| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Clinical Radiology, University Hospitals-Grosshadern,
Ludwig-Maximilians University of Munich, Grosshadern Campus, Marchioninistr.
15, Munich 81377, Germany.
2 Bracco Altana Pharma GmbH, Konstanz, Germany.
3 Worldwide Medical Affairs, Bracco Imaging SpA, Milan, Italy.
4 Department of Cardiology, University Hospitals-Grosshadern, Ludwig-Maximilians
University of Munich, Munich, Germany.
Received April 29, 2005;
accepted after revision September 23, 2005.
Address correspondence to K. Nikolaou
(Konstantin.Nikolaou{at}med.uni-muenchen.de).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Sixty-four patients with suspected coronary
artery disease and a low pretest likelihood of significant disease according
to the criteria of the American Heart Association underwent both MDCT of the
heart and quantitative conventional coronary angiography (QCA). MDCT
examinations were performed on a 16-MDCT scanner. CT data sets were evaluated
on a per-patient basis and a per-segment basis and were classified as
indicating no disease, nonsignificant disease (stenoses 
50%), or
significant disease (stenoses > 50%). Sensitivity, specificity, negative
predictive value (NPV), and positive predictive value (PPV) of 16-MDCT in the
detection or exclusion of significant and nonsignificant coronary artery
disease were evaluated on both per-patient and per-segment bases.
RESULTS. Regarding the success rate of 16-MDCT, 94% (60/64) of patients and 92% (388/420) of vessel segments were of sufficient quality for diagnosis. In the remaining 60 patients evaluated, QCA revealed significant coronary artery disease, nonsignificant disease, and no disease in 8.3% (5/60), 75.0% (45/60), and 16.7% (10/60) of cases, respectively, on a per-patient basis, and in 1.3% (5/388), 23.2% (90/388), and 75.5% (293/388) of cases, respectively, on a per-segment basis. The sensitivity, specificity, NPV, and PPV of 16-MDCT for the detection of significant coronary artery disease were 80.0%, 94.5%, 98.1%, and 57.1%, respectively, on a per-patient basis, and 80.0%, 99.2%, 99.7%, and 57.1% on a per-segment basis.
CONCLUSION. In a population having a low pretest likelihood of significant coronary artery disease, 16-MDCT shows a moderate to high sensitivity and high NPV for the detection or exclusion of significant disease, but has a somewhat reduced PPV compared with QCA.
Keywords: angiography • atherosclerosis • coronary artery disease • MDCT
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In routine practice, however, patients with typical anginal complaints or known or strongly suspected coronary artery disease would possibly undergo invasive cardiac catheterization directly rather than an additional diagnostic MDCT examination [7]. Thus, rather than imaging of patients with a high pretest likelihood of significant coronary artery disease, the primary role of cardiac MDCT would likely be imaging of patients in whom a detailed depiction of the coronary arteries is needed to exclude significant coronary artery disease and therefore avoid invasive cardiac catheterization. A cohort of this type would likely comprise patients with noncardiac and atypical chest pain, asymptomatic patients with a high-risk profile for coronary artery disease, and patients with equivocal stress test findings.
Therefore, the aim of our study was to determine the diagnostic accuracy of 16-MDCT for the detection of significant coronary artery disease in patients with a low pretest likelihood of disease. Determinations of the ability of 16-MDCT to differentiate significant from nonsignificant coronary artery disease were performed on both per-patient and per-segment bases to fully assess the potential value of cardiac MDCT in clinical routines.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patient Population
A total of 64 patients (34 men, 30 women; mean age, 60 ± 10 [SD]
years; range, 37-78 years) were included over a period of 13 months (May
2003-June 2004). After assessment of the CT image quality, 60 patients were
eligible for final image evaluation (see Results section). The mean body mass
index of all patients was 28.1 (minimum, 20.1; maximum, 37.5). Patients
eligible for enrollment included men and women with a low pretest likelihood
of coronary artery disease, as described in the ACC/AHA/ACP-ASIM (American
College of Cardiology/American Heart Association/American College of
Physicians-American Society of Internal Medicine) guidelines for the
management of patients with chronic stable angina
[8]. Deriving the main criteria
for a low pretest likelihood of coronary artery disease from these guidelines,
the following patient cohorts were included: patients 35 years old or more who
had noncardiac chest pain (n = 38 patients: 12 women, 26 men),
patients 35 years old or more (women) or 35-49 years old (men) with atypical
angina (n = 23 patients: 15 women, eight men), and patients 35-49
years old (women only) with typical angina (n = 3). Patients with
known, clinically established coronary artery disease, a high pretest
likelihood of coronary artery disease (men of any age, both men and women of
3 50 years old with typical angina), known allergic reactions to
iodine-containing contrast media, renal insufficiency (serum creatinine >
1.5 mg/dL), hyperthyroidism (thyroid stimulating hormone < 0.3 mU/L),
pregnancy, or a heart rate of more than 60 beats per minute (bpm) at the time
of the MDCT angiography examination were excluded from the study. The IV
administration of 5-20 mg of metoprolol (Beloc, Astra Pharmaceuticals) shortly
before the examination was permitted in patients with a heart rate of more
than 60 bpm, but in this case, only patients with successful and adequate
lowering of the heart rate were included. In total, IV administration of
ß-blockers was necessary in 23 patients who were eligible for the study.
