|
|
||||||||
Eur J Cardiothorac Surg 2004;25:578-584
© 2004 Elsevier Science NL
a Department of Surgery II, Tokyo Medical University, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan
b Department of Internal Medicine II, Tokyo Medical University, 6-7-1 Nishishinjuku, Shinjuku-ku, Tokyo 160-0023, Japan
Received 9 October 2003; received in revised form 7 December 2003; accepted 15 December 2003.
* Corresponding author. Tel.: +81-3-3342-6111x5077; fax: +81-3-3342-6193
e-mail: tshimizu-cvs{at}umin.ac.jp
| Abstract |
|---|
|
|
|---|
Key Words: Coronary artery bypass Coronary artery disease Internal thoracic artery Gastroepiploic artery Saphenous vein
| 1. Introduction |
|---|
|
|
|---|
Wall shear stress has been thought to play an important role in the development of atherosclerosis. Shear stress increases the release of nitric oxide [3] and the production of prostacylin [4] by endothelial cells and inhibits smooth muscle cell proliferation [5]. In other words, low shear stress is associated with the development of atherosclerosis. Bach et al. [6] found that shear rates of the ITA were higher compared with the SVG, and they suggested these differences might have implications regarding the development of degenerative graft disease and long-term conduit patency. Indeed, endothelial cells in the ITA release more prostacylin [7] and nitric oxide [8] than those in the SVG.
Flow competition is a significant factor affecting arterial conduit morphology and patency. The string sign is an atrophic change in the arterial conduit and this phenomenon is due to competitive flow in grafts connected to only mildly stenosed coronary arteries. The fate of the string sign of the ITA is still controversial. In the GEA, patency rates seemed to be reduced by competitive flow. Suma et al. [9] reported that the 10 year patency rate of the GEA was 62.5%, and that anastomosis to a less critically stenosed coronary artery was one of the major causes of late graft occlusion. The flow characteristics of the GEA are yet to be fully clarified. The relationship between the SVG patency and native coronary artery stenosis has also been controversial, however, competitive flow may be a negligible factor in SVG graft patency.
The effect of flow competition on coronary artery bypass conduit shear stress is unknown. The ITA graft diameter has been shown to decrease when the native coronary artery stenosis is less [10]. Greater flow velocity and smaller vessel diameter increase shear stress. It may be beneficial for a bypass conduit to maintain conduit shear stress even if the conduit diameter is reduced by competitive flow.
An intravascular Doppler-tipped guide wire (DGW), developed as a coronary angioplasty guide wire, has been used for analysis of phasic flow velocity of ITA grafts and SVGs during postoperative angiography [6]. In the present study, flow velocity was detected in the ITA, GEA and SVG using a DGW. Shear stress was calculated from the graft flow velocity and diameter. Percent diameter stenosis was taken as an indicator showing native coronary flow. Intermediate coronary stenosis for coronary interventions was estimated between 50 and 75% [11]. The effect of native flow on conduit shear stress was investigated among the ITA, GEA and SVG.
| 2. Methods |
|---|
|
|
|---|
Study grafts consisted of 40 ITA grafts (39 left and 1 right ITA), 27 GEA grafts and 34 SVGs. All ITA grafts were anastomosed to the left anterior descending artery. Of the 27 GEA grafts, 2 were placed to the left anterior descending artery, 15 were grafted to the right coronary system (14 to the posterior descending branch, 1 to the postero-lateral branch) and 10 were grafted to the left circumflex coronary artery. Of the 34 SVGs, 3 were grafted to the left anterior descending artery, 16 were grafted to the right coronary system (3 were grafted to the distal right coronary artery, 12 to the posterior descending branch and 1 to the postero-lateral branch), and 15 were grafted to the left circumflex coronary artery.
2.2. Coronary angiography and flow velocity measurement
Coronary angiography was performed by the standard femoral approach. After ITA, GEA or SVG angiography, a 5 or 6 F catheter was positioned in the origin of the grafts. A 0.015 or 0.018-in., 12-MHz DGW (Jomed Inc., Flowire) was connected to a velocitymeter (Jomed Inc., FloMap), and advanced through the catheter into the graft and introduced to the distal portion of the graft.
2.3. Graft flow volume and shear stress calculation
The graft diameter at the points of flow velocity measurement was determined by angiography using an automated edge-contour detection system (Cardio 500; Kontron Electronic AG, Eching, Germany).
Flow volume (Q ml/min) was calculated using the following equation as previously reported [12]:
![]() |
The SR (s-1) was calculated as:
![]() |
The shear stress (SS dyn/cm2) was calculated using a modified HagenPaiseuille equation [13]:
![]() |
![]() |
2.4. Patient classification
The study patients/grafts were classified according to the grade of native coronary artery percent diameter stenosis at the time of postoperative angiography into group H, which had more than 75% stenosis and group L, which had more than 50% stenosis up to 75% stenosis.
2.5. Statistical analysis
One-way analysis of variance (ANOVA) was used to compare three or six groups for continuous data. A Scheffe test was performed when the ANOVA showed significant differences. An unpaired t-test was used to compare two groups for continuous data. The
2-test was used for nominal data.
| 3. Results |
|---|
|
|
|---|
|
|
|
3.1.3. Velocimetric and calculated data
Average peak velocity in group H significantly differed among the three conduits: ITA 26±7 cm/s; GEA 18±5 cm/s; SVG 12±4 cm/s (ITA vs. GEA P<0.0001; ITA vs. SVG P<0.0001; GEA vs. SVG P=0.013). In group L, average peak velocity was greater (ITA vs. GEA P<0.0001; ITA vs. SVG P=0.0015) in the ITA (19±6 cm/s) than the GEA (9±3 cm/s) or SVG (11±5 cm/s), but did not differ between the GEA and SVG (P=0.495). In the arterial conduits, average peak velocity significantly differed between groups H and L (ITA P=0.0022; GEA P<0.0001), but did not differ in the SVG between groups H and L (P=0.6141).
Flow volume in group H did not significantly differ among the three conduits: ITA 34±11 ml/min; GEA 35±16 ml/min; SVG 41±15 ml/min (ITA vs. GEA P=0.9786; ITA vs. SVG P<0.3370; GEA vs. SVG P<0.5179). In group L, flow volume was greater (P<0.0001) in the SVG (35±16 ml/min) than the ITA (18±6 ml/min) or GEA (13±8 ml/min), but did not significantly differ between the ITA and GEA (P=0.5776). In the arterial conduits, flow volume significantly differed between groups H and L (ITA P<0.0001; GEA P<0.0005), but did not differ in the SVG between groups H and L (P=0.4296).
Shear stress in group H significantly differed among the three conduits: ITA 16.0±4.8 dyn/cm2; GEA 9.1±3.2 dyn/cm2; SVG 4.8±1.6 dyn/cm2 (ITA vs. GEA P<0.0001; ITA vs. SVG P<0.0001; GEA vs. SVG P=0.002). In group L, shear stress was greater (P<0.0001) in the ITA (13.7±4.9 dyn/cm2) than the GEA (5.6±2.0 dyn/cm2) or SVG (4.6±2.0 dyn/cm2), but did not differ between the GEA and SVG (P=0.5832). Between groups H and L, shear stress significantly differed in the GEA (P=0.0041), but did not differ in the ITA (P=0.1354) or SVG (P=0.8021).
The shear rate in group H was 458±136 s-1 in the ITA, 261±93 s-1 in the GEA, and 137±46 s-1 in the SVG. In group L, the shear rate was 391±141 s-1 in the ITA, 159±56 s-1 in the GEA, and 130±57 s-1 in the SVG. Statistical differences in the shear stress among each group were the same as those for shear stress.
| 4. Discussion |
|---|
|
|
|---|
On the other hand, in the group with lower grade coronary stenosis, flow volume was lower in the arterial conduits than in the SVG. In other words, flow volume of the arterial conduits was reduced by native flow competition, however, the effect of competitive flow was not significant in the SVG. The patency of the SVG might be poorly associated with the native coronary stenosis. In contrast, the calculated shear stress of the ITA was also higher than that of the SVG, even when the grafts were anastomosed to the lower grade coronary stenosis, whereas shear stress of the GEA was as low as that of the SVG under flow competitive conditions. These results suggest that the ITA might have more flow adaptability to modulate arterial diameter to maintain flow velocity and shear stress, compared with the GEA. Graft narrowing of the ITA because of flow competition is generally reversible due to progression of native coronary artery stenosis. The string sign is considered to be an atrophic change in the ITA graft occurring with flow competition. It is unknown how long the non-functioning ITA graft maintains anatomical patency but we presume that the presence of the string sign in the GEA would not be prolonged, as this sign is relatively rare in a GEA graft with flow competition. This may be due to morphologic and anatomical differences between these grafts; the GEA is a muscular artery whereas the ITA is an elastic artery [17].
Ofili et al. [18] reported that the average peak velocity measured using a DGW is 23±11 cm/s in the distal LAD, 21±6 in the distal CX and 21±9 in the distal right coronary artery. The shear stress of the distal coronary artery near the anastomosis can be estimated using the formula used in the present study, producing a shear stress of approximately 15 dyn/cm2 (shear rate 420). This value is similar to the shear stress obtained in the distal portion of the ITA grafts in both groups H and L. On the other hand, the mean shear stress of the GEA near the anastomosis in group L was 5 dyn/cm2. This value seems to be too