| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Erasmus Medical Center, Dr. Molewaterplein, 40,
Rotterdam 3015GD, The Netherlands.
2 Department of Cardiology, Erasmus Medical Center, Rotterdam, The
Netherlands.
3 DIBIMEL, Section of Radiological Sciences, University of Palermo, Palmero,
Italy.
4 Computed Tomography, Siemens Medical Solutions, Forchheim, Germany.
Received November 19, 2004;
accepted after revision February 7, 2005.
Address correspondence to F. Cademartiri
(filippocademartiri{at}hotmail.com).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Thirty-eight patients who underwent MDCT
coronary angiography and conventional coronary angiography were enrolled in
the study. The inclusion criterion was the presence of mild heart rhythm
irregularities (i.e., premature beats; atrial fibrillation; mistriggering; or
low heart rate, defined as 40 beats per minute or less) during the scan. All
patients underwent MDCT with the following parameters: 16 detectors;
collimation, 0.75 mm; gantry rotation time, 375 msec; 120 kV; and effective
milliampere-second setting, 500–600. Images were reconstructed in two
settings: before ECG editing and after ECG editing (i.e., arbitrary
modification of temporal windows within the cardiac cycle at the site of mild
heart rhythm irregularities). Data sets were scored for the presence of
significant stenoses (
50% lumen reduction) in coronary segments 2 mm
diameter. The results of the two groups were compared with a McNemar test, and
a p value of less than 0.05 was considered significant.
RESULTS. The sensitivity, specificity, and negative and positive predictive values of MDCT coronary angiography for the detection of significant stenoses before and after ECG editing were 63% (41/65) and 92% (78/85); 97% (251/260) and 96% (305/317); 87% (62/71) and 87% (81/93); 91% (251/275) and 97% (305/313), respectively (p < 0.05). The proportion of nonassessable segments was reduced from 17% (70/416) before ECG editing to 2% (10/416) after.
CONCLUSION. ECG editing significantly improves diagnostic accuracy in a selected population of patients with mild heart rate irregularities.
Keywords: arteriography • coronary angiography • ECG editing • heart rhythm • MDCT angiography
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
50% lumen reduction) in a selected population of patients
[2,
3]. One limitation of MDCT coronary angiography is that high, irregular, or high and irregular heart rates prevent adequate visualization and assessment of the coronary vessels, and for this reason, patients are usually excluded from analysis or scanning [4]. This is because at high heart rates the motion speed of the coronary vessels is increased and the temporal window suited for imaging (e.g., mid- to end-diastole) is shortened. In patients with irregular heart rates, such as premature beats and atrial fibrillation, the temporal variability of the diastolic phase is increased between contiguous heart cycles. This creates motion artifacts due to the inaccurate location of temporal windows, the lack of data, or both. This prevents the application of homogeneous settings for image reconstructions. By arbitrarily modifying the position of the temporal windows within the cardiac cycle, it is possible to correct and compensate for part or all of the artifacts produced by mild heart rhythm irregularities (i.e., ECG editing).
The aim of this study was to compare diagnostic accuracy in a population of patients with mild heart rhythm irregularities using two different image reconstruction protocols for MDCT coronary angiography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
Exclusion criteria were a high heart rate (> 70 bpm) with bigemini or trigemini premature beats, previous allergic reaction to iodine contrast medium, renal insufficiency (serum creatinine > 120 mmol/L), pregnancy, respiratory impairment, unstable clinical status, marked heart failure, and previous bypass surgery or percutaneous coronary interventions.
Patients meeting the inclusion criteria also underwent noninvasive MDCT coronary angiography. The institutional review board approved the study, and the patients gave informed consent.
MDCT
The heart rate of each patient was measured on arrival. Patients with a
prescan heart rate of 65 bpm or greater were given 100 mg of metoprolol orally
and the scan was obtained 1 hr later. Patients were positioned on the CT table
and connected to the ECG trace. The variation in heart rate was observed for 5
min and was then tested during apnea (20 sec). If at rest and during apnea the
variation in heart rates was compatible with the definition of mild heart
rhythm irregularities, the patient underwent the scanning procedure.
In all patients, a bolus (100 mL at 4 mL/sec) of iodinated contrast material (iomeprol 400 mg I/mL, Iomeron, Bracco) followed by a saline chaser (40 mL at 4 mL/sec) was administered using a double-head power injector (Stellant, MedRad) through an 18-gauge cannula positioned in an antecubital vein.
