Ann Thorac Surg 2009;87:1866-1871. doi:10.1016/j.athoracsur.2009.03.054
© 2009 The Society of Thoracic Surgeons
Original Articles: Adult Cardiac
Flow Velocity and Turbulence in the Transverse Aorta of a Proximally Directed Aortic Cannula: Hydrodynamic Study in a Transparent Model
Ikuo Fukuda, MD, PhDa,*,
Shingo Fujimori, MScTb,
Kazuyuki Daitoku, MD, PhDb,
Hideki Yanaoka, Dr Engb,
Takao Inamura, Dr Engb
a Thoracic and Cardiovascular Surgery, Hirosaki University Graduate School of Medicine, Hirosaki, Aomori, Japan
b Department of Intelligent Machines and System Engineering, Faculty of Science and Technology, Hirosaki University, Hirosaki, Aomori, Japan
Accepted for publication March 17, 2009.
* Address correspondence to Dr Fukuda, Thoracic and Cardiovascular Surgery, Hirosaki University Graduate School of Medicine, 5-Zaifucho, Hirosak, 036-8562, Japan (Email: ikuofuku{at}cc.hirosaki-u.ac.jp).
 |
Abstract
|
|---|
Background: The objective of this study was to visualize and characterize the effect of cannula tip direction on flow within transverse aortic arch.
Methods: A hydrodynamic analysis of the Dispersion arterial cannula (Edwards Lifescience LLC, Irvine, CA) was performed using particle image velocimetry in glass perfusion models of healthy and aneurysmal aortic arches. Flow velocity, streamline, distribution of magnitude of the strain rate tensor (function of shear stress), and degree of flow turbulence were comparatively analyzed for cannula tip directed toward the aortic arch (standard direction) and toward the aortic root (root direction).
Results: Standard direction cannulation in the model of the healthy aorta showed the flow velocity in the transverse aortic arch was rapid, the streamlines were nonlinear, and the magnitude of the strain rate tensor was high along aortic curvatures. Conversely, directing the cannula tip toward the aortic root generated slower and less turbulent flow in the transverse aortic arch despite high velocity and turbulence and nonlinear streamlines in the ascending aorta. In the aneurysmal aortic arch model, the flow velocity was more rapid in the area where aortic arch vessels originated, and a reversely directed vortex was observed between the aneurysm and the origination of the arch vessels. In the root direction model, the flow velocity distribution was slower than that in the standard direction.
Conclusions: Directing the cannula tip of the Dispersion cannula toward the aortic root generated slower and less turbulent flow in the transverse arch of the glass models of both healthy and aneurysmal aortic arches.
|
Introduction
|
|---|
Advances in surgical technique and cardiopulmonary bypass (CPB) technology have reduced mortality and the incidence of complications associated with cardiac operations; however, stroke still remains a major complication. Among the various causes of stroke associated with cardiac interventions, atheroembolism from the diseased aorta is an unsolved and emerging problem [1–3]. Breakdown of thickened posterior aortic wall caused by aortic clamping has been well documented [4, 5]; however, cerebral embolism occurs even when the ascending aorta appears healthy.
It is theorized that the sandblast effect of fluid dynamics, caused by rapid flow from aortic cannulas, may cause atheroembolism in patients who have atherosclerosis of the transverse aortic arch. A case of broken aortic atheroma due to jet from an arterial perfusion cannula was reported [6]. The Dispersion cannula (Edwards Lifescience LLC, Irvine, CA, Fig 1) was designed with a flat reflector to provide a fan-shaped flow, thus reducing the flow velocity [7]. Grooters and colleagues [8] reported efficacy of directing the Dispersion cannula toward the aortic valve to reduce atheroembolism during coronary artery bypass grafting. The objective of this study was to elucidate the rheology of the arterial cannula tip and effect of cannula tip direction.
View larger version (54K):
[in this window]
[in a new window]
|
Fig 1. Dispersion cannula (model DCT21A, Edwards Lifescience LLC, Irvine, CA). Flow from the cannula tip creates a fan-shaped flow. (A) Frontal view; (B) lateral view in glass aortic model.
|
|
|
Material and Methods
|
|---|
Aortic Arch Model
Two aortic glass models were used in this experiment. The glass model of the healthy aorta was constructed from an enhanced computed tomography (CT) image of an adult without cardiovascular disease. The aneurysmal aortic glass model was constructed from an enhanced CT of a patient who had an atherosclerotic transverse aortic arch aneurysm. A detailed method of the construction of the glass models was described in a previous report [9]. To simulate aortic perfusion, water at room temperature was perfused into the aortic model by a centrifugal pump (HAP-21, NIKKISO Co, Ltd, Tokyo, Japan). A cannulation site with an 8.5-mm diameter hole was created on the glass model of the ascending aorta 4 cm proximal to the origin of the simulated brachiocephalic artery (BCA). The diameters of the BCA, left common carotid artery, left subclavian artery, and descending aorta were 8, 7, 7, and 15 mm, respectively.
The perfusion rate was set at 4.0 L/min of steady flow. The perfusion pressure was monitored and maintained between 60 and 80 mm Hg by compressing the tube that collected the water from the three arch branches. Water from the three branches was drained into the supply reservoir to be recirculated. Approximately 15% to 20% of total flow went into the arch vessels under this condition.
The flow of the Dispersion cannula (model DCT21A) was analyzed in two directions. One was standard aortic cannulation directing the cannula tip toward the aortic arch (standard direction) and the other was modified cannulation with the cannula tip directed toward the aortic root (root direction). Flow in the healthy aortic model and the aneurysmal aortic model was evaluated.
Visualization of Flow Pattern
Visualization of perfusate in the glass aortic models was performed by a particle image velocimetry (PIV) system (TSI Inc, Shoreview, MN). The PIV measurement system is composed of a double-pulsed neodymium-doped yttrium aluminium garnet (Nd:YAG) laser, a high-resolution charge-coupled device (CCD) camera, a synchronizer (Model 610034, TSI, Inc, Shoreview, MN), and a personal computer (Fig 2). The double-pulsed Nd:YAG laser (Solo II-15, New Wave Research, Inc., Fremont, CA) has a maximum output of 30 mJ/pulse, a repetition rate of 15 Hz, and a wavelength of 532 nm. The beam was expanded into a thin light sheet of approximately 1 mm through a cylindrical lens.
For vector image capturing, an interline CCD camera (MEGAPLUS model ES1.0, REDLAKE MASD Inc, Tucson, AZ) was used. It has a resolution of 1008 x 1018 pixels, a maximum frame rate of 30 frames/s, and a dynamic range of 8 bit. The double-pulsed interval, which was adjusted to give reasonable vector images, was 0.5 msec.
The seeding particles, as the tracer for PIV system, were composed of a high-porous polymer (MCI GEL CHP20P, Mitsubishi Chemical Co, Ltd, Tokyo, Japan) with a diameter of 75 to 150 µm.
The Stokes number (S, S = 
r2U/µD) was 0.048, and was calculated using the particle density ( = 1016.98 kg/m3), dynamic viscosity of water at 25°C (µ = 0.890 x 10–3Pas), mean radius of the particle (r = 50 µm), reference velocity around the aortic arch (U = 0.5 m/s), and lumen diameter of the aortic arch model (D = 30 mm). Under this condition, the Stokes number was small enough for tracers to follow the flow.
The calibration procedure and the PIV cross-correlation analysis were done using INSIGHT 3.34 software (TSI Inc, Shoreview, MN) with an interrogation window size of 64 x 64 pixels, with 50% overlapping. The spatial resolution of the velocity field was 2.2 to 2.8 mm/pixel. After the vector images were captured, spurious vectors were removed by applying filtering procedures installed in the INSIGHT 3.34 software. The statistical analysis was performed by the same software using 40 image pairs for each experimental case.
The aortic arch model was immersed in the cubic water tank to prevent refraction at the glass wall and mounted on a precision mechanical stage (Fig 2). The tip of the cannula was defined as point zero. The PIV measurements were done in a plane that included the center axes of the cannula and the aortic arch (longitudinal plane), and in a plane perpendicular to the center axis of the aortic arch (cross-sectional plane). The PIV measurements in the cross-sectional plane were conducted from 20 to 70 mm upstream and 20 to 70 mm downstream from the tip of the cannula at 5-mm intervals. Those images were captured from the upstream side to downstream side.
The magnitude of the 2-dimensional velocity, U, and streamlines in the longitudinal plane were obtained using Tecplot 9.0 software (Amtec Engineering Inc, Bellevue, WA). A FORTRAN computer program was written for calculating the magnitude of strain rate tensor |D| which expresses the shear stress. It is defined as follows:
| (1) |
where the velocity vectors (u, v) at the coordinates (x, y) were measured by PIV. The root mean square value of velocity magnitude fluctuation, Urms, in the longitudinal plane was calculated using a velocimetry data set.
Quantitative Analysis of Flow Vectors
The velocity magnitude of the flow vectors was measured at 75 points in each cross-sectional plane of the healthy aortic arch model. Absolute values of average, maximum, and minimum flow velocity vectors in each cross-sectional plane were depicted according to distance from cannula tip. This analysis was not performed in an aneurysmal model because of inequality of cross-sectional area in aneurysmal model.
|
Results
|
|---|
Flow Analysis in the Healthy Aortic Model
When the cannula tip was positioned in the standard direction in the model of the healthy aorta, vortical regurgitant flow was observed on the greater curvature side of the cannula near the ostium of the BCA, and an additional vortex appeared in the distal aortic arch. Because the flow ejecting from the dispersion cannula spreads out in a fan-like shape on the plane perpendicular to the 2-dimensional projection, streamtraces ejecting away from the cannula tip were not clearly depicted in the streamline map in the longitudinal plane. The magnitude of flow velocity in the aortic arch was greater in the standard direction than that in root direction (Fig 3).
View larger version (67K):
[in this window]
[in a new window]
|
Fig 3. Streamline and flow velocity analysis in the model of a healthy aorta is shown in the (A) standard direction and (B) root direction. Flow vector map in the cross-sectional plane is shown in the surrounding schema. (C) Enlarged view of streamtrace and flow velocity distribution in the aortic root is shown in the root direction. Flow velocity is depicted with color scale grading. Yellow arrow indicates cannulation site and direction of cannula tip.
|
|
When the cannula tip was directed toward the aortic root, streamtraces formed complicated vortices, indicating that 2 directional flow streams ejected from the cannula tip and reflected off the aortic root (Fig 3C). The velocity distribution revealed fast flow toward the aortic root but very slow flow in the aortic arch. Points of high shear stress appeared along the greater curvature and in the proximal portion of the lesser curvature in the standard direction (Fig 4A).
View larger version (31K):
[in this window]
[in a new window]
|
Fig 4. Distribution of shear stress, |D|, in the ascending and transverse arch is shown in the (A) standard direction and (B) root direction in the model of a healthy aorta. The black arrow indicates cannulation site and direction of cannula tip.
|
|
Although |D| was high in the ascending aorta in the root direction, it was low in the transverse aorta (Fig 4B). In the standard direction, distribution of Urms was relatively homogenous, whereas the distribution of Urms in the root direction was high in the ascending aorta due to the diverse direction of flow vectors caused by the reflected off the aortic root. However, distribution of Urms in the transverse aorta was uniformly low in root direction (data not shown).
Flow Analysis in the Aortic Arch Aneurysm Model
In the aneurysmal aortic model, streamline analysis in the standard direction demonstrated nonlinear flow in the transverse aorta. A rapid flow velocity area and a regurgitant vortex were observed around the aortic arch vessels in the longitudinal plane. Regurgitant streamtraces grazing the lesser curvature eventually flowed back into the arch vessels. In the cross-sectional plane, 2 reversely directed vortices flowing from the lesser curvature were observed at the aortic arch vessels, approximately 40 mm from the cannula tip (Fig 5A). The magnitude of velocity vectors in this area was large. On the other hand, flow velocity in the aneurysmal region was slow. In the root direction, rapid flow was observed in the aortic root region. However, a flow velocity map demonstrated generally slow flow distribution around the aortic arch vessels and in the aneurysmal region (Fig 5B). Two reversely directed vortexes in the arch vessel area were observed in cross-sectional plane at 40 mm, but velocity vectors were smaller than that in the standard direction.
View larger version (42K):
[in this window]
[in a new window]
|
Fig 5. Streamline and flow velocity analysis in the model of the aneurysmal aortic arch is shown in the (A) standard direction and (B) root direction. The flow vector map in cross-sectional plane is shown in the surrounding schema. The magnitude of the velocity vector component is expressed in color and length of vector arrow. Yellow arrow indicates cannulation site and direction of cannula tip.
|
|
In the shear stress analysis, a high |D| value area was observed in the cannula tip, the ostium of the brachiocephalic artery, and the roof of the aneurysm in the standard direction (Fig 6A). In the root direction model, a high |D| value area was observed in the aortic root (Fig 6B).
View larger version (32K):
[in this window]
[in a new window]
|
Fig 6. Distribution of shear stress, |D|, in the ascending and transverse arch in the model of the aneurysmal aorta is shown in the (A) standard direction and (B) root direction. The black arrow indicates cannulation site and direction of cannula tip.
|
|
Quantitative Analysis of Flow Vectors in the Healthy Aortic Model
The absolute value of the flow vectors in the cross-sectional plane was compared in standard direction and aortic root direction in the model of a healthy aorta (Fig 7). Although the range of absolute values of the flow vectors was high in the ascending aorta in the root direction, absolute values of the flow vectors in the transverse aorta was low compared with the standard direction model.
View larger version (8K):
[in this window]
[in a new window]
|
Fig 7. Flow velocity vector distribution in each cross-sectional area is shown in the (A) standard direction and (B) aortic root direction. Point 0 mm indicates the cannulation site. Bars indicate range of vector value. Solid squares indicate average value.
|
|
|
Comment
|
|---|
Recently, more aged patients are undergoing cardiac operations using CPB for ischemic heart disease, thoracic aortic aneurysm, and degenerative valvular disease. As patients advance in age, the prevalence of atherosclerosis of the ascending aorta and aortic arch increases. Incidence of stroke after cardiac procedures increases proportionally to the patient's age [2, 3]. Sometimes cardiac surgeons encounter paste-like material filling in the aortic wall or frail atheroma in the ascending aorta and the aortic arch in operations for the ascending aorta and transverse aortic arch [1, 10]. Katz and colleagues [11] reported protruding atheroma of the aortic arch was a significant risk factor for perioperative stroke after cardiac operations. They documented a 25% incidence of stroke in patients who had mobile atheroma in the aortic arch detected by transesophageal echocardiography. Djaiani and colleagues [12] demonstrated a correlation between the embolic signal count in the area of the middle cerebral artery detected by transcranial Doppler imaging during CPB and postoperative embolic stroke detected by diffusion-weighted magnetic resonance imaging. These reports correspond to our previous clinical observation using preoperative and post-CPB epiaortic ultrasound imaging [6]. Ura and colleagues [4] detected newly generated frail atheroma in the aorta remote from the cannulation site using intraoperative epiaortic echogram in patients who had a stroke after operation.
Weinstein [13] investigated the laterality of stroke after conventional coronary artery bypass grafting and reported a 75% incidence of stroke in the left hemisphere compared with 25% in the right hemisphere. Weinstein speculated that high velocity jet flow from the end-hole cannula might be important in the formation of atheroembolism. These clinical observations suggest that decreasing the sandblast effect of jet flow from the cannula tip is important in preventing postoperative embolic stroke in cardiac operations.
We previously investigated aortic flow in four types of arterial perfusion cannulas and elucidated reduction of flow velocity magnitude and shear force in dispersive type cannulas. In the classic curved end-hole cannula, jet flow from the cannula tip hit the posterior wall of the greater curvature directly, with great shear stress on the aortic wall. It may induce a sandblast effect, causing detachment of atheromatous debris from the diseased aortic arch wall.
The Dispersion cannula has a unique shape, with a flat and obtusely angled flow reflector. It generates a fan-shaped flow without destruction of cellular components of the blood. Flow velocity distribution in the transverse aorta in this experiment was almost the same as an average flow velocity distribution in physiologic conditions [14]. Owing to reduction of flow velocity and shear stress, directing the Dispersion cannula toward the aortic arch is beneficial when flail atheromatous plaque exists in the distal aortic arch. However, when thick soft plague is present on the lesser curvature of the aortic arch, tangential shear force may displace and detach the protruding atheroma in the lesser curvature [6].
Streamline analysis in this study demonstrated reversely directed streamtraces from the aortic arch went into arch vessels in the models of healthy and aneurysmal aortas when the cannula tip was directed toward the arch direction. Considering shear force distribution, detached debris may flow into the brain circulation following these streamtraces.
Grooters and colleagues [8] proposed using a Dispersion cannula directing flow toward the aortic root. They showed clinical efficacy and safety of this method in coronary artery bypass grafting. Our experimental study demonstrated that flow in the aortic arch in both healthy and aneurysmal models was slow and less turbulent when the cannula tip was directed toward the aortic root. Although shear stress was high in the ascending aorta, it was very low in the transverse aortic arch in this method. On the other hand, a higher degree of turbulence was observed in the ascending aorta when the cannula was directed toward the aortic root.
Axillary artery perfusion is one of the alternatives for arterial access in aortic arch interventions. Our previous study using aortic glass model showed there was still a risk of atheroembolism in axillary perfusion when flail plaque was present in the lesser curvature and brachiocephalic ostium [15].
Despite flow velocity reduction in the model of the aneurysmal aorta, flow velocity and shear force was high in the proximal aortic arch where the cranial vessels arose. Reverse flow within the arch may induce atheroembolic stroke because severe intimal change frequently exists in this area. Therefore, aortic cannulation toward the root direction is one of the options for operations in patients who have severe atherosclerosis of the transverse aortic arch. Because shear stress and turbulence are high in the ascending aorta, aortic root direction should not be used if there is calcified vegetation around the aortic root in aged patients with degenerative aortic stenosis. Root direction is also contraindicated in aortic regurgitation due to high velocity magnitude distribution in the aortic root.
This study has several limitations. The arterial perfusate was water in which viscosity was low. In extracorporeal circulation, blood viscosity is decreased as a result of blood dilution. To compare experimental results using various cannulas and various shapes of the aorta, we used water as the perfusate to keep the experimental condition the same. The shape of the ascending aorta and the transverse aorta is diverse. Selection of aortic cannulation sites in clinical practice is limited by aortic morphology, surgical approach, and operative field.
In conclusion, directing the Dispersion cannula tip toward the aortic root generated slower and less turbulent flow in the transverse arch of a transparent aortic model compared with flow with cannulation in a standard direction. The sandblast effect caused by high velocity and shear stress from cannula tip may be attenuated in this method.
|
Acknowledgments
|
|---|
This study was supported by Grant 16390388 from Ministry of Education and Science of Japan.
|
References
|
|---|
- Wareing TH, Davila-Roman VG, Barzilai B, Murphy SF, Kouchoukos NT. Management of the severely atherosclerotic ascending aorta during cardiac operations J Thorac Cardiovasc Surg 1992;103:453-462.[Abstract]
- Rao V, Christakis GT, Weisel RD, et al. Risk factors for stroke following coronary bypass surgery J Card Surg 1995;10:468-474.[Medline]
- Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery) Circulation 2004;110:e340-e437.[Free Full Text]
- Ura M, Sakata R, Nakayama Y, Goto T. Ultrasonographic demonstration of manipulation-related aortic injuries after cardiac surgery J Am Coll Cardiol 2000;35:1303-1310.[Abstract/Free Full Text]
- Calafiore AM, Di Mauro M, Teodori G, et al. Impact of aortic manipulation on incidence of cerebrovascular accidents after surgical myocardial revascularization Ann Thorac Surg 2002;73:1387-1393.[Abstract/Free Full Text]
- Fukuda I, Minakawa M, Fukui K, et al. Breakdown of atheromatous plaque due to shear force from arterial perfusion cannula Ann Thorac Surg 2007;84:e17-e18.[Abstract/Free Full Text]
- Grooters RK, Ver Steeg DA, Stewart MJ, Thieman KC, Schneider RF. Echocardiographic comparison of the standard end-hole cannula, the soft-flow cannula, and the dispersion cannula during perfusion into the aortic arch Ann Thorac Surg 2003;75:1919-1923.[Abstract/Free Full Text]
- Grooters RK, Thieman KC, Schneider RF, Nelson MG. Assessment of perfusion toward the aortic valve using the new dispersion aortic cannula during coronary artery bypass surgery Tex Heart Inst J 2000;27:361-365.[Medline]
- Minakawa M, Fukuda I, Yamazaki J, Fukui K, Yanaoka H, Inamura T. Effect of cannula shape on aortic wall and flow turbulence: hydrodynamic study during extracorporeal circulation in mock thoracic aorta Artif Organs 2007;31:880-887.[Medline]
- Ribakove GH, Katz ES, Galloway AC, et al. Surgical implications of transesophageal echocardiography to grade the atheromatous aortic arch Ann Thorac Surg 1992;53:758-761.[Abstract/Free Full Text]
- Katz ES, Tunick PA, Rusinek H, Ribakove G, Spencer FC, Kronzon I. Protruding aortic atheromas predict stroke in elderly patients undergoing cardiopulmonary bypass: experience with intraoperative transesophageal echocardiography J Am Coll Cardiol 1992;20:70-77.[Abstract]
- Djaiani G, Fedorko L, Borger M, et al. Mild to moderate atheromatous disease of the thoracic aorta and new ischemic brain lesions after conventional coronary artery bypass graft surgery Stroke 2004;35:e356-e358.[Abstract/Free Full Text]
- Weinstein GS. Left hemispheric strokes in coronary surgery: implications for end-hole aortic cannulas Ann Thorac Sugr 2001;71:128-132.
- Segadal L, Matre K. Blood velocity distribution in the human ascending aorta Circulation 1987;76:90-100.[Abstract/Free Full Text]
- Minakawa M, Fukuda I, Inamura T, et al. Hydrodynamic evaluation of axillary artery perfusion for normal and diseased aorta Gen Thorac Cardiovasc Surg 2008;56:215-221.[Medline]
This article has been cited by other articles:
|
|
|
|
 
K. Yamana, T. Ito, A. Maekawa, T. Yoshizumi, M. Sunada, and S. Hoshino
Atherosclerotic Arch Aneurysm Operations With Perfusion Toward the Aortic Valve
Ann. Thorac. Surg.,
February 1, 2010;
89(2):
435 - 439.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
