rate > 70 bpm (n = 4) Nephropathy (n = 1) Patients excluded from cardiac MRI: Technical/logistic (n = 2) Claustrophobia (n = 1) Body weight exceeding table limit (n = 1) Contraindications to adenosine (n = 2) Implanted electronic device (n = 2) an external workstation (Multi-Modality Workplace, Siemens Healthcare) for analysis. Catheter Coronary Angiography Biplane conventional CA was performed according to standard techniques. The angiograms were evaluated by an experienced observer who was blinded to the results from and perfusion cardiac MRI. The coronary arteries were subdivided according to the same scheme used for [15] and were quantitatively assessed with the use of an automated edge-detection system (Xcelera 1.2, Philips Healthcare). Vessel diameter measurements were performed on two different image planes and included the diameter of the reference vessel (proximal and distal to the stenosis), the minimal luminal diameter, and the extent of stenosis (defined as the diameter of the reference vessel minus the minimal luminal diameter, divided by the reference diameter, and multiplied by 100). A significant stenosis was defined as a reduction in luminal diameter of greater than 50%. Perfusion Cardiac MRI All MR studies were performed on a 1.5-T clinical MR system (Achieva, Philips Healthcare). Dedicated cardiac phased-array receiver coils were used for signal reception (five elements). All data were acquired during breath-hold in endinspiration. The true short axis of the left ventricle was determined from a series of scout images. Three representative short-axis sections were obtained, one each in the basal, midventricular, and No CAD at cardiac MRI (n [patients] = 17) (n [arteries] = 75) CAD at CA (n [patients] = 3) (n [arteries] = 25) No CAD at CA (n [patients] = 14) (n [arteries] = 50) CAD at cardiac MRI and CA (n [patients] = 30) (n [arteries] = 54) CAD at cardiac MRI and low-dose (n [patients] = 29) (n [arteries] = 53) No CAD at cardiac MRI and low-dose (n [patients] = 17) (n [arteries] = 87) No CAD at cardiac MRI and CA (n [patients] = 18) (n [arteries] = 88) Fig. 1 Graphic shows flowchart of study. CTA = CTA angiography, bpm = beats per minute, CA = coronary angiography, CAD = coronary artery disease. AJR:194, April

3 TABLE 1: Patient Characteristics Parameter No. of Patients (n = 47) Age (y) 64 ± 9 Women 9 (19) Men 38 (81) Body mass index (kg/m 2 ) 28 ± 4 Suspected CAD 16 (34) Known CAD 31 (66) Single-vessel 2 (4) Two-vessel 8 (17) Three-vessel 21 (45) Previous PCI or stenting 10 (21) Symptoms, Angina pectoris 25 (53) Atypical chest pain 6 (13) Dyspnea 25 (53) None 12 (25.5) Cardiovascular risk factors Diabetes mellitus 9 (19) Hypertension 35 (759) Dyslipidemia 34 (72) Smoker 14 (30) Note Except where indicated otherwise, data are number with percentage in parentheses. CAD = coronary artery disease, PCI = percutaneous coronary intervention. apical regions of the left ventricle according to the standardized 17-segment model of the American Heart Association [16]. Pharmacologic stress was applied using adenosine, which was administered IV at 140 μg per kilogram of body weight over 3 minutes under ECG, oxygen-saturation, and blood pressure monitoring. Acquisition of perfusion cardiac MR images was started immediately after the injection of gadobutrol (Gadovist 1.0, Bayer Schering Pharma). The contrast agent was dosed at 0.1 mmol per kilogram of body weight using a power injector (MR Spectris, Medrad) at an injection rate of 5 ml/s, followed by a 40-mL saline flush. Ten minutes after stress perfusion imaging, a second bolus of 0.1 mmol gadobutrol was injected, and rest perfusion images were obtained with the same orientation and position before and after the administration of adenosine. Spaciotemporal (k-t) sensitivity encoding perfusion cardiac MRI was used in combination with a saturation recovery gradient-echo pulse sequence for both of these sequences (TR/TE, 3.1/1.1; flip angle, 20 ; saturation prepulse delay, 110 milliseconds; partial Fourier sampling; acquisition window, 120 milliseconds; section thickness, 10 mm; k-t factor of five with 11 k-t interleaved training profiles; effective acceleration, 3.7; and three sections acquired sequentially during a single R-R interval), as previously shown [17 19]. High-spatial-resolution perfusion cardiac MRI was performed with an in-plane resolution of mm. Ten minutes after rest perfusion, late gadolinium enhancement (LGE) images were acquired in the continuous short-axis view using