- Email: [email protected]
Dipyridamole stress and rest transmural myocardial perfusion ratio evaluation by 64 detector-row computed tomography
Journal of Cardiovascular Computed Tomography (2011) 5, 443–448
Original Research Article
Dipyridamole stress and rest transmural myocardial perfusion ratio evaluation by 64 detector-row computed tomography Roberto C. Cury, MDa,*, Tiago A. Magalh~aes, MDa, Antonio T. Paladino, MDb, Afonso A. Shiozaki, MDb, Marcela Perini, MDb, Tiago Senra, MDb, Pedro A. Lemos, MDc, Ricardo C. Cury, MDd, Carlos E. Rochitte, MDa a
Cardiovascular MR and CT Section, Heart Institute (InCor), University of S~ao Paulo Medical School and Heart Hospital (HCOR), S~ ao Paulo, Brazil; bCardiovascular CT Section, Dante Pazzannese Institute, S~ao Paulo, Brazil; c Invasive Cardiology Section, Heart Institute (InCor), University of S~ao Paulo Medical School, S~ao Paulo, Brazil and d Baptist Cardiac and Vascular Institute, Cardiovascular MR and CT Program, Miami, FL, USA KEYWORDS: Myocardial ischemia; Multidetector computed tomography; Stress perfusion; Dipyridamole; Transmural perfusion ratio
BACKGROUND: Myocardial stress CT perfusion (CTP) can detect myocardial ischemia. OBJECTIVE: We evaluated the transmural perfusion ratio (TPR) of dipyridamole stress CTP to detect significant coronary stenosis (.70%) defined by quantitative invasive coronary angiography (ICA). METHODS: Twenty-six patients (61.6 6 8.0 years old; 14 males), without prior myocardial infarction, with positive single-photon emission computed tomography (SPECT; ,2 months) and clinical indication for ICA, underwent a customized multidetector-row CT (MDCT) protocol with rest/stress myocardial perfusion evaluation and coronary CT angiography. TPR was defined as mean subendocardial divided by mean subepicardial attenuation and quantified on rest and stress MDCT images. Abnormal TPR was defined as 2 SDs below the mean rest TPR. RESULTS: All 26 patients completed the CT protocol with no adverse events. Rest TPR was measured in all patients with a mean of 1.06 6 0.11, and abnormal TPR was considered ,0.85. For 6 patients with normal coronary arteries by ICA, the mean TPR of territories with a previous positive perfusion defect in SPECT was 1.02 6 0.18 (95% CI, 0.86–1.18; n 5 6), and mean TPR of territories without perfusion defect in SPECT was 1.03 6 0.09 (95% CI, 20.95 to 1.11; n 5 12; P 5 0.83). Mean stress TPR in territories with positive SPECT and significant coronary artery disease by quantitative ICA was 0.71 6 0.13 (95% CI, 20.64 to 0.77) and in the remote myocardial was 1.01 6 0.09 (95% CI, 20.96 to 1.06; P , 0001). In these territories, a significant Pearson’s correlation was observed (r 5 20.74, P , 0.001). CONCLUSION: TPR has a good correlation with SPECT and ICA to detect significant coronary stenosis. Ó 2011 Society of Cardiovascular Computed Tomography. All rights reserved.
Conflict of interest: The authors report no conflicts of interest. * Corresponding author. E-mail address: [email protected]
Submitted August 16, 2011. Accepted for publication October 30, 2011.
1934-5925/$ - see front matter Ó 2011 Society of Cardiovascular Computed Tomography. All rights reserved. doi:10.1016/j.jcct.2011.10.012
444
Journal of Cardiovascular Computed Tomography, Vol 5, No 6, November/December 2011
Introduction In the past decade, multidetector computed tomography (MDCT) has emerged as a powerful anatomical tool to evaluate patients with suspect coronary artery disease (CAD). Sixty-four slices MDCT angiography showed high diagnostic accuracy to detect significant coronary stenosis.1–8 Despite MDCT being considered a modality to detect coronary stenosis, in the past 2 years several studies have shown the ability of MDCT to detect myocardial ischemia.9–12 We previously described the feasibility of dipyridamole stress 64-row CT for combining both anatomical and functional information.9 The aim of this study was to evaluate the ability of dipyridamole stress MDCT transmural myocardial perfusion ratio (TPR) to detect significant coronary stenosis (.70%) by quantitative invasive coronary angiography (QCA). We also sought to investigate the correlation of TPR with single-photon emission CT (SPECT) perfusion defects.
Methods This is a prospective cohort study of consecutive patients with a previous reversible perfusion defect at SPECTwho met eligibility criteria and who agreed to sign informed consent to participate in this CT research stress perfusion study that was approved by our institutional ethic review board. Inclusion criteria included men and nonpregnant women with a clinical suspicion of CAD and a reversible perfusion defect at SPECT, in the past 60 days, and clinical indication of invasive coronary angiography. Only dipyridamole or adenosine stress SPECT studies were included. SPECT was performed with a 1-day protocol with 2-metoxil-isobutilisonitrila-99mTC (sestamibi-99mTv; Cardiolite, BristolMyers Squibb Medical Imaging, New York, NY). The decision to undergo cardiac catheterization was made on a clinical basis by each patient’s cardiologist. Exclusion criteria included history of previous myocardial infarct or CABG asthma, pregnancy, renal insufficiency (serum creatinine . 1.5 mg/dL), and known allergy to iodinated contrast or dipyridamole. Patient selection is shown in Figure 1. Twenty-six patients who met eligibility criteria underwent a customized (20 minutes) MDCT protocol with rest/stress myocardial perfusion evaluation and coronary CT angiography (CTA).9 MDCT was performed in a 64 detector-row scanner (Aquillion 64; Toshiba Medical System, Otawara, Japan). All patients signed informed consent. Results of coronary CTA and myocardial CT perfusion (CTP) were blinded to the attending physician. All patients were instructed to have a diet free of caffeine 24 hours before examination. MDCT customized protocol included 2 contrast-enhanced scans. A calcium score was not performed. The first scan was a stress CTP only, with low resolution. The second scan was done for coronary
Figure 1 Flow diagram showing patient selection. CABG, coronary artery bypass graft.
arteries and rest myocardial perfusion evaluation (Table 1). Stress myocardial perfusion by MDCT image was obtained 2 minutes after intravenous infusion of dipyridamole, 0.56 mg/kg during 4 minutes. During dipyridamole infusion, heart rate, blood pressure, and patients’ symptoms were monitored. Helical acquisition for myocardial perfusion by MDCT was initiated with a real-time bolus tracking technique, at the time of peak left atria filling with contrast, which was determined visually (no specific threshold in Hounsfield units was used). Immediately after the stress scan, infusion with aminophylline (USP) 240 mg was started to revert vasodilatation induced by dipyridamole. Before a second perfusion/CT angiographic scan at rest, intravenous metoprolol was administered until a maximum dose of 20 mg to lower heart rate and improve image quality. Scanning at rest was performed with automatic trigger detection on the descending aorta set to signal density of 180 HU. The standard American College of Cardiology/American Heart Association (ACC/AHA) 17-segment model was used
Cury et al
Dipyridamole stress/rest myocardial TPR
Table 1 Customized protocol used in all patients undergoing stress myocardial CT perfusion First scan
Second scan
Data
Stress TPR
ECG gating Iodinated contrast, mL (mL/s) mA
Retrospective 60 (3)
Coronary CTA Rest TPR Retrospective 80–90 (5)
kV
120
100
Female, 240–270 Male, 370–400 120
ECG, electrocardiogram.
to identify perfusion defects. For comparison of perfusion data with coronary anatomical data derived from MDCT, we consolidated the segmental data into 3 territories according to the ACC/AHA recommendation (per-vessel analysis).13 Myocardial segments were correlated to the specific coronary territory depending on the coronary dominancy. Perfusion datasets were evaluated on a true short axis (apical, medium, and basal slices), with 2- and 4-chamber views, with 8-mm thick average multiplanar reformatted images. Thicker slices were used to improve low-contrast resolution for visual myocardium perfusion analysis, as described previously.10 Display parameters included a narrow window width and level (350 W and 150 L). Initial evaluation of perfusion defects started in the diastolic phase, and, for potential artifacts and motion, readers could use the systolic phase to confirm the perfusion defect. True perfusion defects were defined as subendocardial hypoenhancement, encompassing R25% of transmural extent, present in different phases of the cardiac cycle and within a specific coronary territory. Two independent observers with no knowledge of clinical data or other examinations performed quantitative analysis to measure TPR in all 17 ACC/AHA segment model, excluding the apex. The rest and stress TPR was analyzed with Osirix software (Pixmeo, Geneve, Switzerland). All perfusion images were evaluated in the short-axis view, divided circumferentially in 2 layers (manually) with same thickness to define subendocardium and subepicardium, as shown in Figure 2. For all TPR measurements, endocardial and epicardial contours were traced, excluding papillary muscles. After that, AHA myocardial segmentation was done, and mean myocardial density (Hounsfield unit; HU) was measured in both layers to calculate TPR in each segment (TPR 5 subendocardium mean density/ subepicardium mean density). Mean TPR was measured in each segment at rest and stress phase. The segmental TPR values were assigned to the respective coronary territories, considering the lowest segment TPR. Image quality for perfusion and anatomy evaluation was considered interpretable when in R1 phase of the
445 cardiac cycle; coronary and perfusion evaluations were diagnostic. QCA was performed on all patients submitted to coronary invasive angiography. Blinded readers considered coronary stenosis R 70% by QCA as a hemodynamic significant lesion. Accuracy of TPR in a patient-based analysis was considered for matching segments on SPECT.
Statistical analysis Statistical analyses were performed with Med-Calc version 11.4.4.0 (Meriakerke, Belgium). Data were expressed as mean 6 SD. Interobserver variability was compared with the use of Bland-Altman plots. The relation between percentage of luminal stenosis and TPR was compared with the use of Pearson’s correlation. The area under the curve was calculated and reported with 95% CIs. The threshold of significance was P , 0.05.
Results Twenty-six patients (61.6 6 8.0 years old; 14 males) were enrolled in the study and underwent MDCT (stress perfusion and coronary angiography) and invasive coronary angiography. All 26 patients completed the protocol with no adverse events, with mean CT radiation dose14 of 14.4 6 2.9 mSv and interpretable scans. Rest MDCT images were used to calculate TPR, and the normal segments distribution of myocardial TPR is shown in Figure 3 (n 5 416 myocardial segments). Mean rest TPR was 1.06 6 0.11. Abnormal TPR was considered 2 SDs below the mean (1.06 – 0.22 5 0.84), that is, ,0.85. Dipyridamole stress MDCT images were analyzed, and the TPR was quantified in 416 myocardial segments and
Figure 2 Myocardial segmentation in the short axis of the left ventricle used for TPR measurement.
446
Journal of Cardiovascular Computed Tomography, Vol 5, No 6, November/December 2011
Figure 3 Range distribution of TPR measures on rest images (n 5 416 myocardial segments).
consolidated in 78 territories, which were correlated with the degree of coronary stenosis measured by the QCA. For 6 patients with normal coronary arteries by invasive coronary angiography, 96 myocardial segments were consolidated in 18 territories to calculate each TPR. After calculating the TPR, these territories were divided in 2 groups: one with SPECT perfusion defect and another without. The mean TPR of territories with a previous positive perfusion defect in SPECT was 1.02 6 0.18 (95% CI, 0.86–1.18; n 5 6), and mean TPR of territories without perfusion defect in SPECT was 1.03 6 0.09 (95% CI, 0.95–1.11; n 5 12). No significant statistical difference was observed (P 5 0.83).
The territories TPR analysis for all patients (n 5 78) during stress showed a mean TPR of 0.71 6 0.13 (95% CI, 20.64 to 0.77; n 5 17) in territories corresponding to coronary stenosis R 70% by QCA and positive SPECT (Fig. 4), and 1.01 6 0.09 (95% CI, 20.96 to 1.06; n 5 61) in remote territories corresponding to coronary stenosis , 70% (P , 0.001; Fig. 5). A significant inverse linear correlation was observed between the TPR and the percentage f diameter stenosis (r 5 20.74, P , 0.001; Fig. 6). The sensitivity, specificity, positive predictive value, negative predictive value, and area under the curve for TPR detecting stenosis corresponding to a perfusion defect on SPECT in a patient-based analysis (n 5 26) was 94% (16 of 17; 95% CI, 71.3%–99.9%), 78% (7 of 9; 95% CI, 40.0%–97.2%), 89% (16 of 18; 95% CI, 64.4%–98.7%), 87.5% (7 of 8; 95% CI, 47.3%–99.7%) and 0.85 (95% CI, 0.66–0.96, respectively). Bland Altman plot for interobserver variability in TPR quantification (Fig. 7) showed a mean difference of 20.01 (95% CI, 20.20 to 0.18). Interobserver correlation was good (k 5 0.72; 95% CI, 0.67–0.77).
Discussion Although SPECT myocardial perfusion imaging represents an established technique, limitations include its inability to evaluate transmural myocardial perfusion and
Figure 4 Correlation among TPR by MDCT (A), SPECT (B), coronary angiography by MDCT (C), and invasive coronary angiography (D), mid inferolateral TPR 5 0.78 and left circumflex artery coronary stenosis .70%.
Cury et al
Dipyridamole stress/rest myocardial TPR
Figure 5 Mean TPR analysis comparing territories without significant coronary stenosis (QCA , 70%) with territories with positive SPECT and corresponding coronary stenosis R 70% by QCA.
to accurately quantify the extent of subendocardial perfusion defects because of limited spatial resolution. Other than myocardial perfusion abnormalities, SPECT may also show functional findings that suggest ischemia, such as left ventricular ejection fraction decreases by .5% after stress, pulmonary uptake, and transitory dilatation of the left ventricle.15–17 Blood flow, metabolism, and the contraction and relaxation dynamics of the myocardium are heterogeneous. The subendocardium is more susceptible to ischemic injury than the subepicardium. Myocardial ischemia starts in the subendocardium and spreads like a wavefront toward the subepicardium. Because perfusion initially decreases in the subendocardium as coronary flow is gradually reduced,
447 first-pass imaging with the high spatial and temporal resolution prerequisite allows early detection of coronary stenosis, as indicated by previous magnetic resonance imaging data.18 Cardiac CT can determine the TPR, which is inversely correlated with the degree of coronary stenosis. A recent study measured TPR with the use of automatic delineation of myocardium into 3 layers: subendocardium, mesocardium, and subepicardium.11 The TPR cutoff used was 1 SD below the mean (TPR , 0.99), and the TPRs observed were 1.12 6 0.13 for patients with no obstructive CAD and 0.91 6 0.10 for patients with coronary stenosis . 70%. In that study, a 256-row scanner was used for rest and stress TPR evaluation. In this regard, our results in a 64-row scanner agreed with this previously published study. Although we measured TPR through manual measurement of the myocardium in 2 layers, TPR was inversely correlated with the severity of stenosis. Dividing the myocardial wall in 3 layers would potentially increase the low attenuation of the subendocardial layer and might increase sensitivity. The present study used a lower cutoff for describing a normal TPR (2 SDs below the mean; TPR , 0.85), although we found TPR values in patients with and without obstructive CAD (0.71 6 0.13 vs 1.01 6 0.09, respectively) comparable to that previously described.11 It is worth noting that in our analysis we also used a combined reference method for defining the true ischemic myocardial segment, that is, a myocardial segment with perfusion defect on SPECT combined with a corresponding coronary stenosis . 70% by QCA. In addition, the normal values for TPR observed in patients without CAD were independent of SPECT results, suggesting that TPR was a valuable tool in identifying patients with false-positive SPECT.
Figure 6 Correlation between TPR and percentage of diameter stenosis by QCA in which territories were matched to SPECT territories showing perfusion abnormalities (n 5 26).
448
Journal of Cardiovascular Computed Tomography, Vol 5, No 6, November/December 2011
Figure 7 Bland Altman plot showing the agreement in TPR quantification of 416 myocardial segments on the stress CTP, between observer 1 and observer 2.
Conclusions In conclusion, TPR has a good correlation with SPECT and coronary invasive angiography to detect significant coronary stenosis.
References 1. Leber AW, Knez A, von Ziegler F, Becker A, Nikolau K, Paul S, Wintersperger B, Reiser M, Becker CR, Steinbeck G, Boekstegers P: 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;46:147–54. 2. Leschka S, Alkadhi H, Plass A, Desbiolles L, Grunenfelder J, Marincek B, Wildermuth S: Accuracy of MSCT coronary angiography with 64-slice technology: first experience. Eur Heart J. 2005;26:1482–7. 3. Mollet NR, Cademartiri F, van Mieghem CA, Runza G, McFadden EP, Baks T, Serruys PW, Krestin GP, de Feyter PJ: High-resolution spiral computed tomography coronary angiography in patients referred for diagnostic conventional coronary angiography. Circulation. 2005;112:2318–23. 4. 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;46:552–7. 5. Pugliese F, Mollet NR, Runza G, van Mieghem C, Meijboom WB, Malagutti P, Baks T, Krestin GP, de Feyter PJ, Cademartiri F: Diagnostic accuracy of non-invasive 64-slice CT coronary angiography in patients with stable angina pectoris. Eur Radiol. 2006;16:575–82. 6. Ropers D, Rixe J, Anders K, Kuttner A, Baum U, Bautz W, Daniel WG, Achenbach S: Usefulness of multidetector row spiral computed tomography with 64- x 0.6-mm collimation and 330-ms rotation for the noninvasive detection of significant coronary artery stenoses. Am J Cardiol. 2006;97:343–58.
7. Miller JM, Rochitte CE, Dewey M, Arbab-Zadeh A, Niinuma H, Gottlieb I, Paul N, Clouse ME, Shapiro EP, Hoe J, Lardo AC, Bush DE, de Roos A, Cox C, Brinker J, Lima JA: Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med. 2008; 359:2324–36. 8. Miller JM, Dewey M, Vavere AL, Rochitte CE, Niinuma H, Arbab-Zadeh A, Paul N, Hoe J, de Roos A, Yoshioka K, Lemos PA, Bush DE, Lardo AC, Texter J, Brinker J, Cox C, Cloude ME, Lima JA: Coronary CT angiography using 64 detector rows: methods and design of the multi-centre trial CORE-64. Eur Radiol. 2009;19: 816–28. 9. Cury RC, Magalh~aes TA, Borges AX, Shiozaki AA, Lemos PA, Junior JS, Meneghetti JC, Cury RC, Rochitte CE: Dipyridamole stress and rest myocardial perfusion by 64-detector row computed tomography in patients with suspected coronary artery disease. Am J Cardiol. 2010;106:310–5. 10. Blankstein R, Shturman LD, Rogers IS, Rocha-Filho JA, Okada DR, Sarwar A, Soni AV, Bezerra H, Ghoshhajra BB, Petranovic M, Loureiro R, Feuchtner G, Gewirtz H, Hoffmann U, Mamuya WS, Brady TJ, Cury RC: Adenosine-induced stress myocardial perfusion imaging using dual-source cardiac computed tomography. J Am Coll Cardiol. 2009;54:1072–84. 11. George RT, Arbab-Zadeh A, Miller JM, Kitagawa K, Chang HJ, Bluemke DA, Becker L, Yousuf O, Texter J, Lardo AC, Lima JA: Adenosine stress 64- and 256-row detector computed tomography angiography and perfusion imaging: a pilot study evaluating the transmural extent of perfusion abnormalities to predict atherosclerosis causing myocardial ischemia. Circ Cardiovasc Imaging. 2009;2: 174–82. 12. Ho KT, Chua KC, Klots E, Pankin C: Stress and rest dynamic myocardial perfusion imaging by evaluation of complete time-attenuation curves with dual-source CT. J Am Coll Cardiol. 2010;3:811–20. 13. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS: 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. Circulation. 2002;105:539–42. 14. Shrimpton PC. Assessment of patient dose in CT. Chilton, NRPB-PE/1/2004; 2004. Also published as Appendix C of the 2004 CT Quality Criteria. Available at: http://www.msct.info/CT_Quality_ Criteria.htm. Accessed May 5, 2006. 15. Hida S, Chikamori T, Tanaka H, Usui Y, Igarashi Y, Nagao T, Yamashina A: Diagnostic value of left ventricular function after stress and at rest in the detection of multivessel coronary artery disease as assessed by electrocardiogram-gated SPECT. J Nucl Cardiol. 2007; 14:68–74. 16. Shirai N, Yamagishi H, Yoshiyama M, Teragaki M, Akioka K, Takeuchi K, Yoshikawa J, Ochi H: Incremental value of assessment of regional wall motion for detection of multivessel coronary artery disease in exercise (201) TI gated myocardial perfusion imaging. J Nucl Med. 2002;43:443–50. 17. Desai D, Kozeski G, Akinboboye O: Detection of multivessel coronary artery disease: looking beyond the extent of perfusion abnormalities. J Nucl Cardiol. 2009;16:4–5. 18. Jerosch-Herold M, Wilke N: MR first pass imaging: quantitative assessment of transmural perfusion and collateral flow. Int J Card Imaging. 1997;13:205–18.
