Effects of Sublingual Nitroglycerin on Coronary Lumen Diameter and Number of Visualized Septal Branches on 64-MDCT Angiography

Transcription

1 Cardiac Imaging Original Research Decramer et al. Effects of Nitroglycerin on Coronary Arteries at MDCTA Cardiac Imaging Original Research Isabel Decramer 1,2 Piet K. Vanhoenacker 1 Giovanna Sarno 2 Lieven Van Hoe 1 Olivier Bladt 1 William Wijns 2 Paul M. Parizel 3 Decramer I, Vanhoenacker PK, Sarno G, et al. Keywords: coronary angiography, coronary arteries, coronary vessels, MDCT coronary angiography, nitroglycerin DOI:1214/AJR Received May 30, 2007; accepted after revision July 16, See Editor's comment on page 224 regarding drug safety. 1 Department of Radiology and Imaging, OLV Ziekenhuis, Moorselbaan 164, Aalst, Oost Vlaanderen 9300, Belgium. Address correspondence to P. K. Vanhoenacker (piet@vanhoenacker.be). 2 Cardiovascular Center Aalst, Aalst, Belgium. 3 Department of Radiology, University Hospital Antwerp, Edegem, Belgium. AJR 2008; 190: X/08/ American Roentgen Ray Society Effects of Sublingual Nitroglycerin on Coronary Lumen Diameter and Number of Visualized Septal Branches on 64-MDCT Angiography objective. This article describes the influence of sublingual nitroglycerin spray on the lumen diameter, number of side branches visualized, contrast-to-noise ratio (CNR), and signal-to-noise ratio (SNR) of the coronary arteries with MDCT angiography. Subjects and methods. Forty-two patients were prospectively included in this study: 21 were examined without sublingual nitroglycerin (group A), and 21 were examined after the administration of sublingual nitroglycerin (group B). CT angiography was performed using a 64-MDCT scanner. Two blinded observers quantitatively assessed lumen diameter and volume in the left anterior descending artery () and the right coronary artery (RCA). The number of septal branches was counted. The SNR and CNR in the and RCA were calculated in both groups. The number of clinical side effects was evaluated. results. The lumen diameters and the average volumes were significantly larger in group B than in group A. The number of septal branches visualized in group B was significantly higher than in group A. No statistically significant difference in SNR and CNR between the groups was shown. The number of side effects in the two groups was not significantly different. Conclusion. Sublingual nitroglycerin spray significantly dilates the coronary arteries and allows more septal branches to be visualized at coronary CT angiography without diminishing image quality or increasing the number of side effects. O rganic nitrates are potent vasodilators that dilate both normal and abnormal coronary arteries by relaxing vascular smooth muscle [1, 2]. The administration of nitroglycerin has long been used to obtain maximal dilatation of the epicardial coronary arteries during conventional coronary angiography [3]. The current generation of MDCT systems permits cross-sectional and longitudinal imaging of the coronary arteries with a high spatial and temporal resolution. However, small vessels and side branches and the distal portions of the main coronary arteries are not always optimally visualized [4, 5]. The investigators of several studies on MDCT angiography (MDCTA) of the coronary arteries report the use of nitroglycerin to dilate the lumen of the coronary arteries, and it is a widely accepted practice [6, 7]. However, the effects of nitroglycerin on coronary lumen diameter and volume and on image quality at CT angiography (CTA) have not been extensively documented or quantified. Coronary volume and lumen diameter are thought to be significantly larger in patients who receive nitroglycerin than in patients who do not. We assessed the influence of sublingual nitroglycerin spray on lumen diameter, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and the number of visualized side branches on CTA. In addition, we assessed the number of patients with side effects after the administration of sublingual nitroglycerin. Subjects and Methods Patient Selection and Study Population To avoid scanning patients twice, a prospective study design was followed. In the time period from May through June 2006, all patients scheduled to undergo CTA of the coronary arteries were screened for inclusion in the study. Because this study was focusing on feasibility, determination of lumen size, potential artifacts induced by the administration of nitroglycerin, and the potential occurrence of clinical side effects, some exclusion criteria were used. The exclusion criteria were as follows: previous stent implantation or bypass surgery, age > 80 years, coronary artery calcium (CAC) score of > 400, heart rate > 70 beats per minute (bpm), and body mass index > 35 kg/m 2. AJR:190, January