an inversionrecovery gradient-recalled echo MR sequence with the following parameters: field of view, mm; 7.4/4.3; inversion time, milliseconds; flip angle, 20 ; matrix, ; and slice thickness, 10 mm. The optimal inversion time was chosen to null the signal from normal myocardium. Coronary CTA Data Analysis Coronary CTA data analysis was performed by two independent radiologists who were both blinded to the clinical history and to the results from any other test (including perfusion cardiac MRI). All images were evaluated using transverse source images and multiplanar reformation. All segments with a diameter 1 mm at their origin were included. Vessel segments distal to occlusions were excluded from analysis. Coronary segments were defined according to a scheme proposed by the American Heart Association [15]. The intermediate artery was designated as segment 16, if present, and was considered to belong to the left anterior descending coronary artery (LAD). First, both readers independently rated the image quality of each coronary segment as being diagnostic or nondiagnostic. Reasons for nondiagnostic image quality were assigned to motion or stair step artifacts, image noise, severe vessel wall calcifications, or insufficient contrast attenuation. Then, both readers visually estimated all coronary segments for the presence or absence of significant stenoses, defined as luminal diameter narrowing of less than 50%. In case of disagreement, a consensus reading was appended 1 week after the initial readout. Segments containing stents were rated as either patent or nonpatent. Perfusion Cardiac MRI Data Analysis Perfusion cardiac MRI data analysis was performed visually by two different, independent radiologists who were both blinded to the clinical history and to the results from any other test (including ). In case of disagreement between the readers, a consensus reading was appended within 1 week. Segmental perfusion and LGE were scored using a 4-point scale (0 = definitely normal, 1 = probably normal, 2 = probably pathologic, 3 = definitely pathologic), as previously shown [5, 18 20]. Scores of 2 and 3 were considered abnormal to attain binomial scoring. Segments were considered to contain perfusion defects if either reduced peak signal intensity or delayed wash-in compared with remote segments was shown at stress but not at rest-perfusion or if late gadolinium enhancement was present [20]. Myocardial territories were assigned to the three major coronary arteries according to standard definitions [16]. Assessment of Hemodynamically Relevant Stenoses A hemodynamically relevant coronary stenosis was defined as a lesion with a diameter narrowing exceeding 50%, inducing a perfusion defect in its subtending myocardial territory on perfusion cardiac MRI. To be rated as hemodynamically relevant stenosis, both, luminal narrowing of greater than 50% on CA or as well as a perfusion defect on perfusion cardiac MRI had to be present. A coronary stenosis of greater than 50% without any associated myocardial ischemia was considered to be non-flow-limiting and thus not hemodynamically relevant. Conversely, a perfusion defect in a territory subtended by a coronary artery having a stenosis of 50% or less was considered to represent a false-positive perfusion cardiac MRI result. Comparisons were performed on an intention-to-diagnose basis. Thus, segments with nondiagnostic image quality on were censored as positive for disease. 922 AJR:194, April 2010

4 Statistical Analysis Quantitative data are expressed as mean ± SD and categoric data are given in proportions and percentages. Statistical comparison of means was performed using an unpaired two-tailed Student s t-test and comparison of categorical data using a chi-square test with Yates correction. A p value < 0.05 was considered significant for all tests. Sensitivity, specificity, positive (PPV) and negative predictive value (NPV), and accuracy were obtained from contingency tables and their respective 95% CIs were calculated from binomial expression. Diagnostic performance of was assessed by comparison with the results from CA, which was considered the reference standard for coronary stenosis evaluation. For assessment of performance in diagnosis of hemodynamically relevant stenoses, coronary CTA with perfusion cardiac MRI was compared with CA with perfusion cardiac MRI as the standard of reference. Additionally, diagnostic performance of with perfusion cardiac MRI and of perfusion cardiac MRI alone was assessed by comparison with CA, which served as the reference standard for coronary stenosis evaluation. The significance of differences in diagnostic performance was evaluated using the Wilcoxon s signed rank test. Agreement between methods was assessed by Cohen s kappa statistics. Kappa values of up to 0.4 were considered positive but poor agreement; , good agreement; and greater than 0.76, excellent agreement. Statistical analyses were performed using commercially available software (SPSS release 15.0, SPSS). Results Perfusion Cardiac MRI Perfusion cardiac MRI revealed ischemia in 133 myocardial segments and infarcts in 55 myocardial segments in 28 of 47 (59.6%) and 16 of 47 patients (34.0%), respectively. A myocardial defect (perfusion deficit or infarct according to Klem et al. [20]) was seen in 30 of 47 of the patients (63.8%). The distribution of the segmental defects among the different coronary artery territories is listed in Table 2. Conventional Coronary Angiography CA revealed greater than 50% diameter stenoses in 75 of 141 (53.2%) coronary arteries in 33 of 47 patients (70%) (Table 2). Prospectively ECG-Gated Coronary CTA Thirty-two patients were examined with a tube voltage of 120 kv and 15 patients with 100 kv. The average effective radiation dose per patient was 2.5 ± 1.1 msv. Of the total of 752 coronary segments in 141 coronary arteries, 104 of 752 (13.8%) TABLE 2: Results from Prospectively ECG-Gated Coronary CT Angiography (CTA), Perfusion Cardiac MRI, and Conventional Catheter Angiography (CA) Prospectively ECG-gated Technique No. of Patients (n = 47) Patients with stenoses at 33 LMA and LAD stenoses 32 LCX stenoses 21 RCA stenoses 23 Perfusion cardiac MRI CA Ischemia 28 Infarct 16 LAD defects 23 LCX defects 18 RCA defects 25 Patients with stenoses at CA 33 LMA + LAD stenoses 31 LCX stenoses 22 RCA stenoses 22 Combined plus perfusion cardiac MRI 29 Combined CA plus perfusion cardiac MRI 30 Note LMA = left main artery, LAD = left anterior descending artery, LCX = left circumflex artery, RCA = right coronary artery. coronary segments were either 1 mm or not present. Ten segments (1.3%) were located distal to a coronary occlusion. Hence, 638 of 752 coronary segments were included in the analysis. Three of 638 (0.5%) coronary segments (two in the right coronary artery and one in the LAD) were not assessable because of insufficient image quality due to motion artifacts (n = 2) or heavy calcifications (n = 1). No stair-step artifacts, insufficient contrast attenuation, or image noise was found as a cause of nondiagnostic image quality in the remaining 635 segments. On an intend-to-diagnose basis, coronary CTA revealed coronary stenosis in 135 of 635 segments (21.3%), corresponding to 76 of 141 (53.9%) coronary arteries of 33 of 47 patients (70.2%) (Table 2). Comparison of Prospectively ECG-Gated Coronary CTA and CA of for the detection of coronary stenoses greater than 50% on CA were 96.7%, 92.9%, 92.9%, 96.7%, and 95.7%, respectively, in the patient-based analysis and 90.7%, 87.9%, 89.2%, 89.5%, and 89.4%, respectively, in the vessel-based analysis (Table 3). Overall agreement between and CA for the detection of coronary stenoses was 95.7% (κ = 0.9) and 89.4% (κ = 0.8), in the patient- and artery-based analyses, respectively. Comparison of Perfusion Cardiac MRI and CA of perfusion cardiac MRI for the detection of coronary stenoses greater than 50% on CA were 90.9%, 100%, 82.4%, 100%, and 93.6%, respectively, in the patient-based analysis and 66.7%, 75.8%, 66.7%, 75.8%, and 70.9%, respectively, in the vessel-based analysis (Table 3). Overall agreement between perfusion cardiac MRI and CA for the detection of coronary stenoses was 93.6% (κ = 0.9) and 70.9% (κ = 0.4), in the patient- and arterybased analyses, respectively. Comparison of Prospectively ECG-Gated Coronary CTA Plus Perfusion Cardiac MRI and CA of plus perfusion cardiac MRI for the detection of coronary stenoses on CA were 87.9%, 100%, 77.8%, 100%, and 91.5%, respectively, in the patient-based AJR:194, April

