| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Clinical Radiology, University of Muenster,
Albert-Schweitzer-Strasse 33, 8149 Muenster, Germany.
2 Department of Cardiology, University of Muenster, Muenster, Germany.
Received June 11, 2007;
accepted after revision September 6, 2007.
Address correspondence to H. Seifarth.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. In 90 patients, coronary dual-source CT angiography was performed without ß-blocking agents. Data were reconstructed in 5% steps throughout the R-R interval. Two independent readers selected optimal systolic and diastolic reconstruction windows for each major coronary vessel—the right coronary artery (RCA), left anterior descending artery (LAD), and left circumflex artery (LCX)—using a 3D viewer and volume-rendering displays. The motion score for each vessel was graded from 1 (no motion artifacts) to 5 (severe motion artifacts over entire vessel).
RESULTS. The average heart rate of all patients was 68.7 beats per minute (bpm) (range, 43–119 bpm). The median optimal systolic reconstruction windows were at 35%, 30%, and 35% for the RCA, LAD, and LCX, respectively. The median optimal diastolic reconstruction window was at 75% for all vessels. The mean motion scores (± SD) in the systolic reconstructions were 1.9 ± 0.8 (RCA), 1.7 ± 0.5 (LAD), and 2.0 ± 0.6 (LCX). The mean motion scores for the diastolic reconstructions were 1.7 ± 0.9, 1.5 ± 0.6, and 1.6 ± 0.7, respectively. In patients with a heart rate of < 70 bpm, motion scores were significantly lower in diastole versus systole (1.3 ± 0.4 and 1.9 ± 0.5, respectively; p < 0.01). In most patients with a heart rate of > 80 bpm, motion scores were lower in systolic than in diastolic reconstructions (2.1 ± 0.6 and 2.6 ± 0.8, respectively; p < 0.05).
CONCLUSION. Using dual-source CT, the overall optimal reconstruction window is at 75% of the R-R interval in patients with low or intermediate heart rates. In patients with heart rates of > 80 bpm, systolic reconstructions often yield superior image quality compared with diastolic reconstructions.
Keywords: coronary artery stenosis • coronary arteries • CT angiography • dual-source CT • heart disease • motion artifacts
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The recently introduced dual-source CT scanner is fitted with two sets of image acquisition systems (tube and detector) arranged at an angle of 90°. Thus, a 90° rotation of the gantry is sufficient to acquire the 180° projection data needed for image reconstruction if data from both detectors are combined. With this novel approach, the temporal resolution using a half-scan reconstruction technique is as low as 82 milliseconds in currently available systems and remains constant at all heart rates [4]. Initial studies have shown that with dual-source CT motion-free depiction of the coronary vessels is possible even at high heart rates of up to 90 beats per minute (bpm) [5, 6].
Image reconstruction in patients with low and intermediate heart rates is commonly performed in mid diastole because this phase has been reported to be the most tranquil phase of the cardiac cycle [7, 8]. Early results indicate that with dual-source CT there is a shift of the optimal reconstruction window away from mid diastole [6].
The purpose of our study was to evaluate the presence and, if present, the extent of motion artifacts in the coronary vessels throughout the cardiac cycle and to determine the position of the optimal systolic and diastolic reconstruction intervals for dual-source CT.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patients were examined with a dual-source CT scanner (Somatom Definition, Siemens Medical Solutions). Beta-blocker premedication was not used in any patient.
The scanning delay was determined using a bolus-triggering technique (Care Bolus, Siemens). A single unenhanced scan was obtained at the level of the aortic root. This scan was used to place a 15-mm region of interest (ROI) inside the lumen of the ascending aorta. Then, 80 mL of a nonionic contrast medium (iopromide [Ultravist 370, Bayer Healthcare]) was injected at a flow rate of 5 mL/s through an antecubital catheter. As soon as the attenuation in the ROI exceeded 150 H, scanning was initiated. The scan range—from the aortic root above the ostium of the left coronary artery to the diaphragm—was covered in one breath-hold.
The CT parameters were as follows: detector collimation, 2 x 64 x 0.6 mm; pitch, adapted to heart rate (range, 0.2–0.46); 330-millisecond rotation time; tube current–time product, 400 mAs; and tube voltage, 120 kV. Tube current modulation was used to reduce patient dose. The maximum tube current was used only between 25% and 90% of the R-R interval. During the rest of the R-R interval, the tube current was reduced to 30% of the nominal output. This resulted in a mean dose index of 54.9 mGy. The mean scanning duration was 9 ± 2 (SD) seconds.
For every patient, 20 CT data sets were reconstructed in 5% steps throughout the entire R-R interval using a 512 x 512 matrix. The slice thickness was set to 1.5 mm and the increment, to 1.0 mm. For all images, a soft convolution kernel (B20f) was used. The field of view was set to 180 mm and was increased only if it was not sufficient to include the entire heart.
From the recorded ECG trace, the minimum and maximum heart rates were
extracted. The percentage variability of the heart rate was calculated
according to following equation:
 |
Image Analysis
The CT data sets were transferred to a separate workstation (Multimodality
Workplace, Siemens) with a 4D image viewer (InSpace 4D, Siemens)
installed.
For every patient, all 20 data sets were loaded into the application and displayed in a volume-rendering mode with settings adapted to an optimal delineation of the coronary vessels. The software also permitted the display of multiplanar reformations (MPRs) of each set of CT images.
Two independent readers assessed the three main coronary vessels (RCA, LAD, and LCX) in the volume-rendering mode and determined the reconstruction interval with the least motion artifacts for both the diastolic and the systolic phases. The presence of motion artifacts in this previously selected phase was then evaluated by the same two readers using the curved MPR mode and, if present, the motion artifacts were graded on a 5-point scale with 1 representing no motion artifacts and 5 representing severe motion artifacts over the entire course of the vessel.
Subgroup analysis was performed in patients presenting with heart rates of 50–60, 60–70, 70–80, and > 80 bpm.
Statistical Analysis
Statistical analysis was performed using SPSS software (SPSS version 11.0,
SPSS).
The difference between the image quality in systolic and diastolic reconstructions was compared using the Wilcoxon's signed rank test. Interobserver agreement for the position of the optimal reconstruction interval was calculated using Cohen's kappa statistic.
The influence of heart rate and heart rate variability on the motion score was assessed by means of a Pearson's correlation coefficient analysis for every vessel (RCA, LAD, and LCX) in both systolic and diastolic reconstructions and for the mean motion score of all three vessels. A p value of < 0.05 was considered statistically significant.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 269 vessels were evaluated in the 90 patients. In one patient, the LCX was not evaluated because it was occluded by tumor inversion.
The positions of the optimal reconstruction windows are summarized in Table 1. The distribution of the best systolic and diastolic reconstruction windows for the three arteries is depicted in Figures 1 and 2.
|
|
|
= 0.72
for RCA, = 0.60 for LAD; diastole, =0.65 for RCA, =
0.60 for LAD). For the LCX, the interobserver agreement was 0.54 (systole) and
0.55 (diastole).
Heart Rate and Positions of the Optimal Reconstruction Windows
The position of the optimal reconstruction window shifted toward the next P
wave with increasing heart rates for both systolic and diastolic
reconstructions. The positions of the mean optimal systolic and diastolic
reconstruction window in relation to the heart rate are shown in
Table 2 and
Figure 3.
|
One single systolic reconstruction window was graded as optimal for all three vessels in 19 patients. For diastolic reconstructions, one single reconstruction window was optimal in 31 patients. In the remaining patients, individual optimal reconstruction windows were found for each vessel. In 39 patients, the optimal systolic reconstruction windows for the RCA and the LAD were 5% apart. The optimal diastolic reconstruction windows for these vessels were 5% apart in 39 patients.
The least variability of the optimal reconstruction intervals of the two vessels was found in patients presenting with a heart rate of between 60 and 80 bpm. In these patients, the optimal systolic reconstruction window of both arteries was within a 5% interval in 77.2% of the cases, and the optimal diastolic reconstruction window was not more than 5% apart in 86.0% of the cases. In patients with heart rates of > 80 bpm, the optimal reconstruction windows for both arteries were more than 10% apart in five of 16 cases (31%).
Motion Artifacts
The mean ± SD overall motion scores for the systolic reconstructions
at the optimal reconstruction window were 1.9 ± 0.8 (RCA), 1.7 ±
0.5 (LAD), and 2.0 ± 0.6 (LCX). The mean overall motion scores for the
optimal diastolic reconstructions were 1.7 ± 0.9, 1.5 ± 0.6, and
1.6 ± 0.7, respectively. In all three vessels, the motion scores in the
diastolic reconstructions were significantly lower than those in the systolic
reconstructions (p <0.05).
Influence of Heart Rate on Coronary Motion Artifacts
We found a significant correlation between the mean heart rate and coronary
motion in the diastolic reconstructions (RCA: r = 0.55, p
< 0.01; LAD: r = 0.53, p < 0.01; LCX: r = 0,
51, p < 0.01). No significant correlation was found between the
mean heart rate and coronary motion in the systolic reconstructions (RCA:
r = 0.04, p = 0.73; LAD: r = 0.03, p =
0.80; LCX: r =–0.25, p = 0.81). See
Table 3 and Figures
4A,
4B,
5A,
5B,
6A,
6B,
6C.
|
|
|
|
|
|
|
|
Influence of Heart Rate Variability on Motion Artifacts
Heart rate variability and coronary motion scores in systolic
reconstructions for the RCA and the LAD did not show a significant correlation
(RCA: r = 0.15, p = 0.15; LAD: r = 0.11, p
= 0.29). However, a weak correlation between motion score and heart rate
variability for the systolic reconstruction of the LCX (r = 0.23,
p = 0.04) was found.
For all three vessels, no significant correlation between heart rate variability and coronary motion in the diastolic reconstruction was revealed (RCA: r = 0.04, p = 0.69; LAD: r = 0.08, p = 0.49; LCX: r = 0.07, p = 0.51).
Motion Artifacts in Systolic and Diastolic Reconstructions
In patients with heart rates of < 70 bpm (n = 53), mean motion
scores were significantly lower in diastolic reconstructions (1.3 ±
0.4) compared with systolic reconstructions (1.9 ± 0.5; p <
0.01). In patients presenting with heart rates between 70 and 80 bpm, motion
scores were still lower in diastolic reconstructions than in systolic
reconstructions, but the difference between systolic and diastolic
reconstructions was not statistically significant
(Table 3).
In patients with heart rates of > 80 bpm, the motion scores in the diastolic reconstructions deteriorated rapidly, whereas the extent of coronary motion in the systolic reconstructions remained more constant. In 11 of the 16 patients presenting with a heart rate of > 80 bpm, motion scores were lower in the systolic reconstructions than in the diastolic reconstructions (2.1 ± 0.6 vs 2.6 ± 0.8, respectively; p < 0.05).
The plots of the linear regressions for the motion scores in both reconstructions cross each other at 80 bpm, suggesting that the transition from diastolic to systolic reconstructions should be made at 80 bpm (Fig. 6C).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Image reconstruction in cardiac CT is typically performed in mid to late diastole [8–10]. Diastole consists of four distinct phases: isometric relaxation, early rapid filling, diastasis, and atrial systole. Of these four phases, diastasis is the most tranquil phase. During diastasis, the myocardium is relaxed; atrial and ventricular pressures have equalized; and blood enters the ventricle passively from the lungs, leading to a small, gradual rise of ventricular volume [11, 12]. Although this phase is relatively long in patients with low heart rates, it dramatically shortens with increasing heart rates and even ceases to exist at heart rates of around 80 bpm [13]. However, there is a period of relatively small movement until the peak flow caused by the relaxation of the ventricle and the contraction of the atrium coalesce, which is the case at heart rates of around 100 bpm.
In contrast to the duration of diastole, the duration of systole is less affected by changes in heart rate [14, 15]. Systole consists of three phases: isometric contraction, rapid ejection, and reduced ejection. Isometric contraction is initiated by the QRS complex. In this phase, cardiac motion is small, but the duration of isometric contraction is too short to be suitable for cardiac imaging [12, 16]. Thus, only the phase of reduced ejection that lasts from the moment of peak pressure in the ventricle to the beginning of the closure of the aortic valve can be used for image reconstruction, especially because during the subsequent diastolic phase of isometric relaxation, cardiac motion is also at a minimum [7, 12]. Although the duration of the reduced ejection decreases with increasing heart rates [16–18], the combination of these two phases offers a time frame of up to 140 milliseconds with only little coronary movement.
The shortening of diastole and especially of the diastasis in patients with high heart rates explains the inverse correlation between motion scores and heart rate in diastolic reconstructions that was found in the present study. A negative correlation between the extent of motion artifacts and heart rate has been reported in other studies [19]; however, this correlation seems to be smaller with dual-source CT than with single-source CT. The reason for this smaller correlation is the fact that with dual-source CT the temporal resolution remains constant at 83 milliseconds at all heart rates, whereas the multisegment reconstruction algorithms used in single-source CT reach a comparable temporal resolution only at specific heart rates [2, 20, 21]. Even with dual-source CT, the image quality in diastolic reconstructions in heart rates of > 80 bpm deteriorates markedly. This finding is in concordance with data derived form electron beam CT (temporal resolution, 100 ms), where a sufficient diastolic rest period for the RCA and LCX was reported to exist only at heart rates of < 80 bpm [22, 23].
Compared with mid-diastolic image reconstructions, results from systolic reconstructions are inferior in patients with low heart rates because during the systolic phase of reduced ejection there is slightly more coronary movement than during the diastolic phase of diastasis [22–24].
With increasing heart rates, the duration of the tranquil phase at the end of systole decreases only slowly, whereas the duration of the diastolic phase of diastasis shortens dramatically and even ceases to exist at heart rates of > 80 bpm, explaining why image quality in systolic reconstructions did not deteriorate as much with increasing heart rates as in diastolic reconstructions, a fact that has not been reported before now, to our knowledge. The transition of the optimal reconstruction interval from diastole to systole in this study occurred around 80 bpm, a finding that is supported by Johnson et al. [6] in an early report on dual-source CT of the coronary arteries.
Using single-detector CT, several authors also found that systolic reconstructions can yield image quality superior to diastolic reconstructions in patients with high heart rates [7, 19, 25, 26]. With 16-MDCT, systolic image reconstruction is recommended in patients with heart rates > 67 bpm, whereas with the increased temporal resolution offered by 64-MDCT, the mid-diastolic reconstruction was preferred up to heart rates of 75 or even 85 bpm [7, 26, 27].
Surprisingly, we found no significant correlation between the variability of heart rate and the extent of coronary motion. In contrast, Herzog et al. [7] and Leschka et al. [26] found a strong correlation between coronary motion artifacts and heart rate variability using single-source 16-and 64-MDCT. Single-source CT uses multisegment reconstruction algorithms to optimize the temporal resolution in patients with high heart rates. However, the temporal resolution provided by these algorithms varies with different heart rates, explaining the correlation between the variability of the heart rate and the extent of motion artifacts. In contrast, the temporal resolution in dual-source CT remains constant, leading to a more constant image quality.
The shortening of diastole in relation to systole in patients with high heart rates also influences the position of the optimal reconstruction interval in relation to the duration of the R-R interval. In patients with low heart rates, the optimal systolic reconstruction interval is located at 30–35% of the R-R interval but shifts to approximately 40–45% in patients with high heart rates. Because diastasis shortens while the time needed for filling the ventricles remains fairly constant, the position of the optimal diastolic reconstruction interval shifts from 70–75% of the R-R interval in patients with heart rates of < 70 bpm to 85–90% in patients with heart rates of > 80 bpm. Other authors have confirmed these findings [19, 24].
Using 16-and 64-MDCT, several authors have reported that good image quality for all three coronary arteries was achievable using a single reconstruction in mid diastole in patients with low heart rates [8–10]. In contrast, we found the least variability between the optimal reconstruction intervals for the individual vessels in the present study in patients presenting with intermediate heart rates of between 60 and 80 bpm.
Our study has several limitations. The most important limitation is the fact that tube current modulation was used. Thus, image noise was substantially increased in images reconstructed in sections of the R-R interval where the tube current was reduced. However, the use of tube current modulation significantly reduces patient dose [28], and therefore tube current modulation should not be switched off. In the examinations included in this study, the nominal tube current was available from 25–90% of the R-R interval; thus, the tube current was only reduced in sections of the R-R interval where the optimal reconstruction window was not expected to be found.
Also, we rated image quality only on a per-vessel basis and not on a per-segment basis as several other authors have in previous studies. However, in the present study, the segment that showed the most severe motion artifacts set the score for the entire vessel, thus providing a reliable measure regarding the best image quality possible at any given point during the R-R interval.
Third, we did not analyze diagnostic accuracy; however, that value can be calculated only if corresponding coronary angiography is available, which was not the case for many of our patients.
To conclude, we can state that dual-source CT offers diagnostic image quality even at high heart rates; however, because of the physiologic limitations explained earlier, the best image quality is available in diastolic reconstructions in heart rates of < 80 bpm. Thus, to achieve the optimal image quality, it might be necessary to reduce a patient's heart rate to < 80 bpm.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Stolzmann, H. Scheffel, S. Leschka, A. Plass, S. Baumuller, B. Marincek, and H. Alkadhi Influence of Calcifications on Diagnostic Accuracy of Coronary CT Angiography Using Prospective ECG Triggering Am. J. Roentgenol., December 1, 2008; 191(6): 1684 - 1689. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. G. Lohan, K. Motamedi, K. Chow, R. Habibi, C. Panknin, S. G. Ruehm, and L. L. Seeger Does Dual-Energy CT of Lower-Extremity Tendons Incur Penalties in Patient Radiation Exposure or Reduced Multiplanar Reconstruction Image Quality? Am. J. Roentgenol., November 1, 2008; 191(5): 1386 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Oncel, G. Oncel, A. Tastan, and B. Tamci Evaluation of Coronary Stent Patency and In-Stent Restenosis with Dual-Source CT Coronary Angiography Without Heart Rate Control Am. J. Roentgenol., July 1, 2008; 191(1): 56 - 63. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Colletti Coronary CT Angiography Without {beta}-Blockers Am. J. Roentgenol., December 1, 2007; 189(6): 1324 - 1325. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
= systole.
= data for diastolic reconstructions.