|
|
||||||||
1 Department of Cardiology, University Hospital Aachen, Pauwelsstraße 30,
Aachen 52072, Germany.
2 Department of Diagnostic Radiology, University Hospital Aachen, Aachen 52072,
Germany.
Received March 3, 2004;
accepted after revision May 18, 2004.
Address correspondence to R. Koos.
Abstract
|
|
|---|
SUBJECTS AND METHODS. We conducted a prospective study of 72 patients (38 men and 34 women; mean age ± SD, 69.5 ± 8.8 years) with aortic stenosis who underwent 16-MDCT and cardiac catheterization. Aortic valve calcification was assessed using the aortic Agatston score, aortic mass score, and aortic volume score. Severity of aortic stenosis was classified at cardiac catheterization. Aortic valve area and peak-to-peak and mean transvalvular gradients were correlated with the degree of calcification determined on MDCT.
RESULTS. All measured aortic valve calcification scores were significantly higher in patients with severe aortic stenosis (n = 46) than in patients with moderate (n = 15) or mild (n = 11, p < 0.001) aortic stenosis. Aortic valve calcification scores were inversely related to aortic valve area (r = 0.67, p < 0.001 for aortic mass score) and correlated significantly with peak-to-peak (r = 0.70, p < 0.001) and mean transvalvular (r = 0.72, p < 0.001) gradients. No correlation between the aortic valve calcification and the total coronary calcium scores was observed.
CONCLUSION. Aortic valve calcification assessed on 16-MDCT is associated with severity of aortic stenosis. Thus, aortic valve calcification scores should be calculated routinely in all patients undergoing MDCT for assessment of coronary calcification. High aortic valve calcification scores indicate possibly severe aortic stenosis and should prompt a further functional evaluation.
|
|
|---|
Electron beam CT has shown a good reproducibility for quantification of valvular calcium in aortic stenosis. Previous studies have compared electron beam tomography and echocardiography for assessment of aortic stenosis [46]. A correlation between the degree of valvular calcification and the severity of aortic stenosis has been shown [7].
Retrospectively ECG-gated MDCT has been evaluated primarily for noninvasive coronary angiography and for the detection and quantification of calcified and noncalcified coronary plaques [8, 9]. Because CT is a sensitive method for the detection of calcification, it is potentially useful for the assessment of aortic valve morphology and quantification of the degree of calcification [10, 11]. Preliminary data have shown a good correlation between the degree of aortic valve calcification assessed on 4-MDCT and transvalvular gradients and aortic valve area measured on Doppler echocardiography [12]. Nevertheless, the applicability of these findings to 16-MDCT scanners has yet to be proven.
The aim of this prospective study was to correlate the degree of valvular calcification assessed on 16-MDCT using three aortic valve calcification scores with the severity of aortic stenosis evaluated at cardiac catheterization in patients with aortic stenosis.
|
|
|---|
MDCT
All examinations were performed with a 16-MDCT scanner (SOMATOM Sensation
16, Siemens) using a standardized imaging protocol with retrospective
ECG-gating. Patients were scanned in the supine position during a single
inspiratory breath-hold (mean duration, 21.7 ± 3.4 sec). The mean heart
rate of the patients was 69.3 ± 7.4 beats per minute. No medication was
administered before the examination.
The scanning parameters for the unenhanced MDCT examinations were as follows: collimation, 12 x 0.75 mm; tube voltage, 120 kV with an effective tube current time product of 133 mAs; table feed per rotation, 2.8 mm; and tube rotation time, 420 msec. Axial images were reconstructed at 60% of the R-R interval with an effective slice thickness of 3 mm and a reconstruction increment of 2 mm using a dedicated convolution kernel (B35f, Siemens Medical Solutions). The field of view was 180 x 180 mm with a 512 x 512 matrix.
Scanner quality assurance was performed by calibration using a standard cardiac phantom (CT cardiac phantom, QRM). The software program (WinDose 2.1, Scanditrox Wellhöfer) was used to calculate the effective radiation dose of unenhanced MDCT examinations [13].
MDCT Image Evaluation
Images were assessed in a consensus interpretation by an experienced
radiologist and an experienced cardiologist. Both interpreters were blinded to
all patient data, including findings at cardiac catheterization. Image
analysis was performed on a separate computer workstation (Leonardo, Siemens)
equipped with a dedicated software tool for calcium scoring (Calcium Scoring
CT, Siemens). For quantitative assessment of aortic valve calcification, three
aortic valve calcification scores were calculated with a detection threshold
of 130 H: the aortic Agatston score, the aortic mass score, and the aortic
volume score
[1416].
The Agatston score was calculated by multiplying the lesion area by an
attenuation factor derived from the maximal Hounsfield units within the area,
as described by Agatston et al.
[10]. We used the volume of
calcium measured in cubic millimeters as the aortic volume score. The aortic
mass score was measured in milligrams of calcium hydroxyapatite.
Calcification was attributed to the aortic valve if it was clearly part of the valve cusps. Supravalvular calcifications and calcifications of the coronary arteries including the ostia were removed by manual segmentation. In addition, the amount of coronary calcification was assessed using the Agatston score, which was referred to as "coronary calcium score" in our study.