5 TABLE 3: Diagnostic Performance of Prospectively ECG-Gated Coronary CT Angiography (CTA) and Perfusion Cardiac MRI as Evaluated by Coronary Angiography (CA) and Perfusion Cardiac MRI as Standard of Reference Technique Sensitivity (%) Specificity (%) NPV (%) PPV (%) Accuracy (%) Per artery Coronary CTA vs CA 90.7 (83 98, 68/75) 87.9 (79 97, 58/66) 89.2 (81 98, 58/65) 89.5 (82 97, 68/76) 89.4 (84 95, 126/141) Cardiac MRI vs CA 66.7 (55 78, 50/75) 75.8 (65 87, 50/66) 66.7 (55 78, 50/75) 75.8 (65 87, 50/66) 70.9 (63 79, 100/141) Coronary CTA plus cardiac MRI vs CA 65.3 (54 77, 49/75) 93.9 (87 100, 62/66) 70.5 (60 81, 62/88) 92.5 (84 100, 49/53) 78.7 (72 86, 111/141) Coronary CTA plus cardiac MRI vs CA 90.7 (82 99, 49/54) 95.4 (90 100, 83/87) 94.3 (89 100, 83/88) 92.5 (84 100, 49/53) 93.6 (89 98, 132/141) plus cardiac MRI Per patient Coronary CTA vs CA 96.7 (90 100, 32/33) 92.9 (76 100, 13/14) 92.9 (76 100, 13/14) 96.7 (90 100, 32/33) 95.7 (89 100, 45/47) Cardiac MRI vs CA 90.9 (80 100, 30/33) 100 (96 100, 14/14) 82.4 (61 100, 14/17) 100 (98 100, 30/30) 93.6 (86 100, 44/47) Coronary CTA plus cardiac MRI vs CA 87.9 (75 100, 29/33) 100 (96 100, 14/14) 77.8 (56 100, 14/18) 100 (98 100, 29/30) 91.5 (82 100, 43/47) Coronary CTA plus cardiac MRI vs CA plus cardiac MRI 96.7 (89 100, 29/30) 100 (97 100, 17/17) 94.4 (81 100, 17/18) 100 (98 100, 29/29) 97.9 (93 100, 46/47) Note Data in parentheses are 95% CI and number/total. NPV = negative predictive value, PPV = positive predictive value. analysis and 65.3%, 93.9%, 70.5%, 92.5%, and 78.7%, respectively, in the vessel-based analysis (Table 3). Overall agreement between plus perfusion cardiac MRI and CA for the detection of coronary stenoses was 91.5% (κ = 0.8), and 78.7% (κ = 0.6) in the patient- and artery-based analyses, respectively. Assessment of Hemodynamically Relevant Stenoses On combined analyses of plus perfusion cardiac MRI as well as CA plus perfusion cardiac MRI, 29 of 47 (61.7%) and 30 of 47 patients (63.8%) had flow-limiting stenoses, respectively (Fig. 2). One patient with significant stenosis in the intermediate branch on CA along with perfusion defects in the LAD territory on perfusion cardiac MRI was judged to have no significant stenoses on (Fig. 3). of the combination plus Fig year-old man with coronary stenosis. A and B, Conventional coronary angiography (CTA) (A) and prospectively ECG-gated with multiplanar reformation (B) images depict stenosis in proximal left anterior descending coronary artery (arrowheads). C and D, Perfusion cardiac MR images after adenosine stress (C) and at rest (D) indicate ischemia in anteroseptal midventricular segment (arrowheads, C). Thus, this patient represented true-positive rating in comparison of prospectively ECG-gated coronary CTA with coronary angiography as well as in assessment of hemodynamically relevant stenoses. perfusion cardiac MRI for the detection of flow-limiting coronary stenoses compared with CA plus perfusion cardiac MRI were 96.7%, 100%, 94.4%, 100%, and 97.9%, respectively, in a patient-based analysis and 90.7%, 95.4%, A C 94.3%, 92.5%, and 93.6%, respectively, in an artery-based analysis (Table 3). Overall agreement between plus perfusion cardiac MRI and CA plus perfusion cardiac MRI for the detection of B D 924 AJR:194, April 2010

6 hemodynamically relevant CAD was 97.9% (κ = 0.96) and 93.6% (κ = 0.86), respectively, in a patient- and artery-based analyses. One patient was judged to have significant stenosis on but no significant stenosis on CA. Thus, this patient represented a false-positive result when directly comparing with CA. Because perfusion cardiac MRI in this patient showed no perfusion defects, the initially false-positive result from became a true-negative result when information from coronary CTA and perfusion cardiac MRI was combined. Diagnostic performance of the combined approach for assessing CAD was higher than that of sole morphoanatomic comparison between and CA (Table 3). However, the differences were not significant (p = 0.56 and p = 0.29 for vessel-based and patient-based analysis, respectively). A C Fig year-old man with stenosis in intermediate artery A, Coronary angiography image shows stenosis in intermediate artery (arrowheads). B, On prospectively ECG-gated coronary CT angiography image (CTA), this stenosis (arrowheads) also was depicted retrospectively but was missed during initial readout. C and D, Stress perfusion cardiac MR image (C) shows perfusion deficit in anteroseptal mid ventricular segment (arrowheads, C), which is not seen in similar image under rest (D). Thus, this patient represented falsenegative rating in. Discussion Our study results indicate that the diagnostic performance of combined with perfusion cardiac MRI is comparable to a clinical standard approach using CA and perfusion cardiac MRI in the detection of CAD. The sole morphoanatomic evaluation of CAD as established by CA or has limitations, making functional assessment necessary. Comprehensive noninvasive anatomic and functional imaging best identifies patients who are likely to benefit most from secondary preventive measures and medical therapy (i.e., coronary atherosclerosis without ischemia) or who may be candidates for coronary revascularization (coronary atherosclerosis with ischemia) [21]. The importance of a combined imaging approach (similar to that used in this study) also has been highlighted by the recent controversial debate about the indications for coronary revascularization [22, 23]. Revascularization of non-flow-limiting coronary stenoses has shown no benefit for the patient, neither from a prognostic nor a symptomatic point of view [24]. Furthermore, several studies have shown that revascularization fails to improve the prognosis of patients with stable CAD compared with medical treatment, which was attributed to the lack of a proof of ischemia for guiding the intervention [23, 25]. Thus, current guidelines recommend an objective proof of ischemia before elective revascularization of coronary stenoses [3, 22, 26]. Our study results suggest that a combined noninvasive approach with prospectively ECG-gated and perfusion cardiac MRI detects hemodynamically significant stenoses with similar diagnostic accuracy as the combination of CA and perfusion cardiac MRI. It may thereby serve as a filter test assessing the necessity of coronary revascularization. The advantage of such an approach would be besides its noninvasiveness the very low radiation exposure to patients, on average 2.5 msv, similar to the values reported in the literature [9, 11, 12]. This value is within the lower range of radiation doses reported for conventional CA [27]. The excellent diagnostic performance of prospectively ECG-gated compared with CA in depicting stenoses has recently been shown [11]. By adding functional information from perfusion cardiac MRI and thus assessing for hemodynamically relevant stenoses, the diagnostic performance can further be improved. The increases in specificity, NPV, PPV, and accuracy of the combined approach for assessing CAD compared with the purely morphoanatomic comparison are mainly due to one patient whose stenosis in the LAD was judged to be significant on but not with CA, thus representing a false-positive rating from. Because perfusion cardiac MRI did not show any deficit in the subtending myocardial territories, this patient could be reclassified to represent a true-negative result when assessing hemodynamically relevant coronary stenoses. Our study had some limitations: First, the number of study participants in our preliminary study was limited. Thus, it was not possible to perform subanalyses across different B D AJR:194, April

7 patient groups, such as those with multivessel versus single-vessel disease. Second, prospectively ECG-gated is feasible only in patients with regular heart rates, thus limiting the application of the technique to a broader patient population [9 12]. Third, our patient cohort had a high pretest probability, which may have influenced the NPV and PPV. It has to be taken into account that results may be different in a patient cohort with a lower pretest probability of disease. Finally, nuclear testing rather than cardiac MRI may be considered the clinical reference standard approach for evaluating myocardial perfusion. On the other hand, cardiac MRI has recently shown similar performance characteristics to previously validated nuclear tests [4, 5]. Finally, the influence of the two approaches on therapeutic decision making before percutaneous coronary intervention or coronary artery bypass graft was not evaluated. Further studies should focus on the influence of purely morphoanatomic tests versus combined morphoanatomic and functional tests on clinical decision making. In conclusion, this study shows that a comprehensive workup of hemodynamically relevant CAD, the combination of and perfusion cardiac MRI, shows diagnostic performance comparable to CA plus perfusion cardiac MRI. Our preliminary data suggest therefore that may replace CA in the diagnosis of hemodynamically relevant CAD. References 1. White CW, Wright CB, Doty DB, et al. Does visual interpretation of the coronary arteriogram predict the physiologic importance of a coronary stenosis? N Engl J Med 1984; 310: Topol EJ, Nissen SE. Our preoccupation with coronary luminology: the dissociation between clinical and angiographic findings in ischemic heart disease. Circulation 1995; 92: Smith SC Jr, Feldman TE, Hirshfeld JW Jr, et al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary intervention: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for Percutaneous Coronary Intervention). J Am Coll Cardiol 2006; 47:e1 e Sakuma H, Suzawa N, Ichikawa Y, et al. Diagnostic accuracy of stress first-pass contrast-enhanced myocardial perfusion MRI compared with stress myocardial perfusion scintigraphy. AJR 2005; 185: Schwitter J, Wacker CM, van Rossum AC, et al. MR IMPACT: comparison of perfusion cardiac magnetic resonance with single-photon emission computed tomography for the detection of coronary artery disease in a multicentre, multivendor, randomized trial. Eur Heart J 2008; 29: Heijenbrok-Kal MH, Fleischmann KE, Hunink MG. Stress echocardiography, stress single-photonemission computed tomography and electron beam computed tomography for the assessment of coronary artery disease: a meta-analysis of diagnostic performance. Am Heart J 2007; 154: Leschka S, Alkadhi H, Plass A, et al. Accuracy of MSCT coronary angiography with 64-slice technology: first experience. Eur Heart J 2005; 26: Ropers U, Ropers D, Pflederer T, et al. Influence of heart rate on the diagnostic accuracy of dualsource computed tomography coronary angiography. J Am Coll Cardiol 2007; 50: Earls JP, Berman EL, Urban BA, et al. Prospectively gated transverse coronary CT angiography versus retrospectively gated helical technique: improved image quality and reduced radiation dose. Radiology 2008; 246: 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: Scheffel H, Alkadhi H, Leschka S, et al. Lowdose CT coronary angiography in the step-andshoot mode: diagnostic performance. Heart 2008; 94: Stolzmann P, Leschka S, Scheffel H, et al. Dualsource CT in step-and-shoot mode: noninvasive coronary angiography with low radiation dose. Radiology 2008; 249: Stolzmann P, Scheffel H, Leschka S, et al. Influence of calcifications on diagnostic accuracy of coronary CT angiography using prospective ECG triggering. AJR 2008; 191: Stolzmann P, Donati OF, Scheffel H, et al. Lowdose CT coronary angiography for the prediction of myocardial ischemia. Eur Radiol 2010; 20: Austen WG, Edwards JE, Frye RL, et al. A reporting system on patients evaluated for coronary artery disease: report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 1975; 51: Cerqueira MD, Weissman NJ, Dilsizian V, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Int J Cardiovasc Imaging 2002; 18: Plein S, Ryf S, Schwitter J, Radjenovic A, Boesiger P, Kozerke S. Dynamic contrast-enhanced myocardial perfusion MRI accelerated with k-t SENSE. Magn Reson Med 2007; 58: Gebker R, Jahnke C, Paetsch I, et al. MR myocardial perfusion imaging with k-space and time broad-use linear acquisition speed-up technique: feasibility study. Radiology 2007; 245: Plein S, Schwitter J, Suerder D, Greenwood JP, Boesiger P, Kozerke S. k-space and time sensitivity encoding accelerated myocardial perfusion MR imaging at 3.0 T: comparison with 1.5 T. Radiology 2008; 249: Klem I, Heitner JF, Shah DJ, et al. Improved detection of coronary artery disease by stress perfusion cardiovascular magnetic resonance with the use of delayed enhancement infarction imaging. J Am Coll Cardiol 2006; 47: Meijboom WB, Van Mieghem CA, van Pelt N, et al. Comprehensive assessment of coronary artery stenoses: computed tomography coronary angiography versus conventional coronary angiography and correlation with fractional flow reserve in patients with stable angina. J Am Coll Cardiol 2008; 52: Fox K, Garcia MA, Ardissino D, et al. 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: Boden WE, O Rourke RA, Teo KK, et al. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med 2007; 356: Pijls NH, van Schaardenburgh P, Manoharan G, et al. Percutaneous coronary intervention of functionally nonsignificant stenosis: 5-year follow-up of the DEFER Study. J Am Coll Cardiol 2007; 49: Bucher HC, Hengstler P, Schindler C, Guyatt GH. Percutaneous transluminal coronary angioplasty versus medical treatment for non-acute coronary heart disease: meta-analysis of randomised controlled trials. BMJ 2000; 321: Gibbons RJ, Abrams J, Chatterjee K, et al. ACC/ AHA 2002 guideline update for the management of patients with chronic stable angina summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Chronic Stable Angina). J Am Coll Cardiol 2003; 41: Einstein AJ, Moser KW, Thompson RC, Cerqueira MD, Henzlova MJ. Radiation dose to patients from cardiac diagnostic imaging. Circulation 2007; 116: AJR:194, April 2010