2 Decramer et al. The eligible patients were enrolled and randomized in two arms after determination of CAC score and the other exclusion criteria. The first group of patients (group A) was the control group and did not receive nitroglycerin. The patients in the second group were given sublingual nitroglycerin (group B). The observers were blinded to the group from which each patient originated. Except for the addition of nitroglycerin, the scanning protocol of the two groups was similar. Our institutional review committee approved this study. Informed consent was obtained from all patients. MDCTA Image Acquisition After determination of the CAC score ( collimation, 330-millisecond gantry rotation time, pitch, 3-mm slice thickness, 120-kV tube voltage, 145-mAs tube current), CTA was performed using an MDCT scanner (Somatom 64, Siemens Medical Solutions) with a collimation and 330-millisecond gantry rotation time, 3.8-mm per rotation table feed, -mm reconstruction increment, 0.75-mm slice thickness, 120-kV tube voltage, and 750-mAs tube current. A bolus (average, 90 ml; 5 ml/s) of contrast material (iomeprol, mg/ml [Iomeron 400, Bracco Altana Pharma]) was administered before examination followed by a 50-mL saline flush in all patients. Patients with a baseline heart rate of > 65 bpm were given a b-blocker ( mg of metoprolol, 1 mg/ml) orally before examination and this dose was supplemented by IV administration as required. In group B, one dose of nitroglycerin was administered as a sublingual spray ( mg/l) 5 minutes before initiation of scanning. Analysis of MDCT Angiograms MDCT data were transferred from the scanner to an offline workstation (Aquarius, TeraRecon) for postprocessing and analysis. Visibility of side branches Each patient was evaluated by two observers with experience in the interpretation of more than 1,000 coronary scans, and a consensus decision was made. In each patient, the number of septal branches originating from the proximal and mid left anterior descending artery (), segments 6 and 7 according to the American Heart Association (AHA) [8], was counted. The sinoatrial branch and the conus branch, which are two side branches originating from the proximal part of the right coronary artery (RCA), were assessed and denoted as present or absent. Although the presence or absence of septal branches, the sinoatrial branch, and the conus branch is probably not clinically significant, we wanted to determine whether nitroglycerin improves visualization of these small vessels as a surrogate marker for improved vessel visualization. To obtain the exact number of septal branches, counting was done in the axial plane on source images, on a curved multiplanar reconstruction (MPR), and on different maximal intensity projections (MIPs). The sinoatrial branch and conus branch were assessed in the axial plane on source images, when necessary, and MIP images were reconstructed to confirm their presence or absence. Image quality: SNR and CNR All patients were evaluated by one observer. The SNR was measured in the left ventricular myocardium to assess the potential influence of nitroglycerin on the enhancement of myocardium in groups A and B. The SNR of the myocardium was assessed on a short-axis view in the lateral, septal, anterior, and posterior walls with a fixed region of interest. The SNR was defined as the ratio of attenuation in myocardium and image noise. The SNR and CNR were measured in the left main coronary artery, proximal RCA, crux (the bifurcation of the RCA in the right posterior descending artery and right posterolateral artery), right posterior descending artery, distal part of the circumflex artery, distal part of the, and distal part of the third diagonal branch (whenever visualized) in groups A and B. The SNR was defined as the ratio of attenuation in Hounsfield units (H) in the coronary lumen to image noise. Image noise was defined as the SD of the attenuation (H) in the coronary artery. The CNR was defined as the difference between the attenuation in the coronary artery lumen and the attenuation in the fat surrounding the coronary artery divided by image noise. Volume determination and diameter of vessels The same two independent observers evaluated all patients. Vessel volume and average vessel cross-sectional diameter were evaluated in the and RCA. Measurements of absolute coronary artery diameters were performed in a plane perpendicular to a curved MPR reconstruction along the long axis of the vessel. A 2D profile measuring attenuation along a line crossing through the lumen was made (Fig. 1). The pixels with attenuation values higher than the average attenuation of contrast material in opacified lumen minus 3 SD were supposed to be part of the coronary lumen. We hypothesized that by including values minus 3 SD, 99% of the possible values would be included [9]. The lumen diameter obtained in this way was used for