Cardiac Catheterization
All patients underwent cardiac catheterization 1 or 2 days after MDCT.
Peak-to-peak and mean transvalvular gradients were determined, and the aortic
valve area was calculated using the Gorlin formula
[17]. The severity of aortic
stenosis was classified as mild (aortic valve area
1.5 cm2),
moderate (aortic valve area between < 1.5 and
1.0 cm2), and
severe (aortic valve area < 1.0 cm2), according to the American
College of CardiologyAmerican Heart Association practice guidelines
[18].
Statistical Analysis
Continuous variables are expressed as mean values ± SD. Continuous
variables were compared with Student's t test for unpaired samples.
Kruskal-Wallis H test was used for comparisons of the different grades of
severity of aortic stenosis assuming a nonnormal distribution. Categoric
variables were compared with a chi-square analysis. Spearman's rank
correlation was used to assess the relationship between the aortic valve
calcification identified on MDCT images and the severity of aortic stenosis
identified on angiograms. A p value of less than 0.05 was considered
statistically significant. Statistical analysis was performed with use of
statistical software (SPSS version 10.0 for Windows [Microsoft], Statistical
Package for the Social Sciences).
|
|
|---|
|
Quantitative Assessment of Aortic Valve Calcification and Its Relationship to Cardiac Catheterization
The ranges for the aortic valve calcification scores for the entire group
of patients were 4910,227 by the Agatston method, 458,149 by the
volumetric method, and 72,962 by the mass method. All measured aortic
valve calcification scores were different among the catheterization-determined
severity groups (Table 2).
Patients with severe aortic stenosis showed significantly higher aortic valve
calcification scores than patients with moderate or mild aortic stenosis
(Figs. 1,
2,
3,
4,
5,
6). In addition, patients with
moderate aortic stenosis showed higher aortic valve calcification scores than
those with mild aortic stenosis (p = 0.001 for all aortic valve
calcification scores). Aortic stenosis severity was inversely correlated to
the three different aortic valve calcification scores (p < 0.001).
A significant correlation was found between the invasively determined aortic
valve area and all measured aortic valve calcification scores
(Table 3). In addition, the
aortic valve calcification scores showed significant correlation with the mean
and the peak-to-peak transvalvular gradients
(Table 3).
|
|
|
|
|
|
|
|
To further characterize the usefulness of aortic valve calcification scores in identifying patients with severe aortic stenosis, we constructed receiver operating characteristic (ROC) curves. The best ROC curve for the aortic mass score had an area under the curve (Az) of 0.91 (Fig. 7). The Az values for the aortic volume score and the aortic Agatston score were similar (Table 4). As a tool for distinguishing patients with severe aortic stenosis from those with moderate or mild aortic stenosis, an aortic valve calcification score of 563 derived with the mass method had a sensitivity of 85%, a specificity of 92%, a positive predictive value of 95%, and a negative predictive value of 77%. Table 4 lists the sensitivities and specificities for different aortic valve calcification cutoff scores.
|
|
Coronary Calcification
The coronary calcium score by the Agatston method was not significantly
different among the aortic stenosis severity groups (p = 0.87). We
found no correlation between the aortic valve calcification (aortic Agatston
score) and the total coronary calcium scores (r = 0.17, p =
0.20).
|
|
|---|
Previous studies comparing electron beam CT and echocardiography in patients with aortic stenosis have shown an inverse correlation between the degree of valvular calcification and noninvasively determined aortic valve area [5, 7, 19]. Preliminary MDCT data showed a correlation between the mean echocardiographic gradient and valvular calcification assessed by semiquantitative analysis [20]. In contrast to the findings of a study by Cowell et al. [21] using MDCT without ECG-gating, our data showed a better correlation between aortic valve area or transvalvular gradients and the severity of aortic valve calcification, a finding that emphasizes the importance of ECG-gating. In a recent study using a 4-MDCT scanner, a good correlation was found between the degree of aortic valve calcification as measured on a 3D volume score and the severity of aortic stenosis as assessed on Doppler echocardiography [12]. The results of our study are in agreement with the results of that report. In addition to the factors studied in that report, however, we evaluated all routinely available calcification scores including Agatston, volume, and mass scores, introduced to improve reproducibility, using a 16-MDCT scanner. All aortic valve calcification scores showed a similar correlation between the severity of the aortic stenosis and the degree of aortic valve calcification. Moreover, we used the invasively determined aortic valve area as the reference standard, which is the standard of choice for the evaluation of aortic stenosis, especially in patients whose echocardiograms may be of poor quality. Our results emphasize that a large amount of aortic valve calcification seen on MDCT, especially in patients whose aortic valve calcification scores indicate severe aortic stenosis, should prompt a further cardiologic evaluation, including echocardiography, to assess the hemodynamic burden imposed by the calcified aortic valve.
Aortic valve calcification scores are quantitative and highly reproducible using electron beam CT [5, 8]. The volume score and the mass score have been introduced to decrease the variability of the coronary calcium measurements [15, 16, 22]. In measurements of aortic valve calcification, a similarly low interscan variability has been shown for Agatston and volume scores determined with electron beam CT [4]. Ohnesorge et al. [23] showed that MDCT-derived measurements of coronary calcification have a better reproducibility for calcium mass and volume scores than for the Agatston score. Preliminary MDCT data of patients with aortic valve calcification showed an overall median interscan reproducibility of 7.9% [12], comparable to that found with electron beam CT. Thus, a good reproducibility of aortic valve calcification scores has been shown in previous studies.
Potential of MDCT Technique
Evaluation of aortic valve calcification is of increasing importance as the
number of patients who undergo MDCT for evaluation of the coronary arteries or
unenhanced MDCT for coronary calcium scoring rises. In the future, an accurate
determination of aortic valve calcification may be more important still
because the degree of calcification has been shown to be a strong and
independent risk factor for disease progression and an adverse clinical
outcome [3]. Nevertheless,
prospective studies are needed to evaluate whether the degree of calcification
as determined on MDCT affects treatment. One group of researchers has
suggested that patients with severe aortic stenosis and marked calcification
belong to a high-risk subgroup that may benefit from early aortic valve
replacement, even in the absence of symptoms
[3]. Moreover, several ongoing
trials are assessing the impact of lipid-lowering therapy on the rate of
progression of aortic stenosis. If statin use can be assumed to be associated
with a decreased progression of aortic calcification
[24], our study results
indicate that MDCT is a valuable method for assessing aortic valve
calcification and is potentially useful in monitoring progression of this
calcification.
Study Limitations
Reproducibility of different aortic valve calcification scores in the
measurement of aortic valve calcification was not evaluated in our study.
Nevertheless, preliminary data have already shown a high reproducibility of
aortic valve calcification scores
[12]. Although a high
agreement between different scanners and coronary calcium scores has been
shown [25], the reference
ranges of the aortic valve calcium scores are dependent on the imaging
protocol and CT equipment used.
In conclusion, the results of our study show that a high aortic valve calcification score determined with 16-MDCT may indicate severe aortic stenosis, warranting further hemodynamic examination including echocardiography. We suggest routinely assessing aortic valve calcification scores in patients undergoing MDCT for other reasons, such as coronary calcium scoring. Such routine assessment may identify patients with significant aortic stenosis who warrant further cardiologic evaluation.
|
|
|---|
This article has been cited by other articles:
![]() |
T. M. LaBounty, B. Sundaram, P. Agarwal, W. A. Armstrong, E. A. Kazerooni, and E. Yamada Aortic Valve Area on 64-MDCT Correlates with Transesophageal Echocardiography in Aortic Stenosis Am. J. Roentgenol., December 1, 2008; 191(6): 1652 - 1658. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. F. Tops, S. C. Krishnan, J. D. Schuijf, M. J. Schalij, and J. J. Bax Noncoronary Applications of Cardiac Multidetector Row Computed Tomography J. Am. Coll. Cardiol. Img., January 1, 2008; 1(1): 94 - 106. [Abstract] [Full Text] [PDF] |
||||
![]() |
A.-C. Pouleur, J.-B. le Polain de Waroux, A. Pasquet, J.-L. J. Vanoverschelde, and B. L. Gerber Aortic Valve Area Assessment: Multidetector CT Compared with Cine MR Imaging and Transthoracic and Transesophageal Echocardiography Radiology, September 1, 2007; 244(3): 745 - 754. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. H. Mahnken, G. Muhlenbruch, M. Das, J. E. Wildberger, H. P. Kuhl, R. W. Gunther, M. Kelm, and R. Koos MDCT Detection of Mitral Valve Calcification: Prevalence and Clinical Relevance Compared with Echocardiography Am. J. Roentgenol., May 1, 2007; 188(5): 1264 - 1269. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Hara, W. R. Pedersen, E. Ladich, M. Mooney, R. Virmani, M. Nakamura, T. Feldman, and R. S. Schwartz Percutaneous Balloon Aortic Valvuloplasty Revisited: Time for a Renaissance? Circulation, March 27, 2007; 115(12): e334 - e338. [Full Text] [PDF] |
||||
![]() |
R. Koos, H. P. Kuhl, G. Muhlenbruch, J. E. Wildberger, R. W. Gunther, and A. H. Mahnken Prevalence and Clinical Importance of Aortic Valve Calcification Detected Incidentally on CT Scans: Comparison with Echocardiography Radiology, October 1, 2006; 241(1): 76 - 82. [Abstract] [Full Text] [PDF] |
||||
![]() |
A P leonard, R A Crossley, M Klusmann, and C A Roobottom Cardiac multidetector computed tomography (MDCT): method, indications and applications Imaging, September 1, 2006; 18(3): 151 - 159. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Alkadhi, S. Wildermuth, A. Plass, D. Bettex, B. Baumert, S. Leschka, L. M. Desbiolles, B. Marincek, and T. Boehm Aortic Stenosis: Comparative Evaluation of 16-Detector Row CT and Echocardiography Radiology, July 1, 2006; 240(1): 47 - 55. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |