Intra-Observer Variability of Longitudinal Displacement and Intramural Shear Strain Measurements of the Arterial Wall Using Ultrasound Noninvasively in vivo

Ultrasound in Med. & Biol., Vol. 36, No. 5, pp. 697–704, 2010 Copyright Ó 2010 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$–see front matter

doi:10.1016/j.ultrasmedbio.2010.02.016

d

Original Contribution INTRA-OBSERVER VARIABILITY OF LONGITUDINAL DISPLACEMENT AND INTRAMURAL SHEAR STRAIN MEASUREMENTS OF THE ARTERIAL WALL USING ULTRASOUND NONINVASIVELY IN VIVO ˚ SA RYDE´N AHLGRENy MAGNUS CINTHIO* and A * Electrical Measurements, Faculty of Engineering, LTH, Lund University, Lund, Sweden; and y Department of Clinical Sciences, Clinical Physiology and Nuclear Medicine Unit, Ska˚ne University Hospital, Lund University, Malmo¨, Sweden (Received 19 May 2009; revised 25 February 2010; in final form 28 February 2010)

Abstract—Using a recently developed high-resolution noninvasive ultrasonic method, we recently demonstrated that the intima-media complex of the common carotid artery show a bidirectional multiphasic longitudinal displacement of the same magnitude as the diameter change during the cardiac cycle. The longitudinal movement of the adventitial region was smaller, thus, we identified shear strain and, thus, shear stress, within the arterial wall. The aim of this study was to evaluate the intra-observer variability of measurement of the longitudinal displacement of the intima-media complex and the intramural shear strain of the common carotid artery in vivo using the new ultrasonic method. The evaluation was carried out by comparing two consecutive measurements on the common carotid artery of 20 healthy human subjects. According to the method of Bland Altman, we show that the systematic and random differences for the different phases of movement are acceptable in comparison to the measured displacement and no significant differences between the two measurements could be detected (p . 0.05 for all measured parameters). The coefficient of variation (CV) for measurement of the different phases of movement was #16%, including short-term physiologic variations. The higher variability in the measurement of the intramural shear strain (CV 5 24%) has several explanations, which are discussed. In conclusion, this study shows that the present first ultrasonic method for high-resolution measurement of the longitudinal movement of the arterial wall is reliable and satisfactory for the further research of the longitudinal movement of the arterial wall in vivo. Further studies on the longitudinal movement of the arterial wall are important for developing an improved understanding of the physiology and the pathophysiology of the cardiovascular system. (E-mail: [email protected]) Ó 2010 World Federation for Ultrasound in Medicine & Biology. Key Words: Carotid artery, Vascular mechanics, Arterial wall movements, Vascular ultrasound, Repetitively.

has been assumed that the longitudinal movement of the arterial wall during the cardiac cycle is negligible compared with the radial movement (Nichols and O’Rourke 1998). Lawton and Greene (1956) made measurements of the longitudinal movement of the abdominal aorta by cinematographic observations of beads sutured to the surface of the vessel. The measured longitudinal movement was very small. The small movements present were considered to be mainly due to respiratory movements of the diaphragm. The results were later confirmed by Patel et al. (Patel et al. 1961; Patel and Fry 1969) on the thoracic aorta in dogs. However, using a new high-resolution noninvasive ultrasonic measurement technique in vivo in humans (Persson et al. 2002, 2003; Cinthio et al. 2005a, 2005b), we recently demonstrated that there is a distinct longitudinal movement of the arterial wall during the cardiac cycle of the same magnitude as the diameter change in central elastic arteries as well as in large muscular arteries

INTRODUCTION In cardiovascular research the radial movement of the arterial wall, the diameter change, has been the subject of extensive research and measurement of the radial movements of arteries is now an established tool in cardiovascular research (Lindstro¨m et al. 1987; Ahlgren et al. 1995; Eriksson et al. 2002; Pannier et al. 2002; Laurent et al. 2006), forming the basis for estimation of arterial wall stiffness. Increased stiffness of large central arteries has recently been shown to be an independent risk factor for cardiovascular mortality (Blacher et al. 1999). In contrast to the radial movement, the longitudinal movement of the arterial wall has gained little attention. It

Address correspondence to: Magnus Cinthio, Electrical Measurements, Faculty of Engineering, LTH, Lund University, Box 118, S-221 00 Lund, Sweden. E-mail: [email protected] 697

698

Ultrasound in Medicine and Biology

(Cinthio et al. 2006). In the common carotid artery, the longitudinal movement shows a bi-directional multi-phasic pattern with a distinct antegrade displacement in early systole, i.e., a movement in the direction of the blood flow, which later in systole is followed by a distinct retrograde displacement, i.e., a movement in the direction opposite to the blood flow. In diastole, a second distinct antegrade longitudinal displacement is seen before the vessel wall gradually returns to its initial position (Fig. 1) (Cinthio et al. 2006). Importantly, we have also demonstrated that the inner parts of the vessel wall, the intimamedia complex, shows a larger longitudinal movement than the outer part of the vessel wall, the adventitial region, introducing the presence of substantial shear strain and, thus, shear stress within the vessel wall (Cinthio et al. 2006). This new information seems to be of great importance for the further study and evaluation of vascular biology and hemodynamic and, thus, for the study of arthrosclerosis and cardiovascular disease. This is further supported by results from our first in vivo trial on the porcine carotid artery indicating that adrenaline might influence the longitudinal displacement of the arterial wall and the resulting shear strain within the arterial wall (Ahlgren et al. 2009). Our findings of a longitudinal displacement of the intima-media complex have recently been supported by computational models (Warriner et al. 2008; Hodis and Zamir 2008). Further, it has been suggested that the longitudinal movement may be used to detect the outer boundary of the arterial wall (Numata et al. 2007).

Volume 36, Number 5, 2010

Ultrasound has been extensively used to measure motion from both blood and tissue (Wells 1969; Baker 1970; Bonnefous and Pesque´ 1986; Trahey et al. 1987; de Jong et al. 1990; McDicken et al. 1992). However, one reason why the longitudinal movement of the arterial wall has been neglected is that prior to the most recent improvements of ultrasound systems, the longitudinal movement has been difficult to detect in vivo. Only a few attempts have been made to measure the velocity or the movement of the arterial wall in both the radial and the longitudinal directions (Sunagawa et al. 2000; Golemati et al. 2003; Cinthio et al. 2005a, 2005b; Numata et al. 2007; Hasegawa et al. 2009). However, at present, as far as we know, our method is so far the only one that has been able to perform high-resolution recordings of the longitudinal movement and to detect and measure shear strain within the arterial wall during several cardiac cycles. The aim of this study was to evaluate the intraobserver variability of the ultrasonic measurement (Cinthio et al. 2005b) of the longitudinal displacement and the intramural shear strain of the human common carotid artery in vivo. MATERIAL AND METHODS Material The measurements were performed on 20 healthy normotensive subjects (10 males; age 25–55 years and 10 females; age 25–49 years). None of the subjects

0.3

Longitudinal movement (mm)

0.2

LMov 1

0.1 LMov 2

0.0

LMov 3

- 0.1

- 0.2

- 0.3

- 0.4

T1 0

T2

T3 0.2

0.4

0.6

0.8

1.0

1.2

Time (s)

Fig. 1. The definition of the analysed parameters. LMov1 5 the antegrade longitudinal displacement of the intima-media complex during early expansion of the artery. LMov2 5 the following retrograde longitudinal displacement of the intimamedia complex. LMov3 5 the antegrade longitudinal displacement of the intima-media complex during diastole. The R-peak of the electrocardiogram (ECG, /start cursive/ bottom trace /end cursive/) was used as reference in the measurement of the timing of the different phases of movement. T1 5 the time of initiation of LMov1 in relation to the R-peak of the ECG. T2 5 the time of initiation of LMov2. T3 5 the time of initiation of LMov3.

˚ . R. AHLGREN Repeatability of longitudinal movement measurements d M. CINTHIO and A

699

reported previous cardiopulmonary disease, diabetes or smoking and all were free from current medication. All subjects gave informed consent according to the Helsinki Declaration and the study was approved by the Ethics Committee, Lund University, Sweden. Ultrasonic measurements of the longitudinal movement of the arterial wall The longitudinal movement of the common carotid artery wall was measured 2 to 3 cm proximal to the bifurcation using B-mode ultrasound. All investigations were performed with a commercial ultrasound system (HDI5000; Philips Medical Systems, Bothell, WA, USA). The system was equipped with a 38-mm 5 to 12 MHz linear array transducer. Standard instrument settings were used but it was desirable to achieve the best possible spatial and temporal resolution. Therefore, the setting using the highest frequency was used in combination with that the area of interest was zoomed with a function called HDZoom. HDZoom changed the scan settings so that only the zoomed area is scanned, which an ordinary zoom does not. Furthermore, only one transmit focus was used and the persistence function was off to avoid averaging between images. To enhance tissue texture, the highest available value of the dynamic range was selected. These settings allowed a frame rate of 55 Hz and a spatial quantification of 52 mm in each direction, i.e., a resolution of 19.2 pixels/mm was achieved. No additional filtering was applied. The image data was transferred to a PC for postprocessing and visualized on the PC in HDILab (ATL Ultrasound; Philips Medical Systems, Bothell, WA, USA), a software designed for off-line cineloop analysis. The cineloop was then exported to Matlab (The MathWorks Inc., Natick, MA, USA), where the algorithm for measurement of the longitudinal movement in reference to the transducer (Cinthio et al. 2005a, 2005b) and the diameter change (Cinthio et al. 2004) was implemented. During the measurements, the vessels were scanned in the longitudinal direction and oriented horizontally in the image (Fig. 2). Prerequisites for the measurements include that the double-line pattern from the boundaries of the lumen-intima and media-adventitia is clearly visible at both the near and the far wall. Furthermore, a distinct echo of an inhomogeneity or an irregularity must be visible in all the images during several cardiac cycles. In addition, to ensure that the recording is performed properly, the longitudinal movement must be visible along the visualized vessel wall segment. All recordings were performed in a quiet room with the subject in the supine position after at least 15 min of rest. All subjects were examined by the same experienced ultrasound technician. The transducer was in all instances placed in a fixative clamp to avoid introducing false

Fig. 2. A scan of the common carotid artery. The vessel was scanned in the longitudinal direction. The boxes indicate the positions of the region-of-interests (ROIs) at the intima-media complex and the advential region, respectively. The kernel used in the echo-tracking algorithm was chosen seven times smaller than the ROI and was automatically positioned in the middle of the ROI. The direction of the blood flow is indicated by the arrow. The longitudinal movement was measured in the lateral direction and the radial movement in the axial direction of the ultrasound beam.

movements by the operator. Care was taken to minimize the pressure of the transducer. Electrocardiogram (ECG) was recorded during all the experiments. Blood pressure was measured at the wrist with an oscillometric sphygmomanometer (BPM Wrist 2300; TOPCOM Europe, Heverlee, Belgium) immediately after measuring the two-dimensional (2-D) arterial wall movements. Three recordings were conducted on each subject and each recording contained three to six cardiac cycles. The transducer was removed after each recording. The two first recordings at each examination were analysed. If one of these recordings did not fulfil the prerequisites mentioned above, the third recording was used. Region-of-interests (ROI), the search regions, were positioned around distinct echoes from the luminalintima interface (the innermost layer of the arterial wall) and the medial-adventitial interface, respectively, of the far wall (Fig. 2). Thereafter, the kernel was automatically positioned in the middle of the ROI (Cinthio et al. 2005a). To obtain as stable of a measurement as possible, it is important that surrounding echoes do not interfere, therefore, the ROI only encircle the chosen echo. However, the size of ROI has to be twice the largest movement occurring between two frames. The size of the ROIs in this study varied between 0.26 3 0.57 and 0.47 3 1.09 mm2. The start and the end of the different phases of the longitudinal movement of each cardiac cycle (Fig. 1) were obtained semiautomatically. The procedure was initiated by defining end-diastole in each cardiac cycle with the help of ECG and the diameter change. The minimum of the movement curve around end-diastole was considered as the beginning of the first antegrade

700

Ultrasound in Medicine and Biology

movement. Thereafter, the two next peaks, i.e., the end of the first antegrade movement and the beginning of the retrograde movement, and the end of the retrograde movement and the beginning of the second antegrade movement, respectively, were defined by the next maximum and minimum positions, respectively. The end of the second antegrade movement was not as distinct as the other three positions. Therefore, another approach was chosen to define it. First, the position of the maximum antegrade velocity of the longitudinal movement after the previous peak was found. The end of the second antegrade movement was thereafter defined by the first position where the velocity of the movement changed direction. The intramural shear strain (in radians) of the arterial wall was obtained from measurements of the longitudinal movement at different depths within the wall, i.e., intimamedia and the adventitial region, and was calculated as follows Shear strain   ðLMovAdv 2LMovAdvd Þ 2 ðLMovIM 2LMovIMd Þ 5 arctan DRd

Volume 36, Number 5, 2010

Altman method, the difference between the two repeated measurements was plotted against their mean. In the same graph, the systematic difference as well as the systematic difference 6 two times the random difference was plotted. The latter shows the 95 %-confident interval. The systematic difference is defined as Systematic difference 5

Statistics The intra-observer variability was evaluated in two different ways: (1) using the method of Bland and Altman (Bland and Altman 1986); and (2) by calculating coefficient of variation (CV). According to the Bland and

ð2Þ;

where di is the difference between the measurements of each subject and n is the number of subjects. The random difference is defined as sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi n 1 X ð3Þ: d2 Random difference 5 2,n i51 i The CV was calculated and defined as rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi n P 1 di2 2,n Random difference i51 5 CV 5 2n P Overall mean 1 mi 2,n

ð4Þ;

i51

ð1Þ;

where LMov is longitudinal movement, Adv is adventitia, IM is intima-media, 6R is radial distance between the ROI of intima-media and adventitia, and d is enddiastole. The absolute value of the maximum shearing angle during the cardiac cycle is presented. Results are presented for the first distinct antegrade displacement in early systole (LMov1), the distinct retrograde displacement (LMov2) and the second distinct antegrade longitudinal displacement (LMov3) (Fig. 1). The magnitude and the timing of the different phases of movement, using the R-peak of the ECG as reference, were evaluated. Further, the maximum intramural shear stain of the arterial wall was evaluated. Moreover, to investigate the repeatability of the relations between the different phases of movement, the quotients between the displacements were also evaluated. Quotient 1 was defined as LMov1/LMov2. Quotient 2 was defined as LMov3/ LMov2. Repeatability was evaluated in two different ways: (1) using the mean value of each parameter of all cardiac cycles in a recording; and (2) using one random cardiac cycle in a recording.

n 1 X di 2n i51

where di is the difference between the measurements of each subject, n is the number of subjects and mi is the measurement. Putative differences between the measurements were also evaluated using paired t-test, or, when non-normal distribution, Wilcoxon signed rank test. The significance level was set to 0.05. RESULTS One man and one woman had no traceable echoes in the recorded cineloops and were, therefore, excluded from the analysis. Another two men and one woman had no traceable echoes in the adventitial region and were, therefore, excluded from the analysis of the intramural shear strain. Figure 3 shows recordings of the longitudinal movement of the intima-media complex of the common carotid artery wall of a 28-year-old female. Comparison using mean value of four to six cardiac cycles The number of cardiac cycles in the recordings was mean 4.6; range 3–6. The distinct antegrade movement (LMov 1) in early systole was mean 355 mm; range 27–885 (Table 1). Figure 4a shows difference against mean for LMov 1. The CV was 14.2% (Table 1). The following distinct retrograde movement (LMov 2) later in systole was mean 2711 mm; range 2292 to 21436 (Table 1). Figure 4b shows difference against mean for LMov 2. The CV was 12.5 % (Table 1).

˚ . R. AHLGREN Repeatability of longitudinal movement measurements d M. CINTHIO and A

a

0.6 0.5

Longitudinal Movement (mm)

The subsequent antegrade movement (LMov3) in diastole was mean 563 mm; range 128 to 887 (Table 1). Figure 4c shows difference against mean for LMov 3. The CV was 16.2 % (Table 1). The maximum shear strain between the intima-media complex and the adventitial region during the cardiac cycle was 0.35; range 0.10 to 0.65 (Table 1). Figure 5 shows difference against mean for maximum shear strain. The CV was 24.0 % (Table 1). Figures for the systematic and random differences are given in Table 1. Further, figures for the time from the R-peak to the initiation of the movements as well as the quotients between the different phases of movement are given in Table 1. No significant differences between the two measurements could be detected (p . 0.05 for all measured parameters).

701

0.4 0.3 0.2 0.1 0 -0.1 -0.2 0

0.5

1

1.5

2

2.5

3

3.5

4

Time (s)

b

0.6

Comparison using one random cardiac cycle The distinct antegrade movement (LMov1) in early systole was mean 361 mm; range 10 to 950 (Table 1). The CV was 13.1% (Table 1). The following distinct retrograde movement (LMov2) later in systole was mean 2712 mm; range 2 282 to 21482 (Table 1). The CV was 13.0 % (Table 1). The subsequent antegrade movement (LMov3) in diastole was mean 557 mm; range 113 to 955 (Table 1). The CV was 19.4% (Table 1). The maximum shear strain between the intima-media complex and the adventitial region was 0.34; range 0.11 to 0.72 (Table 1). The CV was 21.3% (Table 1). Figures for the systematic and random differences are given in Table 1. Further, figures for the time from the R-peak to the initiation of the movements as well as the quotients between the different phases of movement are given in Table 1. No significant differences between the two measurements could be detected (p . 0.05 for all measured parameters). DISCUSSION Using a new ultrasonic noninvasive method (Cinthio et al. 2005a), we, for the first time, recently described and measured the longitudinal movement of the arterial wall and the resulting shearing within the arterial wall (Cinthio et al. 2006). The distinct bi-directional longitudinal displacement of the intima-media complex during the cardiac cycle was of the same magnitude as the diameter change (Cinthio et al. 2006), contrary to what hitherto was presumed in the literature. Furthermore, the longitudinal displacement of the inner layers, i.e., the intima-media complex, was larger than that of the outer layer, i.e., the adventitial region, thereby generating shear strain and, thus, shear stress, within the arterial wall (Cinthio et al. 2006). These phenomena were present in central elastic

Longitudinal movement (mm)

0.5 0.4 0.3 0.2 0.1 0 -0.1 -0.2 0

0.5

1

1.5

2

Time (s)

Fig. 3. (a) An original recording of the longitudinal movement of the intima-media complex of the common carotid artery during six cardiac cycle of a 28-year-old female. (b) Two consecutive recordings focusing on three cardiac cycles (— first recording; - - - second recording). The ECG is shown below as reference. The first R-wave was used to synchronise the two recordings.

arteries as well as in large muscular arteries (Cinthio et al. 2006). In this study, we evaluate the intra-observer variability for measurement of the longitudinal movement and the resulting intramural shear strain of the human common carotid artery wall using the above mentioned method. We compared two consecutive recordings performed the same day by the same experienced technician. We evaluated the magnitude of the distinct phases of movements of the intima-media complex, their timing in relation to ECG, the quotient between the different phases of displacements as well as the intramural shear strain. Given the basic prerequisites of the ultrasound scanner and according to the method of Bland Altman, we show that the systemic and random differences are acceptable compared with the measured displacements (Table 1).

702

Ultrasound in Medicine and Biology

Volume 36, Number 5, 2010

Table 1. Summary of the comparison of two consecutive recordings of the longitudinal movement of the arterial wall. All cardiac cycles

Mean

Sys diff

Rand diff

CV

LMov1 (mm) LMov2 (mm) LMov3 (mm) T1 (ms) T2 (ms) T3 (ms) Quotient 1 (LMov1/LMov 2) Quotient 2 (LMov3/LMov 2) ISS (rad)

355 (SD 218) –711 (SD 275) 563 (SD 238) 96 (SD 22) 195 (SD 35) 428 (SD 25) 0.58 (SD 0.46) 0.79 (SD 0.21) 0.35 (SD 0.15)

9 22 –24 1.4 0.6 0.6 0.04 –0.02 0.02

50 89 91 11.7 12.0 16.3 0.12 0.07 0.08

14.2% 12.5% 16.2% 12.1% 6.3% 3.8% 20.0% 9.0% 24.0%

Mean

Sys diff

Rand diff

CV

361 (SD 230) –712 (SD 278) 557 (SD 241) 100 (SD 26) 200 (SD 35) 430 (SD 28) 0.59 (SD 0.50) 0.78 (SD 0.22) 0.34 (SD 0.17)

4 26 236 3.2 1.0 23.0 0.03 20.02 0.01

47 92 108 18.5 14.4 22.6 0.11 0.15 0.08

13.1% 13.0% 19.4% 18.4% 7.4% 5.3% 18.7% 19.6% 21.3%

One random cardiac cycle LMov1 (mm) LMov2 (mm) LMov3 (mm) T1 (ms) T2 (ms) T3 (ms) Quotient 1 (LMov1/LMov 2) Quotient 2 (LMov3/LMov 2) ISS (rad)

LMov1 5 the antegrade longitudinal displacement of the intima-media complex during early expansion of the artery. LMov2 5 the following retrograde longitudinal displacement of the intima-media complex. LMov3 5 the antegrade longitudinal displacement of the intima-media complex during diastole. The R-peak of the ECG was used as reference in the measurement of the timing of the different phases of movement. T1 5 the time of initiation of LMov1 in relation to the R-peak of the ECG. T2 5 the time of initiation of LMov2. T3 5 the time of initiation of LMov3 ISS 5 maximum absolute intramural shear strain. Mean (SD) 5 mean and standard deviation. Sys diff 5 Systematic difference. Rand diff 5 Random difference. CV 5 the coefficient of variation.

The spatial resolution of the system used is 250 to 300 mm at the best laterally; the distance between two ultrasonic beams is 100 to 200 mm and the smallest detectable movement between two frames is 5 mm (Cinthio et al. 2005a). In a previous phantom study, we evaluated our method using triangulation laser (Cinthio et al. 2005a). The systemic and random differences between the two methods were 3 mm and 20 mm, respectively, in the lateral direction and the repeatability for our method was 24 mm (2 SD). As expected, the variability increases in in vivo measurements, including short-term physiologic variations such as minor differences in blood pressure and heart rate and the fact that the measurements are not performed at the very same position, although this was our intention; during the examinations we aimed to measure at the very same position 2 to 3 cm proximal to the bifurcation. If the measurements are not conducted at the very same position along the arterial wall, this may introduce a variation in the measurements. Further, it cannot be excluded that the longitudinal movement can differ to some extent if the measurements are performed on slightly different positions in the circumferential direction. Also the systematic and the random differences of the time from the R-peak to the initiation of the different phases of movement were acceptable in comparison with the resolution of the measurements (18 ms) and the measured times. The CV for measurement of the different phases of displacement and for the timing of the different phases of movement was #16 % when the mean of three to six cardiac cycles

was evaluated. This can be compared with widely used and accepted methods for ultrasonic measurement of arterial stiffness, which are reported to be around 8% to 20% (Hansen et al. 1993; Kool et al. 1994). The higher variation in the measurement of the maximum intramural shear strain has several explanations. First, the calculation of the shear strain angle is a comparison between two different measurements of the longitudinal movement, i.e., ROIs are positioned at two different locations; in the intima-media complex and the adventitial region, respectively. Further, it is often not possible to position the two ROIs at the very same lateral position along the artery, which introduces a measurement error. In addition, the measurement of the radial distance between the ROIs at end-diastole is limited by the pixel size of the ultrasonic B-mode image. Another source of variation, and the reason why three subjects was excluded from the analysis, is that at present the method is implemented off-line, i.e, during scanning it is not always possible for the sonographer to determine whether the prerequisites for the analysis of the longitudinal movement are completely fulfilled. The limitations described above show that it is desirable to develop improved methods using, e.g., real-time implementation of the algorithm, higher transmit-frequency or RF-data. In this study, the evaluation was carried out in two different ways; using a mean value of all cardiac cycle in a recording, as well as using one random cardiac cycle in a recording. The small nonsignificant differences

˚ . R. AHLGREN Repeatability of longitudinal movement measurements d M. CINTHIO and A

a

300

0.2 0.15

Difference (radians)

200

Difference (µm)

703

100 0 -100

0.1 0.05 0 -0.05 -0.1

-200

-0.15 -300

-0.2 0

100

200

300

400

500

600

700

800

900

Mean (µm)

b Difference (µm)

0.3

0.4

0.5

0.6

Fig. 5. Difference against mean for the measurements of the maximum absolute intramural shear strain of the common carotid artery during the cardiac cycle. The mean of all cardiac cycles in a recording was used in the analysis. The dashed curve shows the systematic difference. The dash-dotted curves show systematic difference 62 random difference, i.e., the 95%-confident interval.

300

100 0 -100 -200 -300 -1400

-1200

-1000

-800

-600

-400

Mean (µm)

300 200

Difference (µm)

0.2

Mean (radians)

200

c

0.1

100 0 -100 -200 -300 100

200

300

400

500

600

700

800

900

Mean (µm)

Fig. 4. Difference against mean for the measurements of the different phases of longitudinal displacement of the intimamedia complex of the common carotid artery. The mean of all cardiac cycles in a recording was used in the analyses. The dashed curve shows the systematic difference. The dash-dotted curves show systematic difference 6 2 random difference, i.e, the 95 %-confident interval. (a) The antegrade displacement during systole (LMov 1). (b) The retrograde displacement (LMov 2). (c) The antegrade displacement during diastole (LMov 3).

between the two methods show that the measurement method works well with low beat-to-beat variation. With the present technique, we cannot with certainty discriminate putative differences in movement of the intima, one cell layer, from the media. Therefore, we have chosen to describe the movement of an echo in the intima-media complex. At the present, we cannot tell if there is shear strain and, thus, shear stress, between the intima and the media, i.e., at the internal elastic lamina, or within the media, or if the major shearing occurs at the external elastic lamina, the demarcation between the media and the adventitia, which visually seems to be the case. With further technical improvements and the use of very high frequency ultrasound we hope in a future to be able to further analyze this. In conclusion, this study shows that the present first method for tracking of a specific echo for measurement of the longitudinal movement of the arterial wall is reliable and satisfactory for the further exploration of the longitudinal movements of the arterial wall. Further studies on the longitudinal movements and the resulting intramural shear strain of the arterial wall are important for developing an improved understanding of the physiology and pathophysiology of the cardiovascular system. Acknowledgements—The authors thank Mrs. Ann-Kristin Jo¨nsson for skilful technical assistance. This study was supported by grants from the Swedish Research Council, the Knut and Alice Wallenberg foundation, the Ska˚ne County Council’s Research and Development Foundation, Funds at Malmo¨ University Hospital, the Medical Faculty, Lund University, the Crafoordska Foundation and the Royal Physiographic Society in Lund.

704

Ultrasound in Medicine and Biology

REFERENCES ˚ R, Cinthio M, Steen S, Persson HW, Sjo¨berg T, Lindstro¨m K. Ahlgren A Effects of adrenaline on longitudinal arterial wall movements and resulting intramural shear strain: A first report. Clin Physiol Funct Imaging 2009;29:353–359. ˚ R, La¨nne T, Wollmer P, Sonesson B, Hansen F, Sundkvist G. Ahlgren A Increased arterial stiffness in women, but not in men, with IDDM. Diabetologia 1995;38:1082–1089. Baker DW. Pulsed ultrasonic Doppler blood-flow sensing. IEEE Trans Sonics Ultrason 1970;(SU-17):170–185. Blacher J, Guerin AP, Pannier B, Marchais SJ, Safar ME, London GM. Impact of aortic stiffness on survival in endstage renal disease. Circulation 1999;99:2434–2439. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–310. Bonnefous O, Pesque´ P. Time domain formulation of pulse-Doppler ultrasound and blood velocity estimation by cross-correlation. Ultrason Imaging 1986;8:73–85. ˚ R, Bergkvist J, Jansson T, Persson HW, Lindstro¨m K. Cinthio M, Ahlgren A Longitudinal movements and resulting shear strain of the arterial wall. Am J Physiol Heart Circ Physiol 2006;291:H394–H402. ˚ R, Jansson T, Eriksson A, Persson HW, Cinthio M, Ahlgren A Lindstro¨m K. Evaluation of an ultrasonic echo-tracking method for measurements of arterial wall movements in two dimensions. IEEE Trans Ultrason Ferroelectr Freq Control 2005a;52:1300–1311. ˚ R, Persson HW, Lindstro¨m K. New Cinthio M, Jansson T, Ahlgren A improved and modified method for measurements of arterial wall movements in longitudinal and radial directions. Proceedings in Biomedizinische Technik supp vol 1. Part II 2005b;50:869–870. Cinthio M, Jansson T, Eriksson A, Persson HW, Lindstrom K. A robust and fast algorithm for automatic arterial lumen diameter measurement with ultrasound. In: ed. Ultrasonic Methods for 2D Arterial Wall Movement Measurements, ISSN 0346-6221:5/04. Lund: Lund Institute of Technology, Lund University, 2004. de Jong PG, Arts T, Hoeks AP, Reneman RS. Determination of tissue velocity by correlation interpolation of pulsed ultrasonic echo signals. Ultrason Imaging 1990;12:84–98. Eriksson A, Greiff E, Loupas T, Persson M, Pesque P. Arterial pulse wave velocity with tissue Doppler imaging. Ultrasound Med Biol 2002;28:571–580. Golemati S, Sassano A, Lever MJ, Bharath AA, Dhanjil S, Nicolaides AN. Carotid artery wall motion estimated from B-mode ultrasound using region tracking and block matching. Ultrasound Med Biol 2003;29:387–399. ˚ , Sonesson B, La¨nne T. Hansen F, Bergqvist D, Mangell P, Ryde´n A Noninvasive measurement of pulsatile vessel diameter change and elastic properties in human arteries: A methodological study. Clin Physiol 1993;13:631–643. Hasegawa H, Kanai H. Phase-sensitive lateral motion estimator for measurement of artery-wall displacement–Phantom study. IEEE Trans Ultrasound Ferroelect Freq Control 2009;56:2450–2462.

Volume 36, Number 5, 2010 Hodis S, Zamir M. Solutions of the Maxwell viscoelastic equations for displacement and stress distributions within the arterial wall. Phys Rev E 2008;78. 021914. Kool MJF, van Merode T, Reneman RS, Hoeks APG, Struijker Boudier HAJ, van Bortel LMAB. Evaluation of reproducibility of a vessel wall movement detector system for assessment of large artery properties. Cardiovasc Res 1994;28:610–614. Laurent S, Cockcroft J, Van Bortel L, Boutouyrie P, Giannattasio C, Hayoz D, Pannier B, Vlachopoulos C, Wilkinson I, StruijkerBoudier H. Expert consensus document on arterial stiffness: Methodological issues and clinical applications. Eur Heart J 2006;27: 2588–2605. Lawton RW, Greene LC. A method for the in situ study of aortic elasticity in the dog. Warminister, PA: Authorized Medical Allowance List, US Naval and Air Development Center; 1956(Rep. No. NADC-MA-5603). Lindstro¨m K, Gennser G, Sindberg-Eriksen P, Benthin M, Dahl P. An improved echo tracker for studies on pulse waves in fetal aorta. Fetal Physiol Meas 1987;217–226. McDicken WN, Sutherland GR, Moran CM, Gordon LN. Colour Doppler velocity imaging of the myocardium. Ultrasound Med Biol 1992;18:651–654. Nichols WW, O’Rourke MF. McDonald’s Blood Flow in Arteries. London: Edward Arnold; 1998. Numata T, Hasegawa H, Kanai H. Basic study on detection of outer boundary of arterial wall using its longitudinal motion. Jpn J Appl Phys 2007;46:4900–4907. Pannier BM, Avolio AP, Hoeks APG, Mancia G, Takazawa K. Methods and devices for measuring arterial compliance in humans. Am J Hypertens 2002;15:743–753. Patel DJ, Fry DL. The elastic symmetry of arterial segments in dogs. Circ Res 1969;24:1–8. Patel DJ, Mallos AJ, Fry DL. Aortic mechanics in the living dog. J Applied Physiol 1961;16:293–299. ˚ R, Eriksson A, Jansson T, Persson HW, Persson M, Ahlgren A Lindstro¨m K. Noninvasive measurement of arterial longitudinal movement. Proc IEEE Ultrason Symp 2002;2:1783–1786. ˚ , Jansson T, Eriksson A, Persson HW, Persson M, Ryde´n Ahlgren A Lindstro¨m K. A new noninvasive ultrasonic method for simultaneous measurements of longitudinal and radial arterial wall movements: First in vivo trial. Clin Physiol Funct Imaging 2003;23:247–251. Sunagawa K, Kanai H, Tanaka M. Simultaneous measurement of blood flow and arterial wall vibrations in radial and axial directions. Proc IEEE Ultrason Symp 2000;2:1541–1544. Trahey GE, Allison JW, Vonramm OT. Angle independent ultrasonicdetection of blood-flow. IEEE Trans Biomed Eng 1987;34:965–967. Warriner RK, Johnston KW, Cobbold RSC. A viscoelastic model of arterial wall motion in pulsatile flow: Implications for Doppler ultrasound clutter assessment. Physiol Meas 2008;29:157–179. Wells PNT. A range gated ultrasonic Doppler system. Med Biol Eng 1969;7:641–652.