calculation. The same measurement was repeated along a line perpendicular to the first measurement, and the average of the two values was calculated as the final cross-sectional diameter. The measurements were obtained halfway in the proximal RCA (segment 1, AHA) and halfway in the proximal (segment 6, AHA). The range of volume measured in the and RCA was from the ostium of the left coronary artery to the distal and from the ostium of the RCA to the distal right posterior descending artery. Similar to the method used in the diameter measurements, the lower boundaries for luminal attenuation were set at the mean attenuation of the ostium minus 3 SD, and the upper boundary was set as high as possible so that calcium was not excluded. A dedicated segmentation software algorithm calculated the volumes (Fig. 2). Artifacts Common artifacts, such as stairstep artifacts, blur artifacts, streak artifacts, and slab artifacts, were registered for both groups. Substantial artifacts were registered on a per-patient basis and were denoted as absent or present. Two observers reached a consensus decision. Side effects and clinical parameters The occurrence of three potential well-known side effects of nitrates headache, reflex tachycardia, and drop in blood pressure were registered by the technicians. In all patients, heart rate was measured during scanning and blood pressure was measured before and after scanning. The number of patients who had significant stenosis (> 50% lumen diameter) was registered. Statistical Analysis Continuous data are reported as means with 95% CIs. The sample size was set at 21 patients in each group to reach statistical significance. The unpaired two-sided Student s t test was used for comparison of SNR, CNR, diameter measurements, and volume measurements between group A and group B. The number and proportion of side effects, the presence of artifacts, the patient s heart rate during scanning, the number of side branches in groups A and B, and the prevalence of significant disease were compared using the Fisher s exact test, unpaired two-sided Student s t test, or both. Interobserver agreement between measurements of volume and diameter were analyzed with the concordance coefficient correlation [10 12]. The concordance correlation coefficient, r c, contains a measurement of precision, r, and a measurement of accuracy, C b : r c = r C b, where r is the Pearson s correlation coefficient, which measures how far each observation deviates from the best-fit line, and is a measure of precision; and C b is a bias correction factor, which measures how far the best-fit line deviates from the 45 line through the origin, and is a measure of accuracy. Interobserver variability between measurements of volume in the RCA and were also analyzed with Bland-Altman plots. Agreement between observers for the presence of the sinoatrial branch and conus branch, the number of septal branches, and the presence of artifacts was analyzed with Cohen s kappa. 220 AJR:190, January 2008

3 Effects of Nitroglycerin on Coronary Arteries at MDCTA Fig. 1 Curved multiplanar reconstruction (MPR) along long axis of vessel (left), and two perpendicular 2D profiles (right) at location indicated on curved MPR (TeraRecon Aquarius workstation). Profile is line crossing lumen, resulting in graph representing attenuation (in Hounsfield units) along that line. Pixels with attenuation values higher than average attenuation of contrast material in opacified lumen minus 3 SD are supposed to be part of coronary lumen. A p value of > 5 was considered statistically nonsignificant. MedCalc software (version 7, MedCalc) was used. Results Study Population Of 96 patients who underwent CTA of the coronary arteries in the time period mentioned, 63 patients were eligible according to the exclusion criteria. Forty-two patients (27 men, 15 women; age range, years; mean age, 56.1 TABLE 1: Patient Characteristics Patient Characteristic years) were finally enrolled. Twenty-one of the 63 eligible subjects would not provide consent to participate in the study, leaving a final subject group of 42 patients. Group A consisted of 21 patients, and group B consisted of 21 patients. Patient characteristics are shown in Table 1. In all patients, there was a right dominant anatomy, meaning that the posterior descending artery originated from the RCA and was therefore labeled as the distal right posterior descending artery. Group A (n = 21) Group B (n = 21) 95% CI 95% CI Age (y) a Sex (no. [%] of men) 12 (57) 15 (71) 5 b Blood pressure (mm Hg) 137/ / / /63 77 a Heart rate (beats per minute) a Prevalence (no. [%]) 4 (19) 5 (24) 0.71 b Coronary artery calcification (Agatston) score a Weight (kg) a Height (cm) a Body mass index (kg/m²) a a Student s t test. b Chi-square test. Fig. 2 Curved multiplanar reconstruction along long axis of vessel shows how software on TeraRecon Aquarius workstation automatically calculates volume inside vessel wall corresponding to attenuation between minimal and maximal values in Hounsfield units. p Image Analysis Visibility of side branches There was a statistically significant difference in the number of visible septal branches between groups A and B (Student s t test, p < 001) (Fig. 3). The interobserver agreement for the number of septal branches was excellent to good (Cohen s k = 0.82). Table 2 illustrates the number of patients with a discernible sinoatrial branch and conus branch in groups A and B. The sinoatrial branch was present in 67% of group A compared with No. of Septal Branches Group A Group B Fig. 3 Box-and-whisker diagram shows number of septal branches in groups A and B. AJR:190, January

4 Decramer et al. 95% of group B (Fisher s exact test, p = 69), and the conus branch was present in 33% of group A compared with 62% of group B (Fisher s exact test, p = 0.129). The difference was not statistically significant. The interobserver variability for detecting the sinoatrial branch and conus branch in groups A and B was good (Cohen s k = 0.70 and 0.77, respectively). SNR and CNR There was no significant difference in SNR and CNR between group TABLE 2: Number of Patients with Visible Sinoatrial Branch and Conus Branch in Groups A and B Branch Group A Group B p Sinoatrial Conus SNR LM RCA Crux RPDA CX D3 Fig. 4 Bar graph with 95% CIs shows signal-to-noise ratio (SNR) in groups A (white bars) and B (gray bars) for all segments evaluated: LM = left main artery, RCA = right coronary artery, RPDA = right posterior descending artery, CX = circumflex artery, = left anterior descending artery, D3 = distal part of third diagonal branch. Diameter (mm) CNR LM A and group B (Figs. 4 and 5) for all the vessel segments evaluated (p > 5 in all pairwise comparisons). In addition, there was no difference in the SNR of myocardium between group A and group B (p > 5). Volume determination and diameter of vessels The differences in lumen diameter between group A and group B were statistically significant. The average lumen diameter of the RCA for patients in group A and group B was 2.63 mm (95% CI, mm) and 3.48 mm ( mm) (p < 001), respectively, whereas the average diameter in the in group A and group B was 2.67 mm ( mm) and 3.42 mm ( mm) (p < 001), respectively (Fig. 6). The differences in volume of the lumen between group A and group B were also statistically significant. The average volume of the RCA for patients in group A and group B, respectively, was 8 cm 3 (95% CI, cm 3 ) and 5 cm 3 ( cm 3 ) (p < 001), whereas the average volume of the for patients in group A and group B was 0.35 cm 3 (8 2 cm 3 ) and 0.72 cm 3 ( cm 3 ) (p = 001) (Fig. 7). The mean difference in RCA volume and volume between group A and group B was, respectively, 8 cm 3 ( cm 3 ) and 0.37 cm 3 ( cm 3 ). Bland-Altman plots for agreement of volume measurements are shown in Figure 8. The agreement for measurements of lumen diameters and volume made by the two observers was good to excellent for both groups A and B. The concordance correlation coefficients, Pearson s correlation coefficients (r RCA Crux RPDA CX D3 Fig. 5 Bar graph with 95% CIs shows contrast-to-noise ratio (CNR) in groups A (white bars) and B (gray bars) for all segments evaluated: LM = left main artery, RCA = right coronary artery, RPDA = right posterior descending artery, CX = circumflex artery, = left anterior descending artery, D3 = distal part of third diagonal branch. Volume (cm 3 ) RCA RCA Fig. 6 Bar graph with 95% CIs of lumen diameter in right coronary artery (RCA) and left anterior descending artery () in groups A (white bars) and B (gray bars) for observer 1. Fig. 7 Bar graph with 95% CIs shows volume in right coronary artery (RCA) and left anterior descending artery () in groups A (white bars) and B (gray bars) for observer AJR:190, January 2008

5 Effects of Nitroglycerin on Coronary Arteries at MDCTA for precision), and bias correction factors (C b for accuracy) are displayed in Table 3. Artifacts Table 4 reveals that there was no difference between the groups in the number of scans with substantial artifacts (p > 0.99). Interobserver agreement was excellent (Cohen s k = 0.98). Side effects and clinical parameters Table 4 reveals that there was no statistically significant difference in the number of side effects (p > 0.99), prevalence of significant disease (p = 0.71), and heart rate (p = 3) between groups A and B. The mean blood pressure was lower in group B than in group A (mean blood pressure, 120 over 70 vs 137 over 76 mm Hg, respectively) after scanning (see Table 1 for detailed results). However, this difference was not statistically or clinically significant, suggesting that nitroglycerin lowered blood pressure only slightly. Discussion The results of this study show that more septal branches could be visualized when patients received sublingual nitroglycerin spray before scanning. The administration of sublingual nitroglycerin spray also led to a significant increase in volume in both the RCA and the. Also, the diameters in those arteries were significantly larger and within the range of normal dimensions of the coronary arteries, as seen on conventional angiography [13], when nitroglycerin was applied. The effect is important and reached statistical significance even with a rather small sample, despite the fact that important Fig. 8 Bland-Altman plots show interobserver agreement of volume measurements. A and B, Plots show volume data for left anterior descending artery () (A) and right coronary artery (RCA) (B) of group A. C and D, Plots show volume data for (C) and RCA (D) of group B. A C variability exists between patients concerning the size and anatomy of the coronary tree [14]. The SNR and CNR and the number of artifacts were similar in groups A and B; the use of nitroglycerin did not cause more clinical side effects in group B than in group A. Nitrates are widely used to treat angina and alleviate symptoms of ischemia through a variety of mechanisms [15]. Among their other effects, nitrates increase the coronary blood flow in the epicardial coronary arteries by vasodilation [1, 2]. They have been used widely in invasive coronary angiography to optimize the enhancement and size of the lumen and visualization of small branches and are administrated through an IV or intraarterial route during the procedure [3]. Nitrates are known to have an important effect on the diameter of B D AJR:190, January

6 Decramer et al. TABLE 3: Agreement Between Measurements of Lumen Diameters and Volume of Left Anterior Descending Artery () and Right Coronary Artery (RCA) by Two Observers Diameter Volume Group Concordance Coefficient (r c ) Pearson s Coefficient (r) Bias Correction Factor (C b ) Concordance Coefficient (r c ) Pearson s Coefficient (r) Bias Correction Factor (C b ) Group A : observer 1 vs RCA: observer 1 vs Group B : observer 1 vs RCA: observer 1 vs TABLE 4: Number of Side Effects, Artifacts, and Extra Systoles in Groups A and B Characteristic Group A Group B p a No. of side effects 0 1 > 0.99 No. of artifacts 1 0 > 0.99 No. of extra systoles a Fisher s exact test. the coronary arteries, especially on the size of nonstenotic segments, and to have less effect on stenotic segments; thus, nitrates can enhance the detection of obstructive lesions [3, 16]. Nitroglycerin has been used in the assessment of the microcirculation of myocardium on CT and in imaging the coronary tree with MRI, but the quantification of its effects on the size of the epicardial arteries and the clinical side effects associated with its use were not fully explored [17, 18]. In imaging the coronary tree with CTA, tiny branches (i.e., < 1 2 mm) were usually excluded from analysis, at least in earlier studies [4, 5, 19 21]. Knowing that the overall size of the coronary arteries is small on CTA, every attempt to enhance size and increase enhancement should be made [5]. Sublingual nitroglycerin has been used in one study [22], which showed results comparable to ours. However, the way the drug was administered in that study was different and the number of patients was rather small. In addition, only diameters were measured; neither image quality nor side effects were evaluated. The use of sublingual nitroglycerin spray instead of another nitrate preparation is based on literature stating that sublingual nitroglycerin spray may be more efficacious than other nitrates and may cause fewer side effects [23], which are often seen with the other ways of administration [24]. This was also suggested by our experience because we did not see more clinical side effects, such as tachycardia, flushing, or headache, in group B than in group A. A few limitations of this study must be acknowledged. We did not investigate the influence of sublingual nitroglycerin spray on diagnostic accuracy for the detection of significant lesions (stenosis > 50% lumen diameter) or calcified lesions because this study explored only the capacity of sublingual nitroglycerin spray to enlarge lumen size. The choice not to investigate diagnostic accuracy was made on the basis of the fact that the sample size for such a study, with the population that was included (i.e., low to intermediate pretest probability), would be very high and make the study difficult. Also, this increased sample size might have resulted in selection bias because in our institution only patients who have positive or equivocal scanning results undergo conventional angiography. A selected population would therefore have been investigated, and the influence of sublingual nitroglycerin spray on coronary lumen would have been investigated only in patients with diseased vessels. Another limitation is that we used two different patient samples with our study design; therefore, selection bias could have occurred. We think, however, that scanning patients twice for this observational study was not feasible for ethical reasons. This limitation was counterbalanced to some extent by the randomization procedure. The conclusion of this study is that routine administration of sublingual nitroglycerin spray should be recommended. However, a largescale, randomized study with conventional angiographic correlation should be performed to investigate the influence of sublingual nitroglycerin spray on diagnostic accuracy. Editor's Comment Prior to the administration of nitroglycerin, patients should be questioned regarding the recent use of Viagra (sildenafil), Cialis (tadalafil), or Levitra (vardenafil). Severe side effects, including very low blood pressure and death, have been reported in patients given nitroglycerin within 24 hours after use of these phosphodiesterase type 5 inhibitors. References 1. Brown BG, Bolson E, Petersen RB, Pierce CD, Dodge HT. The mechanisms of nitroglycerin action: stenosis vasodilatation as a major component of the drug response. Circulation 1981; 64: Abrams J. Hemodynamic effects of nitroglycerin and long-acting nitrates. Am Heart J 1985; 110: Feldman RL, Marx JD, Pepine CJ, Conti CR. Analysis of coronary responses to various doses of intracoronary nitroglycerin. Circulation 1982; 66: Hamon M, Biondi-Zoccai GG, Malagutti P, et al. Diagnostic performance of multislice spiral computed tomography of coronary arteries as compared with conventional invasive coronary angiography: a meta-analysis. J Am Coll Cardiol 2006; 48: Schoepf UJ, Becker CR, Ohnesorge BM, Yucel EK. CT of coronary artery disease. Radiology 2004; 232: Gerber TC, Breen JF, Kuzo RS, et al. Computed tomographic angiography of the coronary arteries: techniques and applications. Semin Ultrasound CT MR 2006; 27: Vanovermeire OM, Duerinckx AJ. A practical approach to 64-slice coronary CT angiography. JBR-BTR 2006; 89: Dodge JT, Brown BG, Bolson EL, Dodge HT. In- 224 AJR:190, January 2008

7 Effects of Nitroglycerin on Coronary Arteries at MDCTA trathoracic spatial localization of specified coro- 14. Lemos PA, Ribeiro E, Perin MA, et al. Angio- 20. Kaiser C, Bremerich J, Haller S, et al. Limited di- nary segments on the normal human heart: appli- graphic segment size in patients referred for coro- agnostic yield of non-invasive coronary angiogra- cations in quantitative arteriography, assessment nary intervention is influenced by constitutional, phy by 16-slice multi-detector spiral computed of regional risk and contraction, and anatomic dis- anatomical, and clinical features. Int J Cardio- tomography in routine patients referred for evalu- play. Circulation 1988; 78: vasc Imaging 2007; 23:1 7 ation of coronary artery disease. Eur Heart J 9. Pagano M, Gauvreau K, eds. Theoretical probability 15. Parker JD, Parker JO. Nitrate therapy for stable 2005; 26: distributions. Principles of biostatistics, 2nd ed. Pacific Grove, CA: Duxbury, 2000: Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics 1989; 45: King TS, Chinchilli VM. Robust estimators of the concordance correlation coefficient. J Biopharm Stat 2001; 11: Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res 1999; 8: Goar FG, Pinto FJ, Alderman EL, et al. Intravascular ultrasound imaging of angiographically normal coronary arteries: an in vivo comparison with quantitative angiography. J Am Coll Cardiol 1991; 18: angina pectoris. N Engl J Med 1998; 338: Schnaar RL, Sparks HV. Response of large and small coronary arteries to nitroglycerin, NaNO 2, and adenosine. Am J Physiol 1972; 223: Hackenbroch M, Meyer C, Schmiedel A, et al. The value of applying nitroglycerin in 3D coronary MR angiography with real-time navigation technique [in German]. Rofo 2004; 176: Lerman LO, Siripornpitak S, Maffei NL, Sheedy PF, Ritman EL. Measurement of in vivo myocardial microcirculatory function with electron beam CT. J Comput Assist Tomogr 1999; 23: Achenbach S. Computed tomography coronary angiography. J Am Coll Cardiol 2006; 48: Schuijf JD, Bax JJ, Shaw LJ, et al. Meta-analysis of comparative diagnostic performance of magnetic resonance imaging and multislice computed tomography for noninvasive coronary angiography. Am Heart J 2006; 151: Dewey M, Hoffmann H, Hamm B. Multislice CT coronary angiography: effect of sublingual nitroglycerin on the diameter of coronary arteries. Rofo 2006; 178: Pfister M, Seiler C, Fleisch M, Gšbel H, Lüscher T, Meier B. Nitrate induced coronary vasodilatation: differential effects of sublingual application by capsule or spray. Heart 1998; 80: Thadani U, Rodgers T. Side effects of using nitrates to treat angina. Expert Opin Drug Saf 2006; 5: AJR:190, January