Changes in Carotid Artery Flow Velocities After Stent Implantation: A Fluid Dynamics Study With Laser Doppler Anemometry

Transcription

1 J ENDOVASC THER 275 EXPERIMENTAL INVESTIGATION Changes in Carotid Artery Flow Velocities After Stent Implantation: A Fluid Dynamics Study With Laser Doppler Anemometry Oliver Greil, MD; Gottlieb Pflugbeil, MD; Klaus Weigand, PhD*; Wolfgang Weiß, MD; Dieter Liepsch, PhD; Peter C. Maurer, MD; and Hermann Berger, MD Vascular Center, Klinikum rechts der Isar, Technical University of Munich, Germany; and *Laboratory for Fluid Mechanics, University of Applied Science, Munich, Germany Purpose: To study the influence of stent size and location on flow patterns in a physiological carotid model. Methods: Wallstents were positioned in silicon models of the carotid artery at various locations: 2 stents appropriately sized to the anatomy were placed in (1) the internal carotid artery (ICA) and (2) the ICA extending completely into the common carotid artery so as to cover the external carotid artery (ECA) orifice. Another 2 stents were placed in the ICA extending (1) partially and (2) completely into the bulb to simulate stent displacement and disproportion between stent size and the original vessel geometry. Measurements were performed with laser Doppler anemometry (LDA) using pulsatile flow conditions (Reynolds number 250; flow L/min; ICA:ECA flow rate ratio 70:30) in hemodynamically relevant cross sections. The hemodynamic changes were analyzed with 1-dimensional flow profiles. Results: With the stent in the ICA, no changes of the normal flow profile were seen. For stents positioned in the ICA and extending partially or completely into the carotid bulb, the flow behavior was affected by the resistance of the stent to flow in the ECA. Hemodynamically relevant disturbances were seen in the ICA and ECA, especially in the separation zones (regions along the walls just after a bifurcation, bend, or curve). The ICA:ECA flow rate ratios shifted from 70:30 to 71.3:28.7 and from 70:30 to 75.1:24.9, respectively, in the 2 malpositioned stent models. With the stent placed in the ICA extending completely into the CCA, the ICA:ECA flow rate ratio shifted from 70:30 to 72.4:27.6. In this configuration, there were no notable flow changes in the ICA, but a clear diminishing of the separation zones in the ECA separation zones. Conclusions: Anatomically correct positioning of appropriately sized stents does not lead to relevant flow disturbances in the ICA. In the ECA, depending on the position, size, and interstices of the stent, the physiological flow was considerably disturbed when any part of the stent covered the inflow of the vessel. Disturbances were seen when the stent was positioned into the bulb. For clinical application, stent location and size must be carefully determined so that the stent covers the bifurcation completely or is in the ICA only. J Endovasc Ther Key words: internal carotid artery, external carotid artery, stent, flow velocity, carotid flow model, fluid dynamics, laser Doppler anemometry, atherogenesis Address for correspondence and reprints: Oliver Greil, MD, Department of Interventional Radiology, Klinikum rechts der Isar der Technischen Universität München, Ismaningerstraße 22, München, Germany. Fax: ; greil.oliver@gmx.de 2003 by the INTERNATIONAL SOCIETY OF ENDOVASCULAR SPECIALISTS Available at

2 276 FLOW VELOCITIES IN CAROTID STENTS J ENDOVASC THER Carotid artery stenosis, particularly involving the origin of the internal carotid artery (ICA), is a frequent clinical problem. Patients with ICA stenoses 75% are at risk for stroke at a rate of 2% to 5% within the first year of observation. 1,2 If invasive treatment is necessary, carotid endarterectomy has been firmly established as the gold standard. 1,2 Over the past few years, a great deal of interest has been generated with regards to treating the extracranial carotid stenosis with balloon angioplasty with and without stenting. 3,4 Local hemodynamic factors, such as low wall shear stress, oscillations in shear directions, and long particle residence times appear to have significant impact on the location in which atherosclerotic lesions develop The angle of a bifurcation and the flow division between the parent vessel and branches both affect the local flow profile. Separation from unidirectional laminar flow occurs mainly at a flow divider and distal to stenoses. In such locations, secondary flow patterns tend to form with complex flow profiles, such as vortices and recirculation zones. 6 Plaque formation often occurs in such areas of flow separation, particularly at the carotid bifurcation. The purpose of this study was to investigate the influence of stent size and location on blood flow behavior in a physiological model of the carotid artery using laser Doppler anemometry (LDA) to measure velocities. METHODS Models and Model Fluid True-to-scale transparent silicon rubber models of the common carotid artery (CCA) with an angle of 37 between the ICA and the external carotid artery (ECA) were moulded from casts of human carotid arteries made at autopsy, as described in Liepsch et al. 13 The carotid arteries were pressurized to 60 to 100 mmhg prior to casting. These models had identical geometry and similar compliance as the original vessel; their texture also mimicked biological intima. 13 Four different Wallstents (Boston Scientific International, Paris, France) were positioned at different locations in these physiological carotid models. Stent I Figure 1 (A) Stent I (7 15-mm Wallstent) was placed entirely in the ICA. (B) Stent II (6 15-mm Wallstent) was placed in the ICA and into the bulb of the carotid artery to simulate displacement owing to disproportionate stent size. (C) Stent III (9 25-mm Wallstent) was placed in the CCA and ICA. (D) Stent IV (6 18-mm Wallstent) was placed in the same location as Stent I, but positioned 2 mm back in the bulb to simulate dislocation after implantation of an improperly sized stent. (7-mm diameter) was placed entirely in the ICA (Fig. 1A), while Stent II (6-mm diameter) was positioned in the ICA and in the bulb (Fig. 1B). Stent III (9-mm diameter) was placed in the CCA and ICA, spanning the ECA (Fig. 1C); Stent IV (6-mm diameter) was malpositioned 2 mm back into the bulb (Fig. 1D). Stents II and IV were smaller in size than the original vessel diameter and were positioned unfavorably to simulate stent displacement and size disproportion between the stent and native vessel. To simulate laser light absorption by the hemoglobin in the red cells, a blood-like model fluid was necessary. A dimethyl sulfoxide (DMSO) and water solution with polyacrylamides (Seperan) added served as this medium; it was transparent, had the same refractive index as the model wall, and a density of 1050 kg/m 3. Titanium oxide par-

3 J ENDOVASC THER FLOW VELOCITIES IN CAROTID STENTS 277 ticles (1- m diameter) were added to the solution to act as tracers for LDA measurements. The representative viscosity of the fluid was 4.9 mpa s. 14 The elastic and viscous components were also similar to those of blood. Although stable for 3 weeks, this fluid had to be handled carefully. Experimental Design To study flow in the carotid bifurcation, especially in the bifurcation zones, pulsatile flow conditions are necessary Using an experimental circulatory system, physiological flow was simulated with a computer-controlled piston pump that created pulsatile flow. Fluid from an upstream tank was pumped to a straight tube almost 1.5 m long to achieve fully developed laminar flow at the entrance of the bifurcation. The pulsatile waveform of a healthy human carotid artery, which was obtained by noninvasive ultrasound Doppler velocimetry in a 25-year-old man, was used to drive the piston pump. A mean flow of L/min (Reynolds number [Re] 250) with a frequency of 60 pulses/min was measured with an inductive electromagnetic flowmeter 50 mm in front of the CCA. The pressure was measured 30 mm upstream of the model with inductive pressure transducers. The flow, pressure, and velocity settings used in the experimental circuit are shown in Figure 2. The ICA:ECA flow rate ratio, which was regulated with 2 moving flow regulation containers, was set to 70:30. The model was installed in a Plexiglas container and embedded in a glycerin water solution that had the same refraction index as the model wall. The measuring volume of the 1-dimensional (1D) laser system was exactly positioned when the entire model within the Plexiglas box was mounted on an x-y-z moving table. Velocity was measured with a 5-mW laser Doppler anemometer system (BBC Goerz; Spectrophysics, Munich, Germany) with a wavelength of nm. LDA does not disturb the flow and has a high spatial resolution and a fast linear response; the system is not affected by temperature, pressure, or fluid density. For the laser measurements, the entire system, including model container, container fluid, model, and model fluid, must be Figure 2 Experimental setup with flow (A), pressure (B), and velocity (C) settings. transparent and have the same refractive index. For the calculation of identical measurement planes, the apex of the bifurcation was defined as reference point X 0 (Fig. 3A). LDA measurements were performed in 14 cross sections in the CCA, ICA, and ECA (Fig. 3A), each cross section having 69 defined measuring points (Fig. 3B). The axial velocity component was measured at each position over 8 cycles of the piston pump. For every pulse cycle, 100 measurements were performed, and a mean velocity profile from the 800 measurements per measuring point was calculated (Fig. 2C). Each pulse cycle was divided into 360. For the analysis of the flow profile, the

4 278 FLOW VELOCITIES IN CAROTID STENTS J ENDOVASC THER Figure 3 (A) Cross sections measured with laser Doppler anemometry (LDA) in the CCA, ICA, and ECA. (B) LDA measuring points per cross section. (C) For the entire cross section, a profile of the axial velocity components can be calculated for any point of time in the pulse cycle. Here, the flow velocity profile in the CCA 15 mm proximal to the flow divider (X 0 )at peak systolic velocity (Phase 60 ) is shown. Outer wall 1 (OW1) is continued by the ICA. Proximal to the stents in the CCA, typical laminar flow was found. The velocity profiles in the CCA of all models were similar. L left, U/Um velocity/mean velocity. systolic peak velocity at 60 and the minimum diastolic velocity at 120 were compared with the velocities in the model without a stent. For the entire cross section, a velocity profile of the axial velocity components (Fig. 3C) can be calculated for any time point in the pulse cycle. Definitions Flow separation is a rheological term describing the tendency for fluid slipstreams to continue in a straight line as flow enters a bend, curve, or bifurcation. If the velocity is high enough or the radius of the curvature is small enough, the slipstreams break away or separate from the lesser curvature wall. 15,16 The region along the lesser curve is called a separation zone, which typically contains fluid that is moving sluggishly, swirling, or even sometimes moving retrograde. In bifurcations, the slipstreams pass toward the wall adjacent to the carina and, because fluid is incompressible, velocity increases, whereas in the separation zone at the outer wall, low shear stresses and, sometimes, reversed flow may be found. 15,16 These flow characteristics of a separation zone are portrayed by a typical armchair profile (Fig. 4A) with high velocities at the wall adjacent to the carina and low velocities at the opposite side. High velocity shear gradients are found in the border between low and high velocities. RESULTS In the reference model without a stent (Fig. 4A), the separation zone in the ICA displayed the typical armchair profile with low profiles at the outer wall and decreased velocities in the central slipstreams. At the inner wall, high velocities with laminar flow profiles were observed. The ICA:ECA flow rate ratio in the Stent I model (Fig. 1A) remained at 70.9:29.1, largely unchanged from the values found without the stent (70.3:29.7). Proximal to the

5 J ENDOVASC THER FLOW VELOCITIES IN CAROTID STENTS 279 Figure 4 Flow velocity profiles in the ICA (F) 5 mm distal to the flow divider (X 0 ) at peak systolic velocity (Phase 60 ). Without Stent (A), the typical armchair profile of the separation zone in the ICA was clearly appreciated: low profiles at the outer wall (OW) with decreased velocities in the central slipstreams. At the inner wall, high velocities with laminar flow profiles were seen. Stent I (B) and Stent III (D) showed similar flow profiles without any hemodynamically relevant changes in the separation zone, whereas Stent II (C) and stent IV (E) displayed significant changes with high velocities in the center and low velocities in the separation zone. Every profile was measured in the stent; however, there were some incorrect measurements as the mesh grid of the stent interfered with the laser beam, especially in stent I at the marginal measuring points. L left, U/Um velocity/mean velocity. stent, in the separation zone, there was a systolic velocity increase to 0.24 m/s, while during the diastolic phase, a minimal velocity decrease was apparent. The velocity within the stent increased up to 0.19 m/s during the systolic phase (Fig. 5A); otherwise, the velocity profile remained normal. The separation zone can be clearly seen (Fig. 4B). Ten millimeters within the stent, the profile returned to normal; while at the end of the stent, a 0.1-m/s velocity increase was seen. Slight disturbances were seen behind the stent struts in the marginal measuring points. The central slipstreams showed a similar flow profile as in the model without a stent. Five and 10 mm distal to the stent, there was a velocity decrease toward the inner wall. In contrast to the ICA, slightly decreased velocities were found in the separation zone of the ECA (Fig. 6B and 7A). There were no other differences.

6 280 FLOW VELOCITIES IN CAROTID STENTS J ENDOVASC THER Figure 5 Flow velocity profiles 5 mm distal to the flow divider in the ICA (shown in Fig. 4F) at peak systolic velocity (Phase 60 ) near the outer wall 1 (OW1) in the separation zone. Similar profiles were seen in stents I (A) and III (C), but with slightly increased velocities in stent I (incorrect measurement is marked with a circle); whereas, strong disturbances were seen in stents II (B) and IV (D), with high velocities in the central slipstream. Stent II (Fig. 1B) simulated displacement and disproportion between stent size and the original vessel geometry. The ICA:ECA flow rate ratio changed from 70.3:29.7 in the model without a stent to 75.1:24.9 in Stent II. The peak velocities were 0.46 m/s higher where the stent was located, and the velocity was higher through the entire pulse cycle (Figs. 4C and 5B). Velocities close to the wall distal to the stent were lower compared to the velocities of the model without a stent. The velocity profiles were altered 5 mm and 10 mm distal to the stent and included small vortices. Fifteen millimeters distal to the stent, the flow returned to normal. High velocity disturbances were seen in the ECA (Fig. 6C) and remained up to 20 mm downstream of the apex. The velocity in the local separation region (Fig. 7B) was 0.41 m/s higher than in the model without a stent. No negative velocities were found. Retrograde flow found in the bulb of the unstented model disappeared in the model with a stent. After the measurements, the ICA:ECA flow rate ratio was turned back to 70:30, and the model was measured again to see if the shift in the flow rate ratio had an influence on the flow profiles. There were no relevant changes in the flow behav-

7 J ENDOVASC THER FLOW VELOCITIES IN CAROTID STENTS 281 Figure 6 Flow velocity profiles 2.5 mm distal to the flow divider (F) in the ECA at peak systolic velocity (Phase 60 ). (A) Without Stent, the separation zone was clearly seen. Stent I (B) had a similar flow profile as the model without a stent. (C) Stent II showed strong disturbances at the outer (OW) and inner walls. (D) Stent III had a similar profile as stent II, but not as strong. (E) Stent IV had strong disturbances at the inner wall. L left, U/Um velocity/ mean velocity. ior, just a slight velocity decrease in the ICA and a slight velocity increase in the ECA due to the altered the flow rate ratio in both vessels. For Stent III (Fig. 1C), the flow rate ratio changed to 72.4:27.6 (reference 70.3:29.7). In this configuration, the stent acted as a large mesh grid. The stent grid increased the resistance, especially at the entrance of the ECA. Inside the stent (15 mm proximal to and 10 mm distal from the bifurcation), the marginal velocity was somewhat lower, with a velocity decrease of 0.20 m/s was found. In the separation zone, the systolic values were increased up to 0.27 m/s, whereas the diastolic velocities were slightly decreased. There were no other noticeable changes in the ICA (Figs. 4D and 5C), but in the ECA, an increase of the velocities in the separation zones to values as in an almost laminar flow profile (Figs. 6D and 7C) were seen, with flow disturbances at the outer and inner walls, as in Stent II model. In contrast to Stent II, the velocity profile showed normal flow behavior at a distance of 20 mm distal to the stent. Stent IV (Fig. 1D) was purposefully malpositioned 2 mm back into the bulb, which led to significant disturbances in flow behavior, especially at the inner wall of the ECA and the outer ICA wall. The ICA:ECA flow rate ratio

8 282 FLOW VELOCITIES IN CAROTID STENTS J ENDOVASC THER Figure 7 Flow velocity profiles 2.5 mm in the ECA at peak systolic velocity (Phase 60 ) near the outer wall (OW) of the ECA (F). Stent I (A) had decreased velocities in the separation zone. Stents II (B) and III (C) had high velocity increases in the separation zone. Stent IV (D) near the outer wall had almost no alterations, whereas near the inner wall directly behind the stent (E), high velocity fluctuations occurred. (G) Measuring plane near the inner wall in Stent IV. changed from 70.3:29.7 to 71.3:28.7. Within the stent (Fig. 4E), the central slipstream was accelerated, whereas the marginal velocities tended to be somewhat lower compared to the model without a stent. In the separation zone (Fig. 5D), the typical armchair profile, with high velocities near the wall and low velocities in the center, was disturbed. High velocities were found in the central slipstreams with low velocities in the separation zone. Distal to the stent, the flow was disturbed by the stent grid, which was not attached to the vessel wall. In the ECA (Figs. 6E and 7E), there were high velocity fluctuations near the inner wall up to 0.66 m/s, whereas toward the outer wall (Fig. 6E and 7D), almost no changes occurred. Fifteen millimeters distal in the ECA, the flow profile returned to normal. DISCUSSION Earlier studies have shown that the flow and the flow rate ratio exert a major influence on the velocity distribution in bifurcations. Higher Reynolds numbers (Re 350) lead to an increase in flow separation, with negative velocities in the flow separation regions, and to higher velocity fluctuations and higher velocity shear gradients. Lower Reynolds numbers (Re 200), on the other hand, show no negative velocities. 19 Our studies were performed at Re 250. The influence of the bifurcation angle on the flow separation regions in the ECA and ICA was smaller than that produced by the flow rate ratio, 17 but effects of vessel geometry should not be neglected. Particles, or cells, remain sometimes over several cycles in such recirculation zones and are under constant high and low shear stress. The velocity gradients are very high in such regions between the forward and backward flow, reaching shear rates around 3600 to 4000 s 1 at Re 250 (shear stresses around 20 Pa). 19 Thus, the cell membranes of thrombocytes and endothelial cells are often subjected to enormous stress changes. An angiographic study 20 showed mean durations of the flow

9 J ENDOVASC THER FLOW VELOCITIES IN CAROTID STENTS 283 separation phenomenon from 5 seconds up to 14 seconds. In these separation regions, particles may stick more easily to the wall and to existing stenoses because substrate exchange, thrombocyte activity, and endothelial surface structure are pathologically altered In most clinical applications, the ECA has to be bridged because atherosclerotic lesions are oftentimes located at the entrance of the ICA. When the stent is placed across the bifurcation, it must adapt itself to different diameters: the internal diameter of the ICA varies from 5 to 7 mm (mean 5.7 mm), while the CCA varies from 7 to 10 mm. 3 The stent should have close contact to the vessel wall, otherwise, the flow is disturbed by the stent struts. Improperly positioning a stent and misjudging the stent size may lead to relevant disturbances in the flow behavior. Especially in the situation where stents are misplaced and too small, the central slipstreams are accelerated, while the velocities in the separation zones are decreased. Stents II and IV presented an unfavorable flow pattern: they caused flow acceleration in the ICA. The smaller 6-mm diameters of Stents II and IV compared to the 7-mm and 9-mm stents in models I and III, respectively, meant that the interstices were closer together. The proximal end of the stents protruded into the vessel lumen, which strongly affected the separation zones locally and distally beyond the stent. In the ECA, just behind the stent, dead zones occurred, producing an increased velocity shear gradient in this area. As a consequence, the hemodynamic effects were increased in the separation zone, areas predisposed to atherosclerotic lesion development. Clinically, displacement and disproportionate stent size could lead to hemodynamic alterations that might favor angiogenesis. In contrast, correctly placed and well-adapted stents produce no relevant alterations. The flow pattern in Stent I showed no variations from physiological flow in the carotid artery. When a well-adapted ICA stent crosses the ECA (Stent III), the effects in the separation zones are diminished, but the flow in the ICA remains undisturbed, probably because the stent somewhat reduces the angle between the CCA and ICA. In the ECA, the LDA measurements demonstrated a calming of the flow. However, such straightening of a vessel could lead to severe elongation, kinking, or even stenosis distal to the stent after the procedure. Therefore, in clinical practice, the anatomy of the entire vessel from the CCA to the distal ICA has to be checked prior stent implantation. If any elongation is found, a surgical procedure should be considered instead of stent implantation. In this experimental study, only the hemodynamic alterations after stent implantation at the carotid bifurcation were investigated. This is just one important factor that impacts the flow characteristics in this region. Many questions remain regarding the influence of the stent on vessel wall elasticity, the interaction of blood and the endothelial cell layer, and blood clotting reactions. Moreover, in clinical settings, there are further important factors, such as cardiac disease; the degree, nature, and location of the plaque; the presence of calcification and/or ulceration; the presence of intimal flaps beyond the stent; and plaque protruding through the interstices of the stent. In clinical application, the stent location and size have to be determined carefully so that the stent completely covers the bifurcation or is restricted to the ICA. If stents cannot be placed exactly at the flow divider, then a flexible, large-mesh stent should be placed across the ECA, provided that correct stent position and configuration can be achieved. Inappropriately sized stents and positioning the stent into the bulb must be avoided, as that clearly alters flow behavior and may promote atherogenesis. References 1. Moore WS, Barnett HJM, Beebe HG, et al. Guidelines for carotid endarterectomy. Circulation. 1995;91: Biller J, Feinberg WM, Castaldo JE, et al. Guidelines for carotid endarterectomy. A statement for healthcare professionals from a Special Writing Group of the Stroke Council, American Heart Association. Circulation. 1998;97: Mathias KD. Angioplasty and stenting for carotid lesions: an argument for. Adv Surg. 1999; 32: Yao JS. Angioplasty and stenting for carotid le-

10 284 FLOW VELOCITIES IN CAROTID STENTS J ENDOVASC THER sions: an argument against. Adv Surg. 1999;32: Ku DN, Giddens DP. Hemodynamics of the normal human carotid bifurcation: in vitro and in vivo studies. Ultrasound Med Biol. 1985;11: Glagov S, Zarins C, Giddens DP, et al. Hemodynamics and atherosclerosis. Insights and perspectives gained from studies of human arteries. Arch Pathol Lab Med. 1988;112: Ma P, Li X, Ku DN. Connective mass transfer at the carotid bifurcation. J Biomechan. 1997;30: Zhao SZ, Xu XY, Hughes AD, et al. Blood flow and vessel mechanics in a physiologically realistic model of a human carotid arterial bifurcation. J Biomechan. 2000;33: Davies PF. Flow-mediated endothelial mechanotransduction. Physiol Rev. 1995;75: Davies PF. Mechanisms involved in endothelial responses to hemodynamic forces. Atherosclerosis. 1997;131Suppl:S Davies PF. Overview: temporal and spatial relationships in shear stress-mediated endothelial signaling. J Vasc Res. 1997;34: Barbee KA, Davies PF, Lal R. Shear stress-induced reorganisation of the surface topography of living endothelial cells imaged by atomic force microscopy. Circ Res. 1994;74: Liepsch D, Moravec S, Baumgart R. Some flow visualization and laser Doppler-velocity measurements in a true-to-scale elastic model of a human aortic arch A new model technique. Biorheology. 1992;29: Liepsch D, Thurston G, Lee M. Studies of fluids simulating blood-like rheological properties and applications in models of arterial branches. Biorheology. 1991;28: Kerber CW, Liepsch D. Flow dynamics for radiologists. I. Basic principles of fluid flow. AJNR Am J Neuroradiol. 1994;15: Kerber CW, Liepsch D. Flow dynamics for radiologists. II. Practical considerations in the live human. AJNR Am J Neuroradiol. 1994;15: Liepsch DW, Pflugbeil G, Maurer PC, et al. LDA measurements in anatomically distensible carotid artery models under physiological conditions. In: Liepsch DW, ed. Biofluid Mechanics. Proceedings of the 3rd International Symposium. Düsseldorf: VDI Verlag; 1994: Kerber CW, Hielman CB. Flow dynamics in the human carotid artery: I. Preliminary observations using a transparent elastic model. AJNR Am J Neuroradiol. 1992;13: Liepsch D, Pflugbeil G, Matsuo T, et al. Flow visualisation and 1- and 3-D laser-doppler-anemometer measurements in models of human carotid arteries. Clin Hemorheol Microcirc. 1998;18: Imbesi SG, Kerber CW. Why do ulcerated atherosclerotic artery plaques embolize? A flow dynamics study. AJNR Am J Neuroradiol. 1997;19: