(Stroke. 2000;31:1104.)
© 2000 American Heart Association, Inc.
Original Contributions |
Reproducibility Validation Study Comparing Analog and Digital Imaging Technologies for the Measurement of Intima-Media Thickness
From the E. Grossi Paoletti Center, Institute of Pharmacological Sciences, University of Milan, and the Unit of Biostatistics, Institute H. San Raffaele (F.V.), Milan, Italy.
Correspondence to Prof C.R. Sirtori, Institute of Pharmacological Sciences, University of Milan, Via Balzaretti, 9, 20133, Milan, Italy. E-mail cesare.sirtori{at}unimi.it
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose—New advances in B-mode imaging technologies have led to improved quality in the detection of minute changes in the surface of intima-media thickness (IMT) and plaques. The new digital systems, with increased numbers of imaging channels, multiple frequency probes, and increased microprocessing speeds, now generate images comparable to those of the analog predecessors. Can these digital systems have reproducibility comparable to that of a pure analog system? We compared the Biosound 2000II (analog) system with the Esaote AU4 (digital) system.
Methods—Twenty-two subjects were chosen who had varying degrees of IMT on the far wall of the common carotid artery. Common carotid IMT was determined twice: the first time with the analog system and the second time with the digital system. With each system, replicate scans were made within 2 weeks.
Results—The intramethod agreement was high with the analog system, with a bias between readings of -0.010±0.033 mm, mean absolute difference of 0.027±0.020 mm, repeatability coefficient of 0.067, and correlation coefficient of 0.97. The digital system provided the highest reproducibility with a bias between readings of 0.002±0.016 mm, mean absolute difference of 0.012±0.011 mm, repeatability coefficient of 0.033, and correlation coefficient of 0.99. When the analog and digital systems were compared, the bias between readings was -0.011±0.024 mm with good agreement between the 2 systems; the repeatability coefficient was 0.047, with all points within ±2 SDs of the mean difference. The mean absolute difference between the 2 measurements was 0.018±0.015 mm with a correlation coefficient of 0.98.
Conclusions—The digital system for IMT evaluation compares well with the more widely used analog system and provides a reliable technology for common carotid IMT measurement that can be applied to clinical trials.
Key Words: carotid arteries • intima-media thickness • ultrasonography
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Intima-media thickness (IMT) of the carotid artery1 is increasing in importance and acceptance as an end point in large multicenter clinical trials.2 3 4 5 6 7 8 Clinical trials, including Pravastatin, Lipids, and Atherosclerosis in the Carotid arteries (PLAC-II),4 Kuopio Atherosclerosis Prevention Study (KAPS),9 Carotid Atherosclerosis Italian Ultrasound Study (CAIUS),10 Asymptomatic Carotid Artery Progression Study (ACAPS),11 Regression Growth Evaluation Statin Study (REGRESS),12 LIPID,13 and others,3 6 have shown that long-term studies can be successfully carried out while maintaining a high reproducibility value. Most of these studies were completed with the use of analog-based technology, which has long been the gold standard in IMT imaging. The new clinical trials, however, must also adopt new technologies as analog-based systems slip further and further into the past. The problem with the new digital processing systems is that this technology has no history or background to validate its ability in long-term studies and to maintain a very high reproducibility factor. To supplement this loss of background information, short-term system trials must be performed to ensure a high standard of credibility when this new technology is used. The present study was designed to perform a comparison between analog and digital B-mode ultrasound systems. To this aim, the within-method repeatability in common carotid IMT (CC-IMT) determination and the agreement between the 2 technologies were investigated.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study was performed in 22 healthy subjects who were recruited from the medical staff of E. Grossi Paoletti Center for the Study of Metabolic Disorders. Oral informed consent was obtained from all subjects, and the study was approved by the institutional review board. The subjects were selected on the unique criteria that they had a measurable IMT and could partake in 4 replicate scans: 2 with the analog system and an additional 2 scans with the digital system. The Biosound 2000II system, configured with an 8.0-MHz phased array probe, has pure analog hardware for image processing and maintains an axial resolution of 0.3 mm. The Esaote AU4 system, configured with a 7- to 10-MHz or a 10- to 13-MHz linear array probe, has analog-to-digital hardware for image processing and maintains an axial resolution of
0.2 mm. The IMT images, which were recorded on the same day and by the same sonographer, were taken from longitudinal views of the far wall of left and right common carotid arteries (CCA), which were examined in the anterior, lateral, and posterior planes. Each CCA was analyzed from distal to proximal, starting from 1 cm below the carotid bifurcation. The respective replicate scans were performed within 2 weeks from the initial visit by the same sonographer and according to the same scan protocol.
No specific rules concerning gain controls setting were imposed on the sonographer, who was allowed to regulate gain controls according to his experience.
An additional group of 10 subjects were enrolled so we could compare the reproducibility of analog versus digital systems equipped with comparable probes (transducer frequency). In these subjects, 6 scans were performed: 2 with the Biosound 2000II system, 2 with the Esaote AU4 equipped with probes of 7 to 10 MHz set with a transducer frequency of 7.5 MHz, and 2 with the Esaote AU4 equipped with probes of 10 to 13 MHz set with a transducer frequency of 10 MHz. In addition, respective replicate scans were performed within 2 weeks from the initial visit by the same sonographer according to the same scanning protocol.
All the CC-IMT measurements were performed on the recorded images by a single reader with the software EUREQUA, France, which allows automatic edge detection of IMT images.14 The operator was unaware of the results regarding the different measurements.
To minimize the effect of the cardiac cycles on IMT measurements, each
IMT was obtained through an average of the values of 
4 frames chosen
randomly at different moments of the cardiac cycle. According to this
protocol, the influence of the cardiac cycle was thus considered to be
negligible.15 In addition, to reduce measurement error,
the IMT of each imaged segment was determined 6 times. The individual
subject’s mean CC-IMT values were then calculated through an average
of the values for all carotid segments viewed in the 3 different
projections and for the left and right carotid segments.
Statistical Analysis
The data extracted from the CC-IMT measurements summarize the
ability of each system to reproduce and measure IMT. The intramethod
repeatability and the agreement between the 2 technologies were
evaluated through an estimate of the consistent bias between
readings, as recommended by Bland and Altman.16 The
intramethod comparison is made between the CC-IMT values, obtained with
each system during the initial and the respective replicate scan. The
coefficient of repeatability was calculated according to the method of
the British Standards Institution17 and corresponds to 2
SDs of the relative differences between replicate measurements.
The second and equally important comparison was made between the 2 systems during measurement of the same IMT. Specifically, the agreement was evaluated through an estimation of the consistent bias between the mean value of the 2 replicate readings obtained with the first method versus the mean value of the 2 replicate readings obtained with the second method. To facilitate a comparison of our findings with those provided in other reproducibility studies,18 the intramethod and the intermethod absolute differences, the correlation coefficients, the coefficient of variation, and the percent error between replicate scans also are provided.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Of the 22 subjects recruited for the study (mean±SD age 39±10.3 years, age range 26 to 55 years, mean total serum cholesterol 5.33±0.8 mmol/L), 14 (63.6%) were men and with the exception of 2 smokers, none had significant cardiovascular risk factors (eg, high blood pressure, lipids, diabetes).
Repeatability of Analog System (Biosound 2000II): Initial Versus
Replicate CC-IMT Values
The comparison of CC-IMT values between the initial and replicate
scans performed with the Biosound 2000II analog system is shown in
Table 1
and Figure 1 (left). The relative differences in
CC-IMT values remained constant as the IMT increased from 0.4 to
0.8 mm. Although there was some variability in the measurements of
CC-IMT, the intramethod agreement was high, with a bias between
readings of -0.010±0.033 mm and limits of agreement that ranged
from -0.076 to 0.056 mm, with all points within these limits. The
mean absolute difference was 0.027±0.020 mm (range 0.000 to
0.060 mm), the correlation coefficient was 0.97
(y=0.889x+0.072; P<0.0001; Figure 2, left), and the coefficient of
repeatability was 0.067 mm.
|
|
|
Repeatability of the Digital System (Esaote AU4): Initial Versus
Replicate CC-IMT Values
The comparison of CC-IMT values between the initial and replicate
scans performed with the Esaote AU4 digital system are shown in Table 1
and Figure 1 (right). As in the case of the analog
system, relative differences in CC-IMT values remained constant as the
IMT increased from the lowest to the highest value. Although in this
case there also was some variability in CC-IMT measurements, the
intramethod agreement was even better than that observed with the
analog system, with a bias between readings of -0.002±0.016 mm,
limits of agreement that ranged from -0.030 to 0.040 mm, and all
points but 1 within these limits. The mean absolute difference was
0.012±0.011 mm (range 0.000 to 0.040 mm), the correlation
coefficient was 0.99 (y=0.951x+0.026;
P<0.0001; Figure 2, right), and the coefficient of
repeatability was 0.033 mm.
Agreement Between Analog and Digital Processing in the CC-IMT
Determination
CC-IMT ranged from 0.39 to 0.80 mm, with a mean
value of 0.56±0.12 mm, when measured with the analog system and
from 0.41 to 0.78 mm, with a mean value of 0.57±0.12 mm,
when measured with the digital system. The bias between readings was
-0.011±0.024 mm with an excellent agreement between the 2
systems (limits of agreement from -0.056 to 0.030 mm; Figure 3, left). The mean absolute difference
between the 2 measurements was 0.018±0.015 mm (range 0.000 to
0.056 mm), and the correlation coefficient was 0.98
(y=0.963x+0.032; P<0.0001; Figure 3, right).
|
Effects of Frequency Transducer
A comparison of CC-IMT values between the initial and
replicate scans performed with the analog system and with the digital
system, equipped with 7.5- or 10-MHz probes, is shown in Table 2. Here, the CC-IMTs ranged from 0.46 to
1.12 mm (0.69±0.21 mm) when measured with the analog system
and from 0.48 to 1.04 mm (0.67±0.20 mm) and from 0.47 to
1.02 mm (0.67±0.19 mm) when measured with the digital system
equipped with the probe of 7.5 or 10 MHz, respectively. In this group
of 10 subjects, the absolute differences between replicate scans
observed with the analog system was 0.028±0.017 mm, whereas those
observed with the digital system were 0.019±0.008 and 0.014±0.012
with the 7.5- and 10-MHz probes, respectively. The limits of agreement
ranged from -0.075 to 0.054 mm with the analog system and from
-0.032 to 0.048 mm and from -0.036 to 0.041 with the digital
system equipped with the 7.5- and 10-MHz probes, respectively. Thus,
independent of probes and frequency transducer, the intramethod
agreement of CC-IMT measurements performed with the digital system was
always better than that observed with the analog system. Indeed, in a
comparison of the coefficients of repeatability obtained with the
analog system, those obtained with the digital system were 1.56 and
1.72 times smaller with the 7.5- and 10-MHz probes,
respectively.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Intramethod and Intermethod Accuracy and Reproducibility
The present study shows that the within-method biases, evaluated with a performance of repeated measurements, were small for both analog and digital systems. In addition, because both methods showed a constant SD of 0.4 to 0.8 mm, no obvious systematic relationship between replicate scan differences and their mean values was observed.
Intramethod variability also clearly shows that compared with
analog technology, digital technology can improve both the accuracy and
reproducibility of carotid IMT measurements. In fact, although the
analog system provides high-quality results (bias between readings
-0.01 mm, repeatability coefficient <0.07 mm, and mean
absolute differences <0.03 mm), reproducibility was further
improved with the use of digital technology, with a bias between scans
that was 5-fold less and a repeatability coefficient, a mean absolute
difference, a percent CV, and a percent error that were 2-fold less
than the corresponding values observed with the analog system. Because
the comparison was carried out by the same sonographer (D.B.) and
reader (M.A.) and with all other possible variables kept constant,
including methods for image processing,14 the observed
improvement may be due to a higher performance of digital
instrumentation, probably determined through differences in imaging
frequency (10 MHz digital versus 8 MHz analog), which may result in an
improved axial resolution (Table 2). In addition, the digital
electronics, with increased microprocessing speeds and the possibility
of setting the most adequate imaging characteristics, such as probe
frequency or scan depth, may be relevant for the reproducibility
improvement that was observed.
The comparison of a new method against an established method has
often been evaluated inappropriately through the use of the correlation
coefficient between the results of the 2 methods as an indicator of
agreement. Correlation coefficient measures the strength of a relation
between 2 variables, not the agreement between them. Moreover,
correlation depends on the range of the variables measured, with
wider ranges leading to better correlation.16 More often,
2 methods measuring the same variable will be related, and thus the
test of significance may be irrelevant regarding the question of
agreement. Thus, a very high between-methods correlation coefficient,
such as 0.98, as we determined (Table 1), may still harbor
sufficient disagreement to render the new digital method an unsuitable
substitute for the established analog standard. The analytical approach
of Bland and Altman,16 as used in the present study,
is the most appropriate approach to the evaluation of the
consistency of a new method of measurements compared with
an established method.16 Through the use of this approach,
the analysis of between-method agreement has shown that the 2
methods can result in a discrepancy in IMT measurements of
-0.011 mm (3.4%). On this basis, the 2 methods can be, at
least apparently, considered interchangeable. However, because the
repeatability coefficient of the analog system (0.067 mm) turned
out to be greater than that obtained with the between-method comparison
(0.047 mm), which in turn was greater than that obtained with the
digital system (0.033 mm), we might argue that the replacement of
an analog systems with a digital system is always possible without an
affect on the study results, whereas the replacement of a digital
system with an analog system is possible only after enlargement of the
sample size, which must be recalculated on the basis of analog
variability.
Comparison With Other Reproducibility Studies
Several studies that evaluated the reproducibility of IMT
measurements have been performed, and because of different protocols
for image acquisition, methods for IMT analysis, and methods
for variability quantification, a comparison of the results among these
studies may be difficult.18 As far as variability
quantification is concerned, the analytic approach of Bland and
Altman16 is the most appropriate method to evaluate
between-method agreement as well as the within-method repeatability.
However, because several authors carried out other analytical
approaches,3 14 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 to allow a comparison with our
results, we also provide all parameters most frequently
used in these studies, also to facilitate future users. Specifically,
we provide the correlation coefficient, the CV, the percent error, and
the absolute differences for the within-method analysis as well
as for the between-method agreement (Table 1).
The intraobserver absolute differences of the far wall CC-IMT
assessment between replicate scans of the present study was
compared with those obtained in several other reproducibility studies
(Figure 4). This figure indicate that the
digital system provides even better results than the analog system. It
should be emphasized, however, that an adequate comparison of our
results can be performed only versus those studies in which, as in our
case, the reproducibility of IMT measurements was determined in the far
wall of the CCAs of healthy subjects.14 19 26 34 Touboul
et al14 performed a reproducibility study among 14
subjects with the use of an analog system, and a mean difference (SD)
in IMT within observers of 0.06 (0.06) mm was found by using the
same reading methods to assess carotid IMT as we used in the
present study. Similar data were obtained, again with analog
systems, in the 2 reproducibility studies performed by Persson et
al19 and Salonen et al,26 who found mean
differences (SD) between replicate scans of 0.08 (0.07) and 0.09
(0.11) mm, respectively. In the present study, a significantly
better reproducibility was observed with both analog and digital
systems. This is probably due to the scan protocol we used, with 18
measurement sites considered for each subject. Slightly better results
were found by Gariepy et al28 and Wendelhag et
al,30 who found a mean difference (SD) between replicate
scans of 0.02 (0.02) and 0.007 (0.019) mm, respectively, which are
very similar to those obtained in the present study with the
digital technology.
|
In the present study, the participants were randomly selected from the medical staff of our lipid clinic, and IMT values ranged from only 0.4 to 0.8 mm. We fully recognize the importance in reproducibility studies of data obtained in patients with pathological IMTs. However, we are also convinced that these sources of variability are independent of image collection instrumentation and that they affect analog and digital systems with a similar variability. Thus, because the main objective of the present study was to investigate the variability induced by image collection instrumentation, all other possible sources of variability (eg, sonographer, reader, irregularity of vessel wall profile, instrumentation for image measurements) have been kept constant. Obviously, because in the present study only the IMT of CCA was considered, the reproducibility findings presented cannot be extended to either plaques or bulb or internal carotid arteries. However, we believe that an eventual reduction in measurement repeatability regarding plaques or bulb and internal carotid arteries should not be attributed to the instrumentation for image acquisition but rather to other sources of error, such as biological variability and sonographer/reader subjectiveness.
Practical Implication
We also analyzed the impact that an improved
precision in IMT measurements might have on sample size and duration of
follow-up required for clinical trials based on IMT progression
evaluation. With the assumption of a comparison of the mean slopes in a
treated and a control group, as, for example, in the CAIUS
study,10 sample size depends on the magnitude of the
expected treatment effect and on the within-group SD of the individual
slopes. We estimate that up to 45% of this SD is attributable to
measurement variability (data not shown). With this assumption, we
computed the sample size for a hypothetical clinical trial with
included digital ultrasound technology. Figure 5 reports the required sample size per
group, as a function of the expected treatment effects, in 3 different
cases: (1) the same measurement variability and the same follow-up (3
years) as in the CAIUS study10 ; (2) the variability
expected with the digital equipment, with 3 years’ follow-up; and (3)
the same as in condition 2 but with follow-up shortened to 2 years. We
note that with a treatment effect corresponding to 100% of that
observed for Pravastatin in CAIUS (0.019 mm/y of
lower progression), only 21 patients per group with 3 years of
follow-up, or 35 with 2 years of follow-up, are required.
|
); with digital equipment,
with 3 years’ follow-up (
); and with digital equipment, with 2
years’ follow-up (
).
Further Advantages of Digital Image Processing Systems
Apart from the higher accuracy and reproducibility, digital
systems present a number of technical advantages with respect to
the corresponding analog systems. Digital systems can be equipped with
multifrequency probes, thus increasing the range of depth at which
vessels can be investigated, and they can adapt the frequency and the
scan depth used according to the vessel depth, thus the highest
possible resolution is always obtained.
The characteristics of digital-based echograms can also be of clinical interest because when an image is frozen, the machine automatically stores not just 1 but a number of images, each of which can be zoomed, thus allowing the best image suitable for plaque or IMT measurements to be chosen. In addition, the presence and extent of atherosclerosis, with measurement and automatic calculation of percent of stenosis of transversal images, can be delineated, again in real time, thus highly improving the reliability of clinical reports. Finally, digital technology allows the storage of images not only on videotape but also on optical disc, thus providing the possibility of sending images via electronic mail, which is a very interesting feature for multicenter clinical trials with centralized readings.
Conclusions
Although the present study was performed on a relatively
small sample size, the results strongly support the idea that digital
systems may reduce the effect of variability determined with image
collection instrumentation.
The variability of IMT measurements is determined by the sonographer and reader, with instrumentation for image collection and for image measurements, and with biological differences between subjects (in patients with increased IMT or plaques, measurements become more inaccurate because of tortuous arteries, eccentric plaques, and irregularities). Because biological differences between subjects cannot be influenced, it is important to reduce, whenever possible, the effect of other factors that determine variability. The present results show not only that the digital system is a reliable method for the IMT assessment in clinical trials but also that compared with the analog system, the gold standard technology for IMT imaging, it can further improve accuracy and precision so that an adequate statistical power can be achieved with a smaller sample size and, thus, with lower costs.
In conclusion, the digital system for IMT evaluation compares well with the more widely used analog system and provides a reliable technology for CC-IMT measurement that can be applied to clinical trials.
Received July 9, 1999; revision received January 25, 2000; accepted January 25, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1. Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intima plus media thickness of the arterial wall: a direct measurement with ultrasound imaging. Circulation. 1986;74:1399–1406.
2. Furberg CD, Borhani NO, Byington RP, Gibbons ME, Sowers JR. Calcium antagonist and atherosclerosis: Multicenter Isradipine/Diuretic Atherosclerosis Study. Am J Hypertens. 1993;6:24S–29S.[Medline] [Order article via Infotrieve]
3.
Blankenhorn DH, Selzer RH, Crawford DW, Barth
JD, Chao-ran L, Ci-hua L, Mack WJ, Alaupovic P. Beneficial effects of
colestipol-niacin therapy on the common carotid artery: two- and
four-year reduction of intima-media thickness measured by ultrasound.
Circulation. 1993;88:20–28.
4. Crouse JR III, Byington RP, Bond GM, Espeland MA, Craven TE, Sprinkle JW, McGovern ME. Pravastatin, Lipids, and Atherosclerosis in the Carotid arteries (PLAC-II). Am J Cardiol.. 1995;75:455–459.[Medline] [Order article via Infotrieve]
5. De Groot E, Jukema JW, van Boven AJ, Reiber JH, Zwinderman AH, Lie KI, Ackerstaff RA, Bruschke AV. Effect of pravastatin on progression and regression of coronary atherosclerosis and vessel wall changes in carotid and femoral arteries: a report from the Regression Growth Evaluation Statin Study. Am J Cardiol. 1995;76:40C–46C.[Medline] [Order article via Infotrieve]
6.
Hodis HN, Mack WJ, LaBree L, Selzer RH, Liu CR,
Liu CH, Alaupovic P, Kwong-Fu H, Azen SP. Reduction in carotid
arterial wall thickness using lovastatin and
dietary therapy: a randomised, controlled clinical trial. Ann
Intern Med. 1996;124:548–556.
7.
Kroon AA, Asten WNJC, Stalenhoef AFH. Effects of
apheresis of low-density lipoprotein on peripheral vascular
disease in hypercholesterolemic patients with
coronary artery disease. Ann Intern Med. 1996;125:945–954.
8.
O’Leary DH, Polak JF, Kronmal RA, Manolio TA,
Burke GL, Wolfson SK, for the Cardiovascular Health
Study Collaborative Research Group. Carotid-artery intima-media
thickness as a risk factor for myocardial infarction and stroke in
older adults. N Engl J Med. 1999;340:14–22.
9.
Salonen R, Nyyssönen K, Porkkala E,
Rummukainen J, Belder R, Park JS, Salonen JT. Kuopio
Atherosclerosis Prevention Study (KAPS): a
population-based primary prevention trial of the effect of LDL lowering
on atherosclerotic progression in carotid and femoral arteries.
Circulation. 1995;92:1758–1764.
10. Mercuri M, Bond G, Sirtori CR, Veglia F, Crepaldi G, Feruglio S, Descovich, Ricci G, Rubba P, Mancini M, Gallus G, Bianchi G, D’Alò G, Ventura A. Pravastatin reduces carotid intima-media thickness progression in an asymptomatic hypercholesterolemic Mediterranean population: the Carotid Atherosclerosis Italian Ultrasound Study. Am J Med. 1996;101:627–634.[Medline] [Order article via Infotrieve]
11.
Furberg CD, Adams HP, Applegate WB, Byington RP,
Espeland MA, Hartwell T, Hunninghake DB, Lefkowitz DS, Probstifield J,
Riley WA, Young B, for the Asymptomatic Carotid Artery
Progression Study (ACAPS). Effect of lovastatin on early
carotid atherosclerosis and
cardiovascular events. Circulation. 1994;90:1679–1687.
12.
De Groot E, Jukema JW, Montauban van Swijndregt
AD, Zwinderman AH, Ackerstaff RG, van der Steen AF, Bom N, Lie KI,
Bruschke AV. B-mode ultrasound assessment of pravastatin
treatment effect on carotid and femoral artery walls and its
correlations with coronary arteriographic findings: a report of
the Regression Growth Evaluation Statin Study (REGRESS). J
Am Coll Cardiol. 1998;31:1561–1567.
13.
MacMahon S, Sharpe N, Gamble G, Hart H, Scott J,
Simes J, White H, on behalf of the LIPID Trial Research Group. Effects
of lowering average or below-average cholesterol levels on
the progression of carotid atherosclerosis: results of
the LIPID atherosclerosis substudy.
Circulation. 1998;97:1784–1790.
14. Touboul PJ, Prati P, Scarabin PY, Adrai V, Thibout E, Ducimetiere P. Use of monitoring software to improve the measurement of carotid wall thickness by B-mode imaging. J Hypertens. 1992;10(suppl):S37–S41.
15. Baldassarre D, Werba JP, Tremoli E, Poli A, Pazzucconi F, Sirtori CR. Common carotid intima-media thickness measurement: a method to improve accuracy and precision. Stroke. 1994;25:1588–1592.[Abstract]
16. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–310.[Medline] [Order article via Infotrieve]
17. British Standards Institution. Precision of Test Methods, I: Guide for the Determination and Reproducibility for a Standard Test Method (BS 5497, part 1). London, UK: British Standards Institution; 1979.
18.
Kanters SDJM, Algra A, van Leeuwen MS, Banga JD.
Reproducibility of in vivo carotid intima media thickness measurements:
a review. Stroke. 1997;28:665–671.
19.
Persson J, Stavenow L, Wikstrand J, Israelsson B,
Formgren J, Berglund G. Non-invasive quantification of atherosclerotic
lesions: reproducibility of ultrasonographic measurement of
arterial wall thickness and plaque size. Arterioscler
Thromb. 1992;12:261–266.
20.
Riley WA, Barnes RW, Applegate WB, Dempsey R,
Hartwell T, Davis VG, Bond MG, Furberg CD. Reproducibility of
non-invasive ultrasonic measurement of carotid
atherosclerosis: the Asymptomatic Carotid
Artery Plaque Study. Stroke. 1992;23:1062–1068.
21.
Joakimsen O, Bønaa KH, Stensland-Bugge E.
Reproducibility of ultrasound assessment of carotid plaque occurrence,
thickness, and morphology: the Tromsø Study. Stroke. 1997;28:2201–2207.
22.
Wendelhag I, Wiklund O, Wikstrand J.
Arterial wall thickness in familial
hypercholesterolemia: ultrasound measurement of
intima-media thickness in the common carotid artery. Arterioscler
Thromb. 1992;12:70–77.
23.
Wendelhag I, Wiklund O, Wikstrand J.
Atherosclerotic changes in the femoral and carotid arteries in familial
hypercholesterolemia: ultrasonographic
assessment of intima-media thickness and plaque occurrence.
Arterioscler Thromb. 1993;13:1404–1411.
24. Geroulakos G, Ramaswami G, Veller MG, Fisher GM, Renton S, Nicolaides A, Waldron HA, Diamond J, Elkeles RS. Arterial wall changes in type 2 diabetic subjects. Diabet Med. 1994;11:692–695.[Medline] [Order article via Infotrieve]
25. Veller MG, Fisher CM, Nicolaides AN, Renton S, Geroulakos G, Stafford NJ, Sarker A, Szendro G, Belcaro G. Measurement of the ultrasonic intima-media complex thickness in normal subjects. J Vasc Surg. 1993;17:719–725.[Medline] [Order article via Infotrieve]
26. Salonen R, Haapanen A, Salonen JT. Measurement of intima-media thickness of common carotid arteries with high-resolution B-mode ultrasonography: inter- and intraobserver variability. Ultrasound Med Biol. 1991a;17:225–230.
27.
Salonen JT, Salonen R. Ultrasonographically
assessed carotid morphology and the risk of coronary heart
disease. Arterioscler Thromb. 1991;11:1245–1249.
28.
Gariepy J, Massonneau M, Levenson J, Heudes D,
Simon A. Evidence for in vivo carotid and femoral wall thickening in
human hypertension: Groupe de Prevention Cardio-vasculaire en Medecine
du Travail. Hypertension. 1993;22:111–118.
29.
Pauciullo P, Iannuzzi A, Sartorio R, Irace
C, Covetti G, Di Costanzo A, Rubba P. Increased intima-media thickness
of the common carotid artery in hypercholesterolemic
children. Arterioscler Thromb. 1994;14:1075–1079.
30.
Wendelhag I, Liang Q, Gustavsson T, Wikstrand J.
A new automated computerized analyzing system simplifies readings and
reduces the variability in ultrasound measurement of intima-media
thickness. Stroke. 1997;28:2195–2200.
31. Smilde TJ, Wollersheim H, Van Langen H, Stalenhoef AFH. Reproducibility of ultrasonographic measurements of different carotid and femoral artery segments in healthy subjects and in patients with increased intima-media thickness. Clin Sci. 1997;93:317–324.[Medline] [Order article via Infotrieve]
32.
O’Leary DH, Polak JF, Wolfson SK Jr, Bond MG,
Bommer W, Sheth S, Psaty BM, Sharrett AR, Manolio TA, for the CHS
Collaborative Research Group. Use of sonography to evaluate carotid
atherosclerosis in the elderly: the
Cardiovascular Health Study. Stroke. 1991;22:1155–1163.
33. Selzer RH, Hodis HN, Kwong Fu H, Mack WJ, Lee PL, Liu CR, Liu CH. Evaluation of computerized edge tracking for quantifying intima-media thickness of the common carotid artery from B-mode ultrasound images. Atherosclerosis. 1994;111:1–11.[Medline] [Order article via Infotrieve]
34. Bots ML, Mulder PGH, Van Es GA, Grobbee DE. Reproducibility of carotid vessel wall thickness measurements: the Rotterdam Study. J Clin Epidemiol. 1994;47:921–930.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  J.-H. Bae, W.-S. Kim, C. S. Rihal, and A. Lerman Individual Measurement and Significance of Carotid Intima, Media, and Intima-Media Thickness by B-Mode Ultrasonographic Image Processing Arterioscler Thromb Vasc Biol, October 1, 2006; 26(10): 2380 - 2385. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. R. Sirtori, L. Calabresi, G. Franceschini, D. Baldassarre, M. Amato, J. Johansson, M. Salvetti, C. Monteduro, R. Zulli, M. L. Muiesan, et al. Cardiovascular Status of Carriers of the Apolipoprotein A-IMilano Mutant : The Limone sul Garda Study Circulation, April 17, 2001; 103(15): 1949 - 1954. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||