US Velocimetry in Participants with Aortoiliac Occlusive Disease
Abstract
Background
The accurate quantification of blood flow in aortoiliac arteries is challenging but clinically relevant because local flow patterns can influence atherosclerotic disease.
Purpose
To investigate the feasibility and clinical application of two-dimensional blood flow quantification using high-frame-rate contrast-enhanced US (HFR-CEUS) and particle image velocimetry (PIV), or US velocimetry, in participants with aortoiliac stenosis.
Materials and Methods
In this prospective study, participants with a recently diagnosed aortoiliac stenosis underwent HFR-CEUS measurements of the pre- and poststenotic vessel segments (August 2018 to July 2019). Two-dimensional quantification of blood flow was achieved by performing PIV analysis, which was based on pairwise cross-correlation of the HFR-CEUS images. Visual inspection of the entire data set was performed by five observers to evaluate the ability of the technique to enable adequate visualization of blood flow. The contrast-to-background ratio and average vector correlation were calculated. In two participants who showed flow disturbances, the flow complexity and vorticity were calculated.
Results
Thirty-five participants (median age, 67 years; age range, 56–84 years; 22 men) were included. Visual scoring showed that flow quantification was achieved in 41 of 42 locations. In 25 locations, one or multiple issues occurred that limited optimal flow quantification, including loss of correlation during systole (n = 12), shadow regions (n = 8), a short vessel segment in the image plane (n = 7), and loss of contrast during diastole (n = 5). In the remaining 16 locations, optimal quantification was achieved. The contrast-to-background ratio was higher during systole than during diastole (11.0 ± 2.9 vs 6.9 ± 3.4, respectively; P < .001), whereas the vector correlation was lower (0.58 ± 0.21 vs 0.47 ± 0.13; P < .001). The flow complexity and vorticity were high in regions with disturbed flow.
Conclusion
Blood flow quantification with US velocimetry is feasible in patients with an aortoiliac stenosis, but several challenges must be overcome before implementation into clinical practice.
Clinical trial registration no. NTR6980
© RSNA, 2021
Online supplemental material is available for this article.
Summary
US velocimetry, or echoPIV, can be used in patients with aortoiliac occlusive disease to detect disturbances in blood flow, which could aid in the diagnosis and treatment of this disease.
Key Results
■ In 35 study participants with aortoiliac stenosis, flow quantification was achieved by using particle image velocimetry in 41 of 42 locations.
■ The contrast-to-background ratio was higher during systole than during diastole (11.0 vs 6.9, respectively; P < .001), whereas the vector correlation was lower (0.58 vs 0.4, P < .001).
■ The flow complexity and vorticity were higher compared with regions with undisturbed flow.
Introduction
Time-resolved quantification of blood flow in diseased aortoiliac regions is challenging because of complex flow patterns near the aortic bifurcation and around stenosis. These flow patterns could be used to improve the assessment of stenosis severity and predict disease progression. For example, blood flow patterns have been correlated with the development and progression of atherosclerotic plaques. Specifically, lesions are more likely to form in areas of low wall shear stress (1,2), which induces major changes in endothelial cells, making the vessel wall more prone to atherosclerosis (3,4).
The peak systolic velocity ratio obtained with duplex US is traditionally applied to quantify the severity of a stenosis. This parameter shows mixed results compared with the reference standard, the invasively measured pressure gradient over the stenosis (5,6). This discrepancy can be explained by the angle dependency of duplex US, which only enables estimation of one-dimensional blood flow velocity along the transducer axis, in a complex anatomic region for which assumptions about flow direction are often inaccurate (7,8).
High-frame-rate contrast-enhanced US (HFR-CEUS) combined with particle image velocimetry (PIV), or US velocimetry (echoPIV), is a technique that enables two-dimensional angle-independent blood flow quantification. EchoPIV could be used to improve and expand the evaluation of lesion severity and to predict atherosclerotic disease progression.
A previous study showed that quantifying blood flow in the aortoiliac region by using echoPIV is feasible in healthy volunteers (9). However, US imaging in patients with atherosclerosis is more challenging because of elongated and calcified arteries. This study aimed to investigate the feasibility and clinical application of quantifying blood flow in two dimensions by using echoPIV in participants with aortoiliac stenosis.
Materials and Methods
Study Design
This prospective study was conducted in accordance with Good Clinical Practice guidelines, approved by an institutional review board (NL63077.091.17), and registered with the Netherlands Trial Register (NTR6980). Thirty-five consecutive participants aged over 50 years with intermittent claudication (Rutherford category of 1–3) based on aortoiliac stenosis were included after providing written informed consent. Participants were excluded from the study if the use of contrast microbubbles was contraindicated. Demographic and clinical data were retrieved from electronic health records (10,11). HFR-CEUS was performed between August 2018 and July 2019 within 1 month after diagnosis. Contrast-enhanced CT scans (section thickness, 0.4 mm) were obtained as an anatomic reference. Agatston calcium scores were calculated by performing automatic calcium segmentation (Intuition, TeraRecon), with a threshold of 600 HU being used to account for the contrast agent–filled vessel lumen (12).
HFR-CEUS Measurements
HFR-CEUS was performed with a Vantage 256 Research US System (Verasonics) and a curved-array transducer (GE C1-6D, GE Healthcare). Before HFR-CEUS examination, blood flow velocities were measured by using duplex US, which was performed with an iU22 US machine (Philips Healthcare).
At each location, two doses of contrast microbubbles (SonoVue, Bracco Imaging) were administered (9). Microbubble arrival was monitored by using the Verasonics system with a live imaging sequence at 100 frames/sec. When a quasistable concentration of contrast agent was visually established, two HFR-CEUS measurements with mechanical indexes of 0.05 and 0.1 were performed, with both having a center frequency of 2.2 MHz and a pulse length of one cycle. Images were captured for 2.5 seconds at 2000 frames/sec with use of a three-angled diverging wave acquisition scheme (pulse repetition frequency, 6000 Hz). Subsequent injections were given after complete washout of the contrast agent on the live images (2–10 minutes after injection). Four measurements were obtained for each location (0.5-mL and 1-mL contrast, each acquired with mechanical indexes of 0.05 and 0.1). This protocol was derived from a previous study (9).
Data Analysis
Data were processed off-line with Matlab (R2019b, MathWorks). Raw HFR-CEUS data were reconstructed into images by using coherent compounding of the three transmit angles, increasing contrast and resolution (13). Clutter suppression was performed by using a singular value decomposition–based filter (14). Rank selection (ie, the cutoff between the tissue and blood and the noise) was performed automatically (15). PIV analysis was performed by using a custom implementation in Matlab consisting of two iterations with a square block size of 5.6 mm and two iterations of 2.8 mm with 75% overlap, resulting in a 0.69-mm vector resolution. To improve the signal-to-noise ratio, correlation averaging was performed over 20 frames, resulting in 100 velocity fields per second. All velocity data and a selection of HFR-CEUS data were stored in a repository that can be accessed (with author permission) at https://www.doi.org/10.4121/c.5497704.
Feasibility Scoring
Qualitative scoring of all velocity data was performed by means of visual inspection by five investigators with expertise in blood flow imaging research (M.M.P.J.R., vascular surgeon with 18 years of experience; E.G.J., assistant professor; J.V., postdoctoral researcher; and S.E. and M.V., PhD candidates). For each location, the best measurement was selected and used to assess feasibility, which was classified according to three categories: unfeasible (no meaningful information could be obtained), partial quantification (blood flow was adequately visualized, but this only occurred during part of the heart cycle or in a subregion of the imaged vessel because one or multiple issues occurred), or optimal quantification (without any limiting issues).
Definitions for limiting issues (Table 1) were discussed and agreed on by all authors. The investigators were then trained in the application of these criteria by scoring a separate data set that was discussed afterward. Feasibility scoring was then performed independently as described above. Disagreements were resolved in a consensus meeting, and the final score was used for further analysis. Interobserver agreement on the feasibility categories was calculated by using the intraclass correlation coefficient (SPSS Statistics 27, IBM), which was based on a mean-rating, absolute-agreement, two-way, mixed-effects model (16).
 |
Temporal velocity profiles were acquired at five vector locations along the centerline of the vessel and were used to automatically select the systolic and diastolic phases. Two feasibility parameters were then calculated for both phases. First, the contrast-to-background ratio was obtained from the HFR-CEUS data by selecting multiple regions within and outside the imaged vessel (Fig 1) and calculating the ratio of the average contrast intensity in those regions. The average normalized cross-correlation value of the velocity vectors was used as a measure of confidence in the velocity vectors, with scores ranging from 0 to 1. Paired t tests were performed to review the difference between systole and diastole. P < .05 was considered indicative of statistically significant difference.
Figure 1: Average of 100 high-frame-rate contrast-enhanced US images after postprocessing. To calculate the contrast-to-background ratio, eight regions are selected inside the vessel (orange squares), and one region is selected above and below the vessel as the background (blue squares). CIA = common iliac artery.
Flow Parameters
The flow complexity (a measure of multidirectional blood flow [17,18]) and vorticity (ie, the curl of the vectors [19]) were calculated in two participants with optimal blood flow quantification who showed a region with disturbed blood flow. Both flow parameters were compared with those from an undisturbed region in the same participant.
Results
Thirty-five participants (22 men; median age, 67 years; age range, 56–84 years) were included (Fig 2, Table 2). HFR-CEUS measurements were obtained in 34 of the 35 participants (in one participant, the stenosis could not be visualized with use of either of the US machines). In eight participants, two separate locations were measured, resulting in 42 locations (Fig 2).
Figure 2: Flow diagram shows the inclusion of study participants and the number of measurement locations. echoPIV = US velocimetry.
 |
Feasibility
The intraclass correlation coefficient was 0.30 (95% CI: 0.21, 0.38). Flow quantification was achieved in 98% of measurements (41 of 42). In 38% (16 of 42 measurements), optimal flow quantification was achieved. In 60% (25 of 42) of measurements, only partial flow quantification was possible because of loss of correlation during systole (n = 12) (Fig 3A, (Movie 1 [online]), short vessel segments (n = 7) (Fig 3B, (Movie 2 [online]), shadow regions (n = 8) (Fig 3C, 3D; (Movie 3 [online]), or loss of contrast during diastole (n = 5) (Fig 3E, 3F; (Movie 4 [online]). Calcium scores in the locations where shadows occurred were higher than in other cases (Table 2).
![US velocimetry images show examples of “partial” flow visualization in four participants. Solid red lines indicate the borders of the vessel, and lines within these borders indicate the flow pattern (arrowheads indicate direction). All US images were obtained by placing the transducer on the patient's abdomen and imaging the long axis of the corresponding vessel. (A) A stenotic lesion (red arrow) in the external iliac artery (EIA) of a 61-year-old man caused very fast and disturbed flow patterns (Movie 1 [online]). These patterns could not be adequately visualized. (B) The left common iliac vein (CIV) crosses the right common iliac artery (CIA) in this 58-year-old man, leaving only short arterial vessel segments in the imaged plane (Movie 2 [online]). (C) Shadows caused by calcifications (yellow and blue arrows) limit contrast intensity, and consequently, the visualization of blood flow in this 69-year-old man (Movie 3 [online]). (D) Contrast-enhanced CT images of the same participant in C (coronal image with transverse sections at the location of the red lines) shows calcifications on the anterior side of the left common iliac artery (yellow and blue arrows). (E, F) Images in an 84-year-old woman with a stenosis in the proximal common iliac artery show poststenotic disturbed flow during systole (lateral position at around –10 mm). During diastole, the contrast microbubbles were destroyed, and flow visualization was not possible (Movie 4 [online]).](/cms/10.1148/radiol.2021210454/asset/images/medium/radiol.2021210454fig3.gif "US velocimetry images show examples of “partial” flow visualization in four participants. Solid red lines indicate the borders of the vessel, and lines within these borders indicate the flow pattern (arrowheads indicate direction). All US images were obtained by placing the transducer on the patient's abdomen and imaging the long axis of the corresponding vessel. (A) A stenotic lesion (red arrow) in the external iliac artery (EIA) of a 61-year-old man caused very fast and disturbed flow patterns (Movie 1 [online]). These patterns could not be adequately visualized. (B) The left common iliac vein (CIV) crosses the right common iliac artery (CIA) in this 58-year-old man, leaving only short arterial vessel segments in the imaged plane (Movie 2 [online]). (C) Shadows caused by calcifications (yellow and blue arrows) limit contrast intensity, and consequently, the visualization of blood flow in this 69-year-old man (Movie 3 [online]). (D) Contrast-enhanced CT images of the same participant in C (coronal image with transverse sections at the location of the red lines) shows calcifications on the anterior side of the left common iliac artery (yellow and blue arrows). (E, F) Images in an 84-year-old woman with a stenosis in the proximal common iliac artery show poststenotic disturbed flow during systole (lateral position at around –10 mm). During diastole, the contrast microbubbles were destroyed, and flow visualization was not possible (Movie 4 [online]).")
Figure 3: US velocimetry images show examples of “partial” flow visualization in four participants. Solid red lines indicate the borders of the vessel, and lines within these borders indicate the flow pattern (arrowheads indicate direction). All US images were obtained by placing the transducer on the patient's abdomen and imaging the long axis of the corresponding vessel. (A) A stenotic lesion (red arrow) in the external iliac artery (EIA) of a 61-year-old man caused very fast and disturbed flow patterns (Movie 1 [online]). These patterns could not be adequately visualized. (B) The left common iliac vein (CIV) crosses the right common iliac artery (CIA) in this 58-year-old man, leaving only short arterial vessel segments in the imaged plane (Movie 2 [online]). (C) Shadows caused by calcifications (yellow and blue arrows) limit contrast intensity, and consequently, the visualization of blood flow in this 69-year-old man (Movie 3 [online]). (D) Contrast-enhanced CT images of the same participant in C (coronal image with transverse sections at the location of the red lines) shows calcifications on the anterior side of the left common iliac artery (yellow and blue arrows). (E, F) Images in an 84-year-old woman with a stenosis in the proximal common iliac artery show poststenotic disturbed flow during systole (lateral position at around –10 mm). During diastole, the contrast microbubbles were destroyed, and flow visualization was not possible (Movie 4 [online]).
Movie 1: Vector velocity data of blood flow in the left external iliac artery of a 61-year-old male, during 4 heart cycles (2.5 seconds, slowed down to 8 seconds). Flow velocity is represented by the color and size of the vectors. Vectors with correlation values < 0.1 are coded red and considered to be erroneous. A stenotic lesion caused very fast and disturbed flow patterns in this participant that could not be adequately visualized (Fig 3A).
Movie 2: Vector velocity data of blood flow in the aortic bifurcation of a 58-year-old male, during 4 heart cycles (2.5 seconds, slowed down to 8 seconds). Flow velocity is represented by the color and size of the vectors. Vectors with correlation values < 0.1 are coded red and considered to be erroneous. The left common iliac vein (middle) crosses the right common iliac artery (right side) in this participant, leaving only short arterial vessel segments in the imaged plane (Fig 3B).
Movie 3: Vector velocity data of blood flow in the left iliac artery of a 69-year-old male, during 3 heart cycles (2.5 seconds, slowed down to 8 seconds). Flow velocity is represented by the color and size of the vectors. Vectors with correlation values < 0.1 are coded red and considered to be erroneous. Shadows caused by calcified lesions in this participant (left side) limit contrast intensity and consequently the visualization of blood flow (Fig 3C).
Movie 4: Vector velocity data of blood flow in the right common iliac artery of a 84-year-old female, during 3 heart cycles (2.5 seconds, slowed down to 8 seconds). Flow velocity is represented by the color and size of the vectors. Vectors with correlation values < 0.1 are coded red and considered to be erroneous. A stenosis in this participant causes poststenotic disturbed flow during systole (left side). During diastole, the contrast microbubbles were destroyed and flow visualization was not possible (Fig 3E).
The contrast-to-background ratio was significantly higher during systole than diastole (11.0 ± 2.9 vs 6.9 ± 3.4, respectively; P < .001). The lowest contrast-to-background ratio values during diastole correspond to the five locations where loss of contrast was observed (Fig 4). The mean correlation values of the velocity vectors were significantly lower during systole (0.58 ± 0.21 vs 0.47 ± 0.13; P < .001), except in those same five locations.
![Contrast-to-background ratio (CBR) and mean vector correlation values during systole (blue) and diastole (green). Boxes indicate upper and lower quartiles, and whiskers indicate highest and lowest values. The five cases categorized as having severe contrast destruction at qualitative scoring are presented as separate markers but were included in the paired t test. ** = For both parameters, there is a significant difference between systole and diastole. The mean contrast-to-background was lower during diastole (6.9 ± 3.4 [standard deviation] vs 11.0 ± 2.9; P < .001), and the mean vector correlation value was higher during diastole (0.58 ± 0.21 vs 0.47 ± 0.13; P < .001).](/cms/10.1148/radiol.2021210454/asset/images/medium/radiol.2021210454fig4.gif "Contrast-to-background ratio (CBR) and mean vector correlation values during systole (blue) and diastole (green). Boxes indicate upper and lower quartiles, and whiskers indicate highest and lowest values. The five cases categorized as having severe contrast destruction at qualitative scoring are presented as separate markers but were included in the paired t test. ** = For both parameters, there is a significant difference between systole and diastole. The mean contrast-to-background was lower during diastole (6.9 ± 3.4 [standard deviation] vs 11.0 ± 2.9; P < .001), and the mean vector correlation value was higher during diastole (0.58 ± 0.21 vs 0.47 ± 0.13; P < .001).")
Figure 4: Contrast-to-background ratio (CBR) and mean vector correlation values during systole (blue) and diastole (green). Boxes indicate upper and lower quartiles, and whiskers indicate highest and lowest values. The five cases categorized as having severe contrast destruction at qualitative scoring are presented as separate markers but were included in the paired t test. ** = For both parameters, there is a significant difference between systole and diastole. The mean contrast-to-background was lower during diastole (6.9 ± 3.4 [standard deviation] vs 11.0 ± 2.9; P < .001), and the mean vector correlation value was higher during diastole (0.58 ± 0.21 vs 0.47 ± 0.13; P < .001).
Flow Parameters
The flow complexity was higher in regions with disturbed flow (Fig 5A). In one of the two analyzed participants, this difference was most pronounced during systole (Fig 5C). Both participants showed a similar increase in flow vorticity during systole in both regions, but vorticity was higher in the regions with disturbed flow.
![(A) US were images obtained by placing the transducer on the patient's abdomen and imaging the long axis of the corresponding vessel. US velocimetry data were obtained during systole in two participants who had flow disturbances at visual inspection (Movies 5 and 6 [online]). Solid red lines indicate the borders of the vessel, and lines within these borders indicate the flow pattern (arrowheads indicate direction). Flow parameters were calculated in a region with undisturbed flow (blue boxes) and in a region with disturbed flow (green boxes). CIA = common iliac artery, EIA = external iliac artery. (B) Temporal velocity profiles acquired at five locations along the centerline of the vessel, showing several heart cycles. Shaded error bars represent the range of measured velocities. (C) Flow complexity and (D) vorticity in the region with undisturbed blood flow (blue) and in the region with disturbed blood flow (green). Left side of A–D: blood flow in the left common iliac artery in a 56-year-old woman. Right side of A–D: blood flow in the left external iliac artery in a 69-year-old man.](/cms/10.1148/radiol.2021210454/asset/images/medium/radiol.2021210454fig5.gif "(A) US were images obtained by placing the transducer on the patient's abdomen and imaging the long axis of the corresponding vessel. US velocimetry data were obtained during systole in two participants who had flow disturbances at visual inspection (Movies 5 and 6 [online]). Solid red lines indicate the borders of the vessel, and lines within these borders indicate the flow pattern (arrowheads indicate direction). Flow parameters were calculated in a region with undisturbed flow (blue boxes) and in a region with disturbed flow (green boxes). CIA = common iliac artery, EIA = external iliac artery. (B) Temporal velocity profiles acquired at five locations along the centerline of the vessel, showing several heart cycles. Shaded error bars represent the range of measured velocities. (C) Flow complexity and (D) vorticity in the region with undisturbed blood flow (blue) and in the region with disturbed blood flow (green). Left side of A–D: blood flow in the left common iliac artery in a 56-year-old woman. Right side of A–D: blood flow in the left external iliac artery in a 69-year-old man.")
Figure 5: (A) US were images obtained by placing the transducer on the patient's abdomen and imaging the long axis of the corresponding vessel. US velocimetry data were obtained during systole in two participants who had flow disturbances at visual inspection (Movies 5 and 6 [online]). Solid red lines indicate the borders of the vessel, and lines within these borders indicate the flow pattern (arrowheads indicate direction). Flow parameters were calculated in a region with undisturbed flow (blue boxes) and in a region with disturbed flow (green boxes). CIA = common iliac artery, EIA = external iliac artery. (B) Temporal velocity profiles acquired at five locations along the centerline of the vessel, showing several heart cycles. Shaded error bars represent the range of measured velocities. (C) Flow complexity and (D) vorticity in the region with undisturbed blood flow (blue) and in the region with disturbed blood flow (green). Left side of A–D: blood flow in the left common iliac artery in a 56-year-old woman. Right side of A–D: blood flow in the left external iliac artery in a 69-year-old man.
Movie 5: Vector velocity data of blood flow in the left common iliac artery of a 56-year-old female, during 3 heart cycles (2.5 seconds, slowed down to 8 seconds). Flow velocity is represented by the color and size of the vectors. A recirculation zone is present, right after a stenotic lesion, where vortical flow occurs during the entire cardiac cycle (Fig 5A, left side).
Movie 6: Vector velocity data in the left external iliac artery of a 69-year-old male, during 2 heart cycles (2.5 seconds, slowed down to 8 seconds). Flow velocity is represented by the color and size of the vectors. A recirculation zone is present in the middle of the vessel segment, where multiple vortices occur during systole (Fig 5A, right side).
Discussion
This study showed that US velocimetry, or echoPIV, in the aortoiliac tract was feasible in 98% of locations. Optimal quantification was achieved in 38% of locations, indicating that the technique needs to be optimized. In participants with optimal quantification, disturbed blood flow patterns could be clearly distinguished from undisturbed blood flow patterns in areas without an increased peak systolic velocity that were not identified as problematic areas at duplex US. Partial quantification was achieved in 60% of locations. Here, blood flow could still be visualized, but this was only possible during part of the cardiac cycle or in a subregion of the imaged vessel.
We previously showed a good match between echoPIV and phase-contrast MRI findings at the aortic bifurcation in healthy volunteers (9). In the current study, echoPIV flow patterns closely matched the movement of the contrast microbubbles on HFR-CEUS images and are therefore assumed to be valid. In most participants, we were not able to obtain reliable Doppler velocity measurements because the maximum angle of 60° was not achieved or because the direction of the flow could not be accurately estimated. Therefore, these data could not be used as a reference. This also confirms the inherent limitations of Doppler imaging due to its angle dependency.
Without the use of duplex US, a reference standard to compare the blood flow velocities measured with echoPIV is lacking. A suitable alternative would have been to use phase-contrast MRI. However, this was not available at our institution during the study.
Despite thorough training, the interobserver agreement was still poor (intraclass correlation coefficient, 0.30), emphasizing the complexity of this unvalidated quality assessment developed in-house. During the consensus meeting, disagreements were mostly caused by different interpretations of the exact cutoff for each limiting issue. Consensus was achieved more easily for each subsequent case, indicating that the interpretation of the scoring criteria converged. Rescoring would therefore likely improve interobserver agreement, but not the scoring method itself.
Optimal blood flow quantification was not achieved in all imaged vessel segments because of several limiting issues. Some of them, including calcifications and out-of-plane blood flow, affect US imaging in general. Loss of contrast (ie, microbubble destruction) is a problem that is exacerbated with the use of HFR-CEUS because of the increased exposure of individual microbubbles to ultrasound waves. Severe bubble destruction occurred in five participants, whereas only minor destruction occurred in a previous study in healthy volunteers for whom similar US intensities were used (9). This could be explained by stagnant blood flow during diastole in patients with atherosclerosis, which does not occur in healthy volunteers. Decreasing the mechanical index further to prevent this destruction would have resulted in an inadequate signal-to-noise ratio. This issue could be addressed by using novel contrast agents that are either more stable during insonification or produce stronger US reflections at a lower mechanical index (20).
In addition, echoPIV requires capturing an entire vessel segment in a single image plane. This is challenging in patients with atherosclerosis, who typically have elongated and curved arteries. Three-dimensional US acquisitions are needed to properly capture these out-of-plane vessels. However, this technique is still in early development (21,22).
In our study, data processing and subsequent flow quantification were performed off-line, without having flow information as feedback to optimize measurement settings and transducer positioning. Translation of echoPIV to daily clinical practice would greatly benefit from real-time flow quantification capabilities, which would require direct data processing (23). This could then be used to search for clinically relevant flow features instead of relying on anatomic features.
Despite these technical limitations, flow disturbances were successfully quantified with use of echoPIV by measuring vector complexity and vorticity. These parameters could be used as an alternative for local wall shear stress values, which currently cannot be calculated accurately with use of echoPIV, and may predict atherosclerotic disease progression. In the future, longitudinal studies with a larger sample size will be needed to show the prognostic value and clinical impact of this technology.
In conclusion, blood flow quantification is feasible by using US velocimetry (echoPIV) in patients with aortoiliac stenosis. Technical challenges—such as microbubble stability, three-dimensional imaging methods, and direct data processing—must be addressed for clinical implementation. Nonetheless, echoPIV already enables acquisition of additional information, including vector complexity and vorticity, that can be used to distinguish disturbed from undisturbed blood flow in regions that are not identified as problematic areas by using duplex US.
Acknowledgments
The authors thank Bastiaan Bongers, Laura Bouwmeister, Jochem Noordzij, Pinel Schrijver, and Frans Tak for their assistance during the high–frame-rate contrast-enhanced US measurements.
Author Contributions
Author contributions: Guarantors of integrity of entire study, S.E., M.M.P.J.R.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, S.E., M.v.H.; clinical studies, S.E., M.v.H., J.V., E.G.J., M.M.P.J.R.; statistical analysis, S.E.; and manuscript editing, all authors
Supported in part by a private charitable organization, the Lijf en Leven Foundation, based in Krimpen aan den IJssel, the Netherlands.
References
- 1. . Correlation between intimal thickness and fluid shear in human arteries. Atherosclerosis 1981;39(3):425–436. Crossref, Medline, Google Scholar
- 2. . Pulsatile flow and atherosclerosis in the human carotid bifurcation: positive correlation between plaque location and low oscillating shear stress. Arteriosclerosis 1985;5(3):293–302. Crossref, Medline, Google Scholar
- 3. . Hemodynamic shear stress and its role in atherosclerosis. JAMA 1999;282(21):2035–2042. Crossref, Medline, Google Scholar
- 4. . The role of shear stress in the generation of rupture-prone vulnerable plaques. Nat Clin Pract Cardiovasc Med 2005;2(8):401–407. Crossref, Medline, Google Scholar
- 5. . The value of the peak systolic velocity ratio in the assessment of the haemodynamic significance of subcritical iliac artery stenoses. Eur J Vasc Endovasc Surg 2001;22(5):424–428. Crossref, Medline, Google Scholar
- 6. . Hemodynamic significance assessment of equivocal iliac artery stenoses by comparing duplex ultrasonography with intra-arterial pressure measurements. J Cardiovasc Surg (Torino) 2018;59(1):37–44. Medline, Google Scholar
- 7. . Measurement of blood flow by ultrasound: accuracy and sources of error. Ultrasound Med Biol 1985;11(4):625–641. Google Scholar
- 8. . Interobserver variability in aortoiliac and femoropopliteal duplex scanning. J Vasc Surg 2001;33(3):540–545. Crossref, Medline, Google Scholar
- 9. . High-frame-rate contrast-enhanced US particle image velocimetry in the abdominal aorta: first human results. Radiology 2018;289(1):119–125. Link, Google Scholar
- 10. . Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg 1997;26(3):517–538. Crossref, Medline, Google Scholar
- 11. . Reporting standards of the Society for Vascular Surgery for endovascular treatment of chronic lower extremity peripheral artery disease. J Vasc Surg 2016;64(1):e1–e21. Crossref, Medline, Google Scholar
- 12. . Low iliofemoral calcium score may predict higher survival after EVAR and FEVAR. Ann Vasc Surg 2020;68(283):291. Google Scholar
- 13. . Coherent plane-wave compounding for very high frame rate ultrasonography and transient elastography. IEEE Trans Ultrason Ferroelectr Freq Control 2009;56(3):489–506. Crossref, Medline, Google Scholar
- 14. . Spatiotemporal clutter filtering of ultrafast ultrasound data highly increases Doppler and fUltrasound sensitivity. IEEE Trans Med Imaging 2015;34(11):2271–2285. Crossref, Medline, Google Scholar
- 15. . High-frame-rate contrast-enhanced ultrasound for velocimetry in the human abdominal aorta. IEEE Trans Ultrason Ferroelectr Freq Control 2018;65(12):2245–2254. Crossref, Medline, Google Scholar
- 16. . A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 2016;15(2):155–163 [Published correction appears in J Chiropr Med 2017;16(4):346.]. Crossref, Medline, Google Scholar
- 17. . In vivo blood velocity vector imaging using adaptive velocity compounding in the carotid artery bifurcation. Ultrasound Med Biol 2019;45(7):1691–1707. Crossref, Medline, Google Scholar
- 18. . Novel flow quantification of the carotid bulb and the common carotid artery with vector flow ultrasound. Ultrasound Med Biol 2014;40(11):2700–2706. Crossref, Medline, Google Scholar
- 19. . Accuracy and precision of a plane wave vector flow imaging method in the healthy carotid artery. Ultrasound Med Biol 2018;44(8):1727–1741. Crossref, Medline, Google Scholar
- 20. . Microbubble agents: new directions. Ultrasound Med Biol 2020;46(6):1326–1343. Crossref, Medline, Google Scholar
- 21. . High-frame-rate echo-particle image velocimetry can measure the high-velocity diastolic flow patterns. Circ Cardiovasc Imaging 2019;12(4):e008856. Crossref, Medline, Google Scholar
- 22. . 4-D echo-particle image velocimetry in a left ventricular phantom. Ultrasound Med Biol 2020;46(3):805–817. Crossref, Medline, Google Scholar
- 23. . Ultrasound open platforms for next-generation imaging technique development. IEEE Trans Ultrason Ferroelectr Freq Control 2018;65(7):1078–1092. Crossref, Medline, Google Scholar
Article History
Received: Feb 22 2021Revision requested: Apr 7 2021
Revision received: June 23 2021
Accepted: June 25 2021
Published online: Aug 24 2021
Published in print: Nov 2021