In 18 of these 23 patients, an adequate lowering of the heart rate to less
than 60 bpm was achieved.
All patients included underwent conventional X-ray coronary angiography within 72 hr before or after MDCT. The decision to perform X-ray angiography was made by a cardiologist independently and for clinical reasons only. The patients selected for the study were consecutive patients who fit the selection criteria as described, had already undergone or were scheduled to undergo conventional X-ray angiography for clinical reasons, and had a heart rate of 60 bpm or less immediately before the CT examination.
MDCT Protocol
CT angiography was performed using a 16-MDCT scanner (Somatom Sensation 16,
Siemens Medical Solutions), with the following acquisition parameters:
detector collimation, 16 x 0.75 mm; spatial resolution, 0.6 x 0.6
x 0.75 mm; gantry rotation, 370 msec; temporal resolution, 93-185 msec;
table feed, 3 mm per rotation; tube voltage, 120 kV; tube current, 750 mAs;
scanning duration, 20 sec; dose modulation (ECG-pulsing)
[9]. For optimal motion-free
image quality, data sets were reconstructed in mid-diastole (mean interval,
633 ± 125 msec after the R wave). In cases with insufficient image
quality of the primary data set (reconstructed at 70% of the R-R interval),
additional reconstructions (e.g., 65%, 75%, or 80% of the R-R interval) were
performed, and the data sets with the optimal image were chosen for further
evaluation. These additional reconstructions were required for 29 of 60
patients (48%). Image reconstruction in systole was not possible because dose
modulation was used on a routine basis to lower the tube voltage during
systole to 20% of the nominal value. This resulted in significant reduction of
the signal-to-noise ratio of the systolic reconstructions.
Contrast Medium Administration
Enrolled patients were randomly assigned to one of two study groups to
receive commercially available solutions of either iomeprol 300 (Iomeron 300,
containing 300 mg I/mL, Bracco Altana Pharma; group A) or iomeprol 400
(Iomeron 400, containing 400 mg I/mL; group B). Administration of pre-warmed
(37°C) contrast medium was performed IV into the left antecubital vein at
a rate of 1 g/sec of iodine to achieve an equivalent total iodine dose of 25 g
in all patients. For iomeprol 300, a total volume of 103 mL (20-mL test bolus
+ 83-mL main bolus) was administered at a rate of 3.3 mL/sec. For iomeprol
400, a reduced total volume of 78 mL (15-mL test bolus + 63-mL main bolus) was
administered at a rate of 2.5 mL/sec. The test bolus and main bolus injections
for both study groups were administered using a power injector (Stellant,
Medrad) and an 18-gauge needle, and each injection was followed by a saline
flush (50 mL) injected at the same rate.
The individual delay between the injection of contrast medium and the arrival of the contrast bolus in the ascending aorta was determined by evaluating the test bolus curve with commercially available software (DynEva, Siemens Medical Solutions), requiring an additional delay of 6 sec.
Conventional X-Ray Coronary Angiography
All patients underwent conventional coronary angiography within 72 hr
before or after the MDCT angiography examination. None of the patients
underwent intervening interventional or surgical treatment. Invasive coronary
angiograms were evaluated quantitatively (QuantCor QCA [quantitative coronary
angiography], Siemens Medical Solutions) by an independent, blinded
investigator. QCA was performed for all coronary artery lesions detected.
Determinations were made of the mean diameter reduction of the coronary lumen
in two projections and of the segmental location of the stenosis according to
the 15-segment AHA coronary artery model
[10]. Significant coronary
artery disease on a per-patient basis was diagnosed if at least one
significant stenosis (> 50%) was detected.
MDCT Angiography Analysis
All 16-MDCT data sets were analyzed by two independent experienced
reviewers who were blinded to the concentration of contrast medium
administered and to the results of X-ray coronary angiography. A commercially
available postprocessing workstation (Leonardo, Siemens) was used that
permitted assessment not only of the original axial images but also of
multiplanar reformations (multi-planar reconstruction), maximum intensity
projections, and 3D volume-rendered reconstructions.
Assessment of Image Quality
Image quality was determined subjectively on a per-patient basis by both
reviewers in consensus. A 4-point grading system was used in which image
quality was rated as excellent, good, sufficient, or insufficient. Only data
sets considered excellent, good, or sufficient were evaluated further; scans
considered to be of insufficient quality were excluded.
Image sets considered to be of adequate quality overall were evaluated further on a per-segment basis to determine whether individual coronary artery segments were of diagnostic or nondiagnostic image quality. Assessment was performed of the proximal and middle coronary artery segments according to the 15-segment model of the AHA [10]—that is, segments 1, 2, and 3 of the right coronary artery (RCA); segment 5 of the left main artery (LM); segments 6 and 7 of the left anterior descending coronary artery (LAD); and segment 11 of the left circumflex artery (LCX).
Before subsequent assessments of diagnostic efficacy, evaluations were performed of the inter-group homogeneity of the patient populations in the two study groups to ensure that the overall diagnostic image quality achieved with the two injection protocols was similar. The comparability of the two injection protocols in terms of overall vascular enhancement was ascertained by means of contrast density measurements (in Hounsfield units [H]) determined at standardized regions of interest (ROIs) positioned in the proximal coronary vessels.
Assessment of Plaque Numbers and Plaque Types
All plaques detected were recorded by the two reviewers independently in
terms of the number of plaques detected per patient. The segmental location of
plaques was determined immediately there-after in a final consensus
interpretation. Detected plaques were evaluated for morphology according to
the following classification: 1, noncalcified plaque; 2, calcified plaque; and
3, mixed plaque.
Assessment of Diagnostic Accuracy
The accuracy of 16-MDCT for the detection of coronary artery stenoses was
assessed preliminarily by both reviewers independently. Determinations of the
presence or absence of disease were performed for each coronary artery segment
using a 3-point scale in which 1 indicated no signs of atherosclerosis
("no disease"); 2, atherosclerotic vessel wall changes present,
but no significant stenosis ("nonsignificant disease"; 50%
lumen narrowing); and 3, significant coronary artery lesion present
("significant disease"; > 50% lumen narrowing).
After the independent evaluations, a consensus interpretation was performed to obtain a final MDCT diagnosis on both per-patient and per-segment bases. On a per-patient basis, significant coronary artery disease was diagnosed if one or more significant stenoses were detected, independently of the segmental location.
The ability of 16-MDCT to correctly grade the degree of stenosis was determined retrospectively by comparing the findings obtained on 16-MDCT (in terms of no disease, nonsignificant disease, and significant disease) with the results of conventional coronary angiography.
Statistical Analysis
Data were analyzed statistically using two Microsoft Windows-based software
products (Med-Calc, version 7.0.0.2, 2002, MedCalc Software; and SPSS 12.0.1.,
2003, LEAD Technologies). Continuous variables were presented as mean ±
SD, and all statistical tests were determined at a significance level of
p < 0.05.
The ability of 16-MDCT to detect significant coronary artery lesions was determined using QCA as the standard of reference. Sensitivity, specificity, diagnostic accuracy, error rate, negative predictive value (NPV), and positive predictive value (PPV) were calculated on both per-patient and per-segment bases. Determination of the ability of 16-MDCT versus QCA to correctly grade detected disease was performed using the chi-square test after calculation of a frequency table and determination of the contingency coefficient (a measure of the degree of relationship, describing the dependence of the classifications in the frequency table; values range between 0, no association, and 1, maximal association).
Interreviewer agreement for the detection or exclusion of significant coronary artery disease was evaluated on per-patient and per-segment bases using kappa statistics, in which a kappa value of less than 0.20 implies poor agreement; 0.21-0.40, fair agreement; 0.41-0.60, moderate agreement; 0.61-0.80, good agreement; and 0.81-1.00, excellent agreement.
To compare the diagnostic accuracy of the iomeprol 300 group and the iomeprol 400 group, receiver operating characteristic (ROC) curve analysis was performed and the values for the areas under the curve (AUC) were evaluated.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The 60 patients included in the efficacy analysis comprised 17 men and 13 women (mean age, 58.1 ± 11.2 years; range, 37-74 years) in group A (iomeprol 300) and 15 men and 15 women (mean, 62.2 ± 8.2 years; range, 40-78 years) in group B (iomeprol 400). No clinically relevant differences existed in the distribution of age or sex between the two groups. The mean heart rate of the patient population during the MDCT examination was 57.5 ± 4 bpm (range, 45-60 bpm).
A total of 420 coronary artery segments (60 assessable data sets, seven segments per patient) were evaluated for diagnostic image quality. Of these 420 segments, 388 (92.4%) were considered to be of sufficient diagnostic quality for assessment.
The mean contrast densities of the coronary arteries were 259.1 ± 46.7 H for group A and 251.6 ± 51.0 H group B. No noteworthy differences in terms of vascular enhancement or diagnostic image quality were apparent between patients who received iomeprol 300 and those who received iomeprol 400. On the basis of these findings, patients in groups A and B were combined for subsequent evaluations of the overall diagnostic efficacy of coronary 16-MDCT angiography.
Conventional X-Ray Angiography
Invasive coronary angiography revealed significant coronary artery disease
(i.e., one or more stenoses of > 50%) in five (8.3%) of 60 patients,
nonsignificant disease (i.e., one or more stenoses of 50%) in 45 (75.0%)
of 60 patients, and no disease in 10 (16.7%) of 60 patients. All five patients
with significant stenoses were diagnosed as having disease in just one vessel
(one-vessel disease).
Of the 388 coronary artery segments assessable on 16-MDCT, five (1.3%) had significant (> 50%) coronary artery stenoses on QCA, 90 (23.2%) showed nonsignificant disease on QCA, and 293 (75.5%) had no disease on QCA. The five significant lesions were located in segment 1 of the proximal RCA (n =2), segment 7 of the mid LAD (n = 2), and segment 11 of the proximal LCX (n =1).
MDCT Diagnostic Accuracy
Plaque number and plaque types—Reviewers 1 and 2 detected
plaque in 22 (73.3%) patients in group A (44 and 45 plaques, respectively) and
in 21 (70.0%) patients in group B (64 and 66 plaques). After consensus
interpretation and pooling the data from both groups, 17 (28.3%) of 60
patients had no signs of coronary atherosclerosis, whereas 43 (71.7%) of 60
patients had at least one coronary artery plaque. On consensus interpretation,
a total of 109 plaques were detected in these 43 patients. These 109 plaques
comprised seven (6.4%) that were considered to be noncalcified, 18 (16.5%)
that were considered to be mixed, and 84 (77.1%) that were considered to be
calcified.
Diagnostic accuracy per patient—The findings for diagnostic
accuracy, error rate, sensitivity, specificity, NPV, and PPV are shown in
Table 1 by patient and by
segment for both significant and nonsignificant disease. Four of five patients
with significant coronary artery disease according to QCA were correctly
identified on 16-MDCT. The one significant lesion that was underestimated on
16-MDCT (CT diagnosis: nonsignificant disease 50%) was located in
coronary artery segment 1 of the RCA. On a per-patient basis, the specificity,
diagnostic accuracy, error rate, NPV, and PPV for the correct detection or
exclusion of significant coronary artery disease were 94.5%, 93.3%, 6.7%,
98.1%, and 57.1%, respectively.
|
A similar sensitivity (78%; 39/50 patients) was noted for 16-MDCT for the
diagnosis of any disease (i.e., including also nonsignificant stenoses
50%) on a per-patient basis (Table
1). However, although specificity remained high (90%; 9/10
patients), the some-times minimal lumen narrowing seen on QCA for
nonsignificant stenoses resulted in slightly lower overall accuracy for
16-MDCT (80%; 48/60 patients).
Excellent interreviewer agreement (
= 0.91) was noted on 16-MDCT for
the detection or exclusion of significant coronary artery disease on a
per-patient basis. Figures 1A,
1B,
1C,
1D and
2A,
2B,
2C and
2D show examples of correct
diagnoses made on 16-MDCT, and Figure
3A,
3B,
3C and
3D shows an example of a
false-positive diagnosis.
|
|
|
|
|
|
|
|
|
|
|
|
Slightly lower overall values for diagnostic accuracy, error rate,
sensitivity, specificity, NPV, and PPV were noted for the assessment of any
disease (i.e., including nonsignificant stenoses) by coronary artery segment,
again reflecting the sometimes minimal lumen narrowing seen on QCA
(Table 1). However,
interreviewer agreement for the detection or exclusion of significantly
diseased coronary artery segments was excellent ( = 0.89).
Comparison of diagnostic accuracy of iomeprol 300 and iomeprol 400—No significant differences between iomeprol 300 and iomeprol 400 were noted in terms of the overall diagnostic accuracy of 16-MDCT for the detection or exclusion of significant coronary artery disease when injection of these two contrast media is adjusted to deliver an equivalent total iodine dose (Table 2). The sensitivity, specificity, and overall diagnostic accuracy for the detection of significant coronary artery disease on a per-patient basis were 75.0% (3/4), 92.3% (24/26), and 90.0% (27/30) for group A, and 100% (1/1), 96.6% (28/29), and 96.7% (29/30) for group B, respectively. Figure 4A, 4B and 4C displays the results of ROC analysis to compare the diagnostic accuracies of the iomeprol 300 group, the iomeprol 400 group, and the combined patient population. Differences among the AUC values for iomeprol 300 (AUC = 0.837), iomeprol 400 (AUC = 0.856), and the combined patient population (AUC = 0.846) were not statistically significant.
|
|
|
|
Grading of coronary artery stenoses on MDCT—The 16-MDCT data were interpreted in blinded fashion on a per-segment basis according to a 3-point scale (no disease, nonsignificant disease, and significant disease) to correlate different degrees of coronary artery disease with findings from QCA. After retrospective matching of these three diagnostic 16-MDCT classes for all 388 assessable segments with the respective QCA data, a moderate to high contingency coefficient of 0.60 was obtained, indicating good agreement of CT angiography (CTA) and QCA diagnoses on a per-segment basis. Chi-square analysis revealed a highly significant (p < 0.001) concordance between CTA and QCA diagnoses (X2 = 364.71 and 314.60 for CTA and QCA, respectively). The frequency table for this analysis and the proportion of diagnoses on CTA and QCA are shown in Table 3.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our study was designed to determine the accuracy of MDCT in a typical clinical setting on both a per-patient basis and a per-segment basis. Importantly, only patients with a low pretest likelihood for having significant coronary artery disease were included, because this is likely to represent the type of patient referred for cardiac CT in a clinical environment [14]. The hypothesis to be proven was that in this specific patient population, MDCT can reliably help in the further stratification of these patients by facilitating the decision as to whether to perform invasive coronary intervention.
In contrast to earlier studies, detection of nonsignificant disease (i.e.,
including also nonsignificant lesions causing 50% lumen reduction as
determined by QCA) was also evaluated to assess the sensitivity of MDCT
coronary angiography for earlier stages of coronary atherosclerosis.
Detection of Significant Coronary Artery Stenoses
In the per-patient-based analysis of 60 patients, only five (8.3%) patients
were shown to have significant disease on QCA, indicating that the patient
cohort evaluated was representative of a low-risk population.
In the per-patient-based analysis, we found moderate to high sensitivity (4/5; 80%) and high specificity (52/55; 94.5%) and NPV (52/53; 98.1%) for the correct detection or exclusion of significant coronary artery disease. Results from the per-segment analysis were similar. Overestimation of stenosis on 16-MDCT was observed in only three patients overall. In a typical clinical setting, these false-positive findings on CTA might have led to further, more invasive diagnostic procedures despite the absence of significant disease. One possible reason for this overestimation of disease is the purely qualitative way in which the CTA data sets were assessed (i.e., by subjectively relating plaque size to adjacent lumen diameter). However, atherosclerotic lesions tend to grow both inside the lumen and outside the vessel border, exaggerating the true size of the lumen-obstructing plaque components. Automated analysis tools may help in this respect by facilitating comparison of lumen diameter or lumen area in the stenotic area with an adjacent luminal diameter without atherosclerotic lesions.
The precise plaque composition may be another reason for false-positive assessments of the degree of stenosis on 16-MDCT angiography. In the one case of a false-negative MDCT diagnosis and in all three false-positive diagnoses of significant coronary artery disease, the plaques were either fully calcified or of a mixed type containing calcifications. These results suggest not only that heavily calcified segments are a potential limitation to diagnosis, even on 16-MDCT scanners, but also that plaque composition has an impact on the accuracy of the CT diagnosis.
Detection of Nonsignificant Coronary Artery Disease and Stenosis Grading with CT
In the population described, a high percentage of patients (45/60; 75.0%)
showed signs of coronary atherosclerosis on QCA despite not having significant
coronary artery disease. In this patient cohort, especially if patients
present with atypical symptoms, further clinical stratification is often
difficult. Although low sensitivity (56/95; 59.0%) was noted on 16-MDCT for
the detection of nonsignificant coronary atherosclerotic disease on a
per-segment basis, a markedly higher sensitivity (39/50; 78.0%) was noted on a
per-patient basis. For nonsignificant stenoses in particular, correct grading
is difficult, and many lesions are overestimated because calcifications
preclude the accurate determination of luminal diameter. In the present study,
a low PPV (56/109; 51.4%) was found for the per-segment-based analysis.
In general, QCA and 16-MDCT use different references to measure lumen narrowing. QCA compares an apparently narrowed segment, seen in the plane in which it appears most narrowed, with an adjacent apparently normal segment, whereas 16-MDCT compares the lumen diameter to the full diameter of the vessel from its outer wall, including the plaque. Also, CTA has the potential to perform cross-sectional imaging through the narrowed portion of the lumen and to measure the area of the vessel lumen versus the area of an adjacent, apparently normal segment, which are not possible using QCA. Concluding, MDCT might even be superior to QCA for the detection of the early stages of disease, because CT can detect coronary artery plaque even in vessels for which the vessel lumen is apparently preserved because of positive remodelling effects. In such cases, QCA may be negative, because a positive diagnosis of atherosclerotic disease requires at least a minimal reduction in luminal diameter. The high number of false-positive diagnoses in the per-segment-based analysis may thus not be false-positive results at all. Rather, they may indicate plaques that are correctly detected by CTA but missed by QCA. This possibility is also reflected by the results for the grading of coronary artery stenosis on 16-MDCT compared with QCA. After retrospective matching of the three diagnostic CT classes (no disease, nonsignificant disease, and significant disease) with QCA data, a contingency coefficient of 0.60 was found for the 388 assessable coronary artery segments, indicating good agreement of the CTA and QCA diagnoses on a per-segment basis. Chi-square analysis further revealed highly significant (p < 0.001) concordance between the CTA and QCA diagnoses. In fact, for a detailed assessment of eccentric, extraluminal plaque, intravascular sonography would be the ideal gold standard for comparison with coronary CTA data [15]. However, intravascular sonography was not performed in our patient cohort.
Because a diagnosis of nonsignificant disease on CTA requires plaques merely to be visible, even if the artery lumen seems preserved, disagreement between QCA and CTA in the present study is likely to have arisen for plaques that were visible on CT but that were detected in segments having a lumen that remains preserved, giving a negative finding on QCA. For this reason, intravascular sonography might have been a more appropriate technique for comparison than QCA. Intravascular sonography has been shown previously to be an excellent standard of reference for determinations of stenosis grade on CTA [15].
Finally, the findings of our study suggest that inclusion of additional criteria such as plaque composition (contents of mixed, calcified, and noncalcified plaque components) may permit a more exact assessment of risk in patients with nonsignificant coronary artery lesions [16, 17]. Although few data are available on the risk stratification of coronary artery disease patients on the basis of CT plaque composition, plaque morphology as assessed on CT has been shown to have an impact on patient prognosis [18].
Limitations
A principal limitation of our study is that only 60 patients were evaluated
for the presence of significant coronary artery disease and that, of these,
only five had significant disease. Giving a sensitivity value in such a small
number of significantly diseased vessels (4/5 stenoses detected; 80%
sensitivity) might not be very statistically useful. Although the patients
enrolled were representative of a truly low pretest likelihood population, a
more complete evaluation of the true diagnostic value of 16-MDCT angiography
in these patients would require a much larger patient cohort.
A second potential limitation is that the data for the two patient groups with different contrast injection protocols were pooled before the determinations of diagnostic accuracy. In this regard, however, the total delivered amount of iodine (25 g) and the iodine injection rate (1 g/sec) were identical for the two groups, and quantitative analysis of the lumen enhancement in the coronary arteries revealed no significant differences. Furthermore, no differences were observed in the diagnostic accuracy achieved with iomeprol 300 and iomeprol 400, as determined by ROC analysis. A potential advantage of using iomeprol 400 is that the injection rate and overall administered volume can be reduced compared with that in the iomeprol 300 group (injection rate: 2.5 vs 3.3 mL/sec; contrast volume: 78 vs 103 mL, respectively).
A final limitation of our study is that only the proximal and middle segments of the main coronary arteries were included in the analysis. Unfortunately, despite the considerable technical improvements of MDCT in recent years, the temporal and spatial resolutions of interventional X-ray coronary angiography are still unmatched; the spatial resolution even on advanced MDCT scanners is still not sufficient to reliably identify stenoses in distal vessel segments. For precise and reproducible image analysis, both CT reviewers and the QCA reviewer (a cardiologist) had to review the identical seven proximal and middle coronary artery vessel segments, as defined by the AHA. This permitted a detailed and unquestionable matching of all results. Although in routine clinical practice larger side branches, such as diagonal branches and sometimes the marginal branches, may be of clinical relevance and might be targeted for intervention; these were not included in the analysis. As described, a total of only five significant stenoses were found in the coronary artery segments included in this comparison. However, two more significant coronary artery stenoses (> 75%) were identified in the eight coronary segments not compared with CTA. One significant stenosis was found in segment 10 of the LAD (second diagonal branch), and a second significant stenosis was found in segment 15 of the distal LCX. However, neither stenosis was treated by an intervention.
Future studies should focus on the potential of new MDCT scanners allowing acquisition of 64 slices per gantry rotation. These scanners enable further improved temporal and spatial resolution and permit the complete coronary artery tree to be included in the analysis [19].
In conclusion, in a population at a low pretest likelihood of significant coronary artery disease, 16-MDCT shows a satisfactory sensitivity for the detection of significant coronary artery disease and a moderate to high sensitivity for the detection of nonsignificant disease. Our findings show that the high NPV reported in earlier MDCT studies is preserved. Although 16-MDCT appeared to overestimate diagnoses as compared with QCA, resulting in a relatively high number of false-positive diagnoses of significant coronary artery disease, this may have been a consequence either of the relatively small patient population or of false-negative diagnoses on QCA. Finally, the use of a contrast medium containing 400 mg I/mL (iomeprol 400) permits the injection rate and total volume of administered contrast material to be reduced with no impact on 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:
|
|
 |
|
  C. Plumhans, G. Muhlenbruch, A. Rapaee, K.-H. Sim, T. Seyfarth, R. W. Gunther, and A. H. Mahnken Assessment of Global Right Ventricular Function on 64-MDCT Compared with MRI Am. J. Roentgenol., May 1, 2008; 190(5): 1358 - 1361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hamon, R. Morello, J. W. Riddell, and M. Hamon Coronary Arteries: Diagnostic Performance of 16- versus 64-Section Spiral CT Compared with Invasive Coronary Angiography Meta-Analysis Radiology, December 1, 2007; 245(3): 720 - 731. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Andreini, G. Pontone, M. Pepi, G. Ballerini, A. L. Bartorelli, A. Magini, C. Quaglia, E. Nobili, and P. Agostoni Diagnostic Accuracy of Multidetector Computed Tomography Coronary Angiography in Patients With Dilated Cardiomyopathy J. Am. Coll. Cardiol., May 22, 2007; 49(20): 2044 - 2050. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Andreini, G. Pontone, G. Ballerini, E. Bertella, E. Nobili, A. Parolari, and M. Pepi Bypass Graft and Native Postanastomotic Coronary Artery Patency: Assessment With Computed Tomography Ann. Thorac. Surg., May 1, 2007; 83(5): 1672 - 1678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hamon, G. G.L. Biondi-Zoccai, P. Malagutti, P. Agostoni, R. Morello, M. Valgimigli, and M. Hamon Diagnostic Performance of Multislice Spiral Computed Tomography of Coronary Arteries as Compared With Conventional Invasive Coronary Angiography: A Meta-Analysis J. Am. Coll. Cardiol., November 7, 2006; 48(9): 1896 - 1910. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||