The scanning parameters for MDCT coronary angiography (Sensation 16, Siemens Medical Solutions) were number of detectors, 16; individual detector width, 0.75 mm; gantry rotation time, 375 msec; effective temporal resolution, 188 msec; 120 kV; 500–600 effective mAs; feed/rotation, 3.0 mm; and scan direction, craniocaudal.
Synchronization between the passage of contrast material and data acquisition was achieved with real-time bolus tracking (CARE bolus, Siemens Medical Solutions) using a region of interest in the ascending aorta and a threshold of 100 H above the baseline attenuation to trigger the scan. The heart rate and ECG trace were recorded during scanning [5].
Image Reconstruction
Two study data sets were reconstructed using a retrospective ECG-gating
technique. The first study data set—the standard reconstruction
protocol—was selected among three reconstructed data sets without any
ECG editing using temporal windows starting -350, -400, and -450 msec before
the next R wave. Two experienced operators selected in consensus the data set
with the fewest motion artifacts. The second study data set—the
ECG-editing reconstruction protocol—was reconstructed in consensus by
two experienced operators using ECG editing (available at the MDCT scanner
console) at the level of anomalies of the ECG signal due to premature beats
(Appendix 1, which is available online at
www.ajronline.org,
and Figs. 1A,
1B,
1C,
1D,
1E, and
1F), atrial fibrillation,
mistriggering, and low heart rate. The ECG-editing procedure consisted of an
arbitrary modification (i.e., deletion, insertion, or repositioning) of the
number and position of temporal windows to provide the reconstruction software
with information with the least residual motion. Both data sets were
reconstructed with the following parameters: effective slice width, 1 mm;
reconstruction interval, 0.5 mm; field of view, 160 mm; and convolution
filter, medium smooth (B30f).
Image Analysis
Two experienced observers (the same who selected the data sets), blinded to
the results of the conventional coronary angiography and the identity of
patients, assessed in consensus the data set for the presence of significant
stenoses ( 50% lumen reduction). The evaluation of the data was performed
3 months after the reconstruction and selection of the data sets, and the
observers were unaware whether the data set was edited or unedited. The
evaluation was performed following the classification for coronary segments
established by the American Heart Association
[6]. Significant stenoses were
scored as absent (< 50%) or present ( 50%). Significant lesions had an
average diameter stenosis of 50% or more measured in two longitudinal views
orthogonal to each other.
Conventional Coronary Angiography
Conventional coronary angiography was performed within 2 weeks of MDCT
coronary angiography using a standard technique. One experienced observer,
unaware of the results of MDCT coronary angiography, determined the diameter
of all coronary branches and evaluated for the presence of significant
stenoses using quantitative coronary angiography (CAAS, Pie Medical Imaging).
Lesions with an average diameter stenosis of 50% or more in two orthogonal
views were considered significant obstructions.
Statistical Analysis
The result of quantitative coronary angiography performed in two orthogonal
projections was used as the reference standard to classify stenoses as
significant. The number of assessable segments on MDCT coronary angiography
was used to calculate sensitivity, specificity, positive predictive value, and
negative predictive value with 95% confidence intervals. A McNemar
nonparametric test was used to compare the diagnostic accuracy, and a
p value of less than 0.05 was considered significant.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Therefore, 38 patients (33 men; mean age, 59 ± 11 years; mean heart rate during the scan, 54 ± 9 [SD] bpm) were suitable for MDCT coronary angiography, image reconstruction, and data analysis. After exclusion of segments with a diameter of less than 2 mm, 416 segments remained available for analysis, 88 of which had a significant stenosis (21% per segment, corresponding to 2.3 lesions per patient). Of the 38 patients with mild heart rhythm irregularities, 21 irregularities were classified as premature beats, four as atrial fibrillation, six as mistriggering, and seven as low heart rate.
Sensitivity, specificity, negative predictive value, and positive predictive value for the detection of significant stenoses for the standard and ECG editing protocols were 63% (41/65) and 92% (78/85); 97% (251/260) and 96% (305/317); 87% (62/71) and 87% (81/93); 91% (251/275) and 97% (305/313), respectively (p < 0.05).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The main drawbacks of a high heart rate are the increased speed of the coronary arteries and reduced duration of diastole [7]. With a temporal resolution of approximately 200 msec, this results in a limited possibility to adapt the position of the temporal window in the cardiac cycle.
This limitation is less compromising at a low heart rate, for which a longer diastolic phase is available. In these conditions, the position of the temporal window in the cardiac cycle can be modified arbitrarily to compensate for motion artifacts resulting from arrhythmias. This tool can be defined as ECG editing. In fact, mild heart rhythm irregularities can be ruled out after careful ECG editing, thereby providing increased diagnostic accuracy (Table 1).
|
In our institution during the period of this study (10 months), we could include 42 additional patients, while the overall population included for MDCT coronary angiography was 398 patients. This finding suggests that ECG techniques could have been used for evaluation of an additional 42 patients (11%).
In our study, we compared the reconstruction capabilities of an MDCT scanner in the presence of mild heart rhythm irregularities before and after ECG editing. We found that the sensitivity and negative predictive value for the detection of significant stenoses were significantly different before and after ECG editing.
The effect of ECG editing on diagnostic accuracy depends on the number and distribution of lesions throughout the coronary tree. A population with a lower probability of disease (e.g., younger patients or those with less evident symptoms) would show a smaller variation in the negative predictive value than that shown in our study.
Our results suggest that from the technical point of view heart rhythm irregularities are not an absolute limitation of the technique. In fact, our technique exploits the flexibility of retrospective reconstruction to the limit. Another observation is that the number of assessable segments is increased after ECG editing.
The motion speed of the coronary vessels, especially if compared with effective temporal resolution of MDCT, remains a limitation. To compensate for this limitation, a slow heart rate, faster rotation time, or both are required.
Patients with heart rates above 70 bpm were not enrolled to prevent image degradation from other sources than those targeted by our study. Only cases that we defined as mild heart rhythm irregularities could be included in the evaluation because of the still limited effective temporal resolution of MDCT coronary angiography.
Scanning of patients with mild heart rhythm irregularities does not allow the use of X-ray reduction software such as ECG pulsing [8]. This is because the algorithm for dose reduction works with prospective triggering based on the R wave. In the presence of heart rate abnormalities, the location of the low-dose period will be variable and can fall within the diastole. Therefore, the presence of mild heart rhythm irregularities prevents the application of dose reduction protocols that are based on prospective tube current modulation.
In addition, the presence of heart rhythm irregularities, with the exclusion of low heart rates, does not allow the application of multisegmental reconstruction algorithms [9, 10]. This is because the variable diastolic filling of the heart prevents proper interpolation between the data originating from neighboring heart cycles.
Another disadvantage is related to the additional time required to edit the ECG. That time in our study was, on average, 7 min.
Editing the ECG in MDCT coronary angiography reduces the negative impact of mild heart rhythm irregularities on the diagnostic accuracy in a selected population of patients with stable angina. Improvements in scanner technology (e.g., faster rotation time or increased temporal resolution) and software capabilities (e.g., automatic recognition and editing of heart rhythm irregularities) could allow patients with mild arrhythmias to be scanned routinely with good diagnostic accuracy.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Matsumoto, T. Kondo, S. Watanabe, R. Kikumoto, T. Shimada, Y. Hiraoka, and K. Ueda ECG-Edited Middiastolic Phase Reconstruction Improves Image Quality at 64-MDCT Coronary Angiography of Patients with Atrial Fibrillation Am. J. Roentgenol., December 1, 2008; 191(6): 1659 - 1666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Hirai, J. Horiguchi, C. Fujioka, M. Kiguchi, H. Yamamoto, N. Matsuura, T. Kitagawa, H. Teragawa, N. Kohno, and K. Ito Prospective versus Retrospective ECG-gated 64-Detector Coronary CT Angiography: Assessment of Image Quality, Stenosis, and Radiation Dose Radiology, August 1, 2008; 248(2): 424 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Horiguchi, M. Kiguchi, C. Fujioka, Y. Shen, R. Arie, K. Sunasaka, and K. Ito Radiation Dose, Image Quality, Stenosis Measurement, and CT Densitometry Using ECG-Triggered Coronary 64-MDCT Angiography: A Phantom Study Am. J. Roentgenol., February 1, 2008; 190(2): 315 - 320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Feuchtner, T. Schachner, J. Bonatti, G. J. Friedrich, P. Soegner, A. Klauser, and D. z. Nedden Diagnostic Performance of 64-Slice Computed Tomography in Evaluation of Coronary Artery Bypass Grafts Am. J. Roentgenol., September 1, 2007; 189(3): 574 - 580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Attili and P. N. Cascade CT and MRI of Coronary Artery Disease:Evidence-Based Review. Am. J. Roentgenol., December 1, 2006; 187(6 Suppl): S483 - S499. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