Original Research Article
Dipyridamole stress and rest transmural myocardial perfusion ratio evaluation by 64 detector-row computed tomography Roberto C. Cury, MDa,*, Tiago A. Magalh~aes, MDa, Antonio T. Paladino, MDb, Afonso A. Shiozaki, MDb, Marcela Perini, MDb, Tiago Senra, MDb, Pedro A. Lemos, MDc, Ricardo C. Cury, MDd, Carlos E. Rochitte, MDa a
Cardiovascular MR and CT Section, Heart Institute (InCor), University of S~ao Paulo Medical School and Heart Hospital (HCOR), S~ ao Paulo, Brazil; bCardiovascular CT Section, Dante Pazzannese Institute, S~ao Paulo, Brazil; c Invasive Cardiology Section, Heart Institute (InCor), University of S~ao Paulo Medical School, S~ao Paulo, Brazil and d Baptist Cardiac and Vascular Institute, Cardiovascular MR and CT Program, Miami, FL, USA KEYWORDS: Myocardial ischemia; Multidetector computed tomography; Stress perfusion; Dipyridamole; Transmural perfusion ratio
BACKGROUND: Myocardial stress CT perfusion (CTP) can detect myocardial ischemia. OBJECTIVE: We evaluated the transmural perfusion ratio (TPR) of dipyridamole stress CTP to detect significant coronary stenosis (.70%) defined by quantitative invasive coronary angiography (ICA). METHODS: Twenty-six patients (61.6 6 8.0 years old; 14 males), without prior myocardial infarction, with positive single-photon emission computed tomography (SPECT; ,2 months) and clinical indication for ICA, underwent a customized multidetector-row CT (MDCT) protocol with rest/stress myocardial perfusion evaluation and coronary CT angiography. TPR was defined as mean subendocardial divided by mean subepicardial attenuation and quantified on rest and stress MDCT images. Abnormal TPR was defined as 2 SDs below the mean rest TPR. RESULTS: All 26 patients completed the CT protocol with no adverse events. Rest TPR was measured in all patients with a mean of 1.06 6 0.11, and abnormal TPR was considered ,0.85. For 6 patients with normal coronary arteries by ICA, the mean TPR of territories with a previous positive perfusion defect in SPECT was 1.02 6 0.18 (95% CI, 0.86–1.18; n 5 6), and mean TPR of territories without perfusion defect in SPECT was 1.03 6 0.09 (95% CI, 20.95 to 1.11; n 5 12; P 5 0.83). Mean stress TPR in territories with positive SPECT and significant coronary artery disease by quantitative ICA was 0.71 6 0.13 (95% CI, 20.64 to 0.77) and in the remote myocardial was 1.01 6 0.09 (95% CI, 20.96 to 1.06; P , 0001). In these territories, a significant Pearson’s correlation was observed (r 5 20.74, P , 0.001). CONCLUSION: TPR has a good correlation with SPECT and ICA to detect significant coronary stenosis. Ó 2011 Society of Cardiovascular Computed Tomography. All rights reserved.
Conflict of interest: The authors report no conflicts of interest. * Corresponding author. E-mail address: [email protected]
Submitted August 16, 2011. Accepted for publication October 30, 2011.
1934-5925/$ - see front matter Ó 2011 Society of Cardiovascular Computed Tomography. All rights reserved. doi:10.1016/j.jcct.2011.10.012
444
Journal of Cardiovascular Computed Tomography, Vol 5, No 6, November/December 2011
Introduction In the past decade, multidetector computed tomography (MDCT) has emerged as a powerful anatomical tool to evaluate patients with suspect coronary artery disease (CAD). Sixty-four slices MDCT angiography showed high diagnostic accuracy to detect significant coronary stenosis.1–8 Despite MDCT being considered a modality to detect coronary stenosis, in the past 2 years several studies have shown the ability of MDCT to detect myocardial ischemia.9–12 We previously described the feasibility of dipyridamole stress 64-row CT for combining both anatomical and functional information.9 The aim of this study was to evaluate the ability of dipyridamole stress MDCT transmural myocardial perfusion ratio (TPR) to detect significant coronary stenosis (.70%) by quantitative invasive coronary angiography (QCA). We also sought to investigate the correlation of TPR with single-photon emission CT (SPECT) perfusion defects.
Methods This is a prospective cohort study of consecutive patients with a previous reversible perfusion defect at SPECTwho met eligibility criteria and who agreed to sign informed consent to participate in this CT research stress perfusion study that was approved by our institutional ethic review board. Inclusion criteria included men and nonpregnant women with a clinical suspicion of CAD and a reversible perfusion defect at SPECT, in the past 60 days, and clinical indication of invasive coronary angiography. Only dipyridamole or adenosine stress SPECT studies were included. SPECT was performed with a 1-day protocol with 2-metoxil-isobutilisonitrila-99mTC (sestamibi-99mTv; Cardiolite, BristolMyers Squibb Medical Imaging, New York, NY). The decision to undergo cardiac catheterization was made on a clinical basis by each patient’s cardiologist. Exclusion criteria included history of previous myocardial infarct or CABG asthma, pregnancy, renal insufficiency (serum creatinine . 1.5 mg/dL), and known allergy to iodinated contrast or dipyridamole. Patient selection is shown in Figure 1. Twenty-six patients who met eligibility criteria underwent a customized (20 minutes) MDCT protocol with rest/stress myocardial perfusion evaluation and coronary CT angiography (CTA).9 MDCT was performed in a 64 detector-row scanner (Aquillion 64; Toshiba Medical System, Otawara, Japan). All patients signed informed consent. Results of coronary CTA and myocardial CT perfusion (CTP) were blinded to the attending physician. All patients were instructed to have a diet free of caffeine 24 hours before examination. MDCT customized protocol included 2 contrast-enhanced scans. A calcium score was not performed. The first scan was a stress CTP only, with low resolution. The second scan was done for coronary
Figure 1 Flow diagram showing patient selection. CABG, coronary artery bypass graft.
arteries and rest myocardial perfusion evaluation (Table 1). Stress myocardial perfusion by MDCT image was obtained 2 minutes after intravenous infusion of dipyridamole, 0.56 mg/kg during 4 minutes. During dipyridamole infusion, heart rate, blood pressure, and patients’ symptoms were monitored. Helical acquisition for myocardial perfusion by MDCT was initiated with a real-time bolus tracking technique, at the time of peak left atria filling with contrast, which was determined visually (no specific threshold in Hounsfield units was used). Immediately after the stress scan, infusion with aminophylline (USP) 240 mg was started to revert vasodilatation induced by dipyridamole. Before a second perfusion/CT angiographic scan at rest, intravenous metoprolol was administered until a maximum dose of 20 mg to lower heart rate and improve image quality. Scanning at rest was performed with automatic trigger detection on the descending aorta set to signal density of 180 HU. The standard American College of Cardiology/American Heart Association (ACC/AHA) 17-segment model was used
Cury et al
Dipyridamole stress/rest myocardial TPR
Table 1 Customized protocol used in all patients undergoing stress myocardial CT perfusion First scan
Second scan
Data
Stress TPR
ECG gating Iodinated contrast, mL (mL/s) mA
Retrospective 60 (3)
Coronary CTA Rest TPR Retrospective 80–90 (5)
kV
120
100
Female, 240–270 Male, 370–400 120
ECG, electrocardiogram.
to identify perfusion defects. For comparison of perfusion data with coronary anatomical data derived from MDCT, we consolidated the segmental data into 3 territories according to the ACC/AHA recommendation (per-vessel analysis).13 Myocardial segments were correlated to the specific coronary territory depending on the coronary dominancy. Perfusion datasets were evaluated on a true short axis (apical, medium, and basal slices), with 2- and 4-chamber views, with 8-mm thick average multiplanar reformatted images. Thicker slices were used to improve low-contrast resolution for visual myocardium perfusion analysis, as described previously.10 Display parameters included a narrow window width and level (350 W and 150 L). Initial evaluation of perfusion defects started in the diastolic phase, and, for potential artifacts and motion, readers could use the systolic phase to confirm the perfusion defect. True perfusion defects were defined as subendocardial hypoenhancement, encompassing R25% of transmural extent, present in different phases of the cardiac cycle and within a specific coronary territory. Two independent observers with no knowledge of clinical data or other examinations performed quantitative analysis to measure TPR in all 17 ACC/AHA segment model, excluding the apex. The rest and stress TPR was analyzed with Osirix software (Pixmeo, Geneve, Switzerland). All perfusion images were evaluated in the short-axis view, divided circumferentially in 2 layers (manually) with same thickness to define subendocardium and subepicardium, as shown in Figure 2. For all TPR measurements, endocardial and epicardial contours were traced, excluding papillary muscles. After that, AHA myocardial segmentation was done, and mean myocardial density (Hounsfield unit; HU) was measured in both layers to calculate TPR in each segment (TPR 5 subendocardium mean density/ subepicardium mean density). Mean TPR was measured in each segment at rest and stress phase. The segmental TPR values were assigned to the respective coronary territories, considering the lowest segment TPR. Image quality for perfusion and anatomy evaluation was considered interpretable when in R1 phase of the
445 cardiac cycle; coronary and perfusion evaluations were diagnostic. QCA was performed on all patients submitted to coronary invasive angiography. Blinded readers considered coronary stenosis R 70% by QCA as a hemodynamic significant lesion. Accuracy of TPR in a patient-based analysis was considered for matching segments on SPECT.
Statistical analysis Statistical analyses were performed with Med-Calc version 11.4.4.0 (Meriakerke, Belgium). Data were expressed as mean 6 SD. Interobserver variability was compared with the use of Bland-Altman plots. The relation between percentage of luminal stenosis and TPR was compared with the use of Pearson’s correlation. The area under the curve was calculated and reported with 95% CIs. The threshold of significance was P , 0.05.
Results Twenty-six patients (61.6 6 8.0 years old; 14 males) were enrolled in the study and underwent MDCT (stress perfusion and coronary angiography) and invasive coronary angiography. All 26 patients completed the protocol with no adverse events, with mean CT radiation dose14 of 14.4 6 2.9 mSv and interpretable scans. Rest MDCT images were used to calculate TPR, and the normal segments distribution of myocardial TPR is shown in Figure 3 (n 5 416 myocardial segments). Mean rest TPR was 1.06 6 0.11. Abnormal TPR was considered 2 SDs below the mean (1.06 – 0.22 5 0.84), that is, ,0.85. Dipyridamole stress MDCT images were analyzed, and the TPR was quantified in 416 myocardial segments and
Figure 2 Myocardial segmentation in the short axis of the left ventricle used for TPR measurement.
446
Journal of Cardiovascular Computed Tomography, Vol 5, No 6, November/December 2011
Figure 3 Range distribution of TPR measures on rest images (n 5 416 myocardial segments).
consolidated in 78 territories, which were correlated with the degree of coronary stenosis measured by the QCA. For 6 patients with normal coronary arteries by invasive coronary angiography, 96 myocardial segments were consolidated in 18 territories to calculate each TPR. After calculating the TPR, these territories were divided in 2 groups: one with SPECT perfusion defect and another without. The mean TPR of territories with a previous positive perfusion defect in SPECT was 1.02 6 0.18 (95% CI, 0.86–1.18; n 5 6), and mean TPR of territories without perfusion defect in SPECT was 1.03 6 0.09 (95% CI, 0.95–1.11; n 5 12). No significant statistical difference was observed (P 5 0.83).
The territories TPR analysis for all patients (n 5 78) during stress showed a mean TPR of 0.71 6 0.13 (95% CI, 20.64 to 0.77; n 5 17) in territories corresponding to coronary stenosis R 70% by QCA and positive SPECT (Fig. 4), and 1.01 6 0.09 (95% CI, 20.96 to 1.06; n 5 61) in remote territories corresponding to coronary stenosis , 70% (P , 0.001; Fig. 5). A significant inverse linear correlation was observed between the TPR and the percentage f diameter stenosis (r 5 20.74, P , 0.001; Fig. 6). The sensitivity, specificity, positive predictive value, negative predictive value, and area under the curve for TPR detecting stenosis corresponding to a perfusion defect on SPECT in a patient-based analysis (n 5 26) was 94% (16 of 17; 95% CI, 71.3%–99.9%), 78% (7 of 9; 95% CI, 40.0%–97.2%), 89% (16 of 18; 95% CI, 64.4%–98.7%), 87.5% (7 of 8; 95% CI, 47.3%–99.7%) and 0.85 (95% CI, 0.66–0.96, respectively). Bland Altman plot for interobserver variability in TPR quantification (Fig. 7) showed a mean difference of 20.01 (95% CI, 20.20 to 0.18). Interobserver correlation was good (k 5 0.72; 95% CI, 0.67–0.77).
Discussion Although SPECT myocardial perfusion imaging represents an established technique, limitations include its inability to evaluate transmural myocardial perfusion and
Figure 4 Correlation among TPR by MDCT (A), SPECT (B), coronary angiography by MDCT (C), and invasive coronary angiography (D), mid inferolateral TPR 5 0.78 and left circumflex artery coronary stenosis .70%.
Cury et al
Dipyridamole stress/rest myocardial TPR
Figure 5 Mean TPR analysis comparing territories without significant coronary stenosis (QCA , 70%) with territories with positive SPECT and corresponding coronary stenosis R 70% by QCA.
to accurately quantify the extent of subendocardial perfusion defects because of limited spatial resolution. Other than myocardial perfusion abnormalities, SPECT may also show functional findings that suggest ischemia, such as left ventricular ejection fraction decreases by .5% after stress, pulmonary uptake, and transitory dilatation of the left ventricle.15–17 Blood flow, metabolism, and the contraction and relaxation dynamics of the myocardium are heterogeneous. The subendocardium is more susceptible to ischemic injury than the subepicardium. Myocardial ischemia starts in the subendocardium and spreads like a wavefront toward the subepicardium. Because perfusion initially decreases in the subendocardium as coronary flow is gradually reduced,
447 first-pass imaging with the high spatial and temporal resolution prerequisite allows early detection of coronary stenosis, as indicated by previous magnetic resonance imaging data.18 Cardiac CT can determine the TPR, which is inversely correlated with the degree of coronary stenosis. A recent study measured TPR with the use of automatic delineation of myocardium into 3 layers: subendocardium, mesocardium, and subepicardium.11 The TPR cutoff used was 1 SD below the mean (TPR , 0.99), and the TPRs observed were 1.12 6 0.13 for patients with no obstructive CAD and 0.91 6 0.10 for patients with coronary stenosis . 70%. In that study, a 256-row scanner was used for rest and stress TPR evaluation. In this regard, our results in a 64-row scanner agreed with this previously published study. Although we measured TPR through manual measurement of the myocardium in 2 layers, TPR was inversely correlated with the severity of stenosis. Dividing the myocardial wall in 3 layers would potentially increase the low attenuation of the subendocardial layer and might increase sensitivity. The present study used a lower cutoff for describing a normal TPR (2 SDs below the mean; TPR , 0.85), although we found TPR values in patients with and without obstructive CAD (0.71 6 0.13 vs 1.01 6 0.09, respectively) comparable to that previously described.11 It is worth noting that in our analysis we also used a combined reference method for defining the true ischemic myocardial segment, that is, a myocardial segment with perfusion defect on SPECT combined with a corresponding coronary stenosis . 70% by QCA. In addition, the normal values for TPR observed in patients without CAD were independent of SPECT results, suggesting that TPR was a valuable tool in identifying patients with false-positive SPECT.
Figure 6 Correlation between TPR and percentage of diameter stenosis by QCA in which territories were matched to SPECT territories showing perfusion abnormalities (n 5 26).
448
Journal of Cardiovascular Computed Tomography, Vol 5, No 6, November/December 2011
Figure 7 Bland Altman plot showing the agreement in TPR quantification of 416 myocardial segments on the stress CTP, between observer 1 and observer 2.
Conclusions In conclusion, TPR has a good correlation with SPECT and coronary invasive angiography to detect significant coronary stenosis.
References 1. Leber AW, Knez A, von Ziegler F, Becker A, Nikolau K, Paul S, Wintersperger B, Reiser M, Becker CR, Steinbeck G, Boekstegers P: 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;46:147–54. 2. Leschka S, Alkadhi H, Plass A, Desbiolles L, Grunenfelder J, Marincek B, Wildermuth S: Accuracy of MSCT coronary angiography with 64-slice technology: first experience. Eur Heart J. 2005;26:1482–7. 3. Mollet NR, Cademartiri F, van Mieghem CA, Runza G, McFadden EP, Baks T, Serruys PW, Krestin GP, de Feyter PJ: High-resolution spiral computed tomography coronary angiography in patients referred for diagnostic conventional coronary angiography. Circulation. 2005;112:2318–23. 4. 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;46:552–7. 5. Pugliese F, Mollet NR, Runza G, van Mieghem C, Meijboom WB, Malagutti P, Baks T, Krestin GP, de Feyter PJ, Cademartiri F: Diagnostic accuracy of non-invasive 64-slice CT coronary angiography in patients with stable angina pectoris. Eur Radiol. 2006;16:575–82. 6. Ropers D, Rixe J, Anders K, Kuttner A, Baum U, Bautz W, Daniel WG, Achenbach S: Usefulness of multidetector row spiral computed tomography with 64- x 0.6-mm collimation and 330-ms rotation for the noninvasive detection of significant coronary artery stenoses. Am J Cardiol. 2006;97:343–58.
7. Miller JM, Rochitte CE, Dewey M, Arbab-Zadeh A, Niinuma H, Gottlieb I, Paul N, Clouse ME, Shapiro EP, Hoe J, Lardo AC, Bush DE, de Roos A, Cox C, Brinker J, Lima JA: Diagnostic performance of coronary angiography by 64-row CT. N Engl J Med. 2008; 359:2324–36. 8. Miller JM, Dewey M, Vavere AL, Rochitte CE, Niinuma H, Arbab-Zadeh A, Paul N, Hoe J, de Roos A, Yoshioka K, Lemos PA, Bush DE, Lardo AC, Texter J, Brinker J, Cox C, Cloude ME, Lima JA: Coronary CT angiography using 64 detector rows: methods and design of the multi-centre trial CORE-64. Eur Radiol. 2009;19: 816–28. 9. Cury RC, Magalh~aes TA, Borges AX, Shiozaki AA, Lemos PA, Junior JS, Meneghetti JC, Cury RC, Rochitte CE: Dipyridamole stress and rest myocardial perfusion by 64-detector row computed tomography in patients with suspected coronary artery disease. Am J Cardiol. 2010;106:310–5. 10. Blankstein R, Shturman LD, Rogers IS, Rocha-Filho JA, Okada DR, Sarwar A, Soni AV, Bezerra H, Ghoshhajra BB, Petranovic M, Loureiro R, Feuchtner G, Gewirtz H, Hoffmann U, Mamuya WS, Brady TJ, Cury RC: Adenosine-induced stress myocardial perfusion imaging using dual-source cardiac computed tomography. J Am Coll Cardiol. 2009;54:1072–84. 11. George RT, Arbab-Zadeh A, Miller JM, Kitagawa K, Chang HJ, Bluemke DA, Becker L, Yousuf O, Texter J, Lardo AC, Lima JA: Adenosine stress 64- and 256-row detector computed tomography angiography and perfusion imaging: a pilot study evaluating the transmural extent of perfusion abnormalities to predict atherosclerosis causing myocardial ischemia. Circ Cardiovasc Imaging. 2009;2: 174–82. 12. Ho KT, Chua KC, Klots E, Pankin C: Stress and rest dynamic myocardial perfusion imaging by evaluation of complete time-attenuation curves with dual-source CT. J Am Coll Cardiol. 2010;3:811–20. 13. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, Pennell DJ, Rumberger JA, Ryan T, Verani MS: 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. Circulation. 2002;105:539–42. 14. Shrimpton PC. Assessment of patient dose in CT. Chilton, NRPB-PE/1/2004; 2004. Also published as Appendix C of the 2004 CT Quality Criteria. Available at: http://www.msct.info/CT_Quality_ Criteria.htm. Accessed May 5, 2006. 15. Hida S, Chikamori T, Tanaka H, Usui Y, Igarashi Y, Nagao T, Yamashina A: Diagnostic value of left ventricular function after stress and at rest in the detection of multivessel coronary artery disease as assessed by electrocardiogram-gated SPECT. J Nucl Cardiol. 2007; 14:68–74. 16. Shirai N, Yamagishi H, Yoshiyama M, Teragaki M, Akioka K, Takeuchi K, Yoshikawa J, Ochi H: Incremental value of assessment of regional wall motion for detection of multivessel coronary artery disease in exercise (201) TI gated myocardial perfusion imaging. J Nucl Med. 2002;43:443–50. 17. Desai D, Kozeski G, Akinboboye O: Detection of multivessel coronary artery disease: looking beyond the extent of perfusion abnormalities. J Nucl Cardiol. 2009;16:4–5. 18. Jerosch-Herold M, Wilke N: MR first pass imaging: quantitative assessment of transmural perfusion and collateral flow. Int J Card Imaging. 1997;13:205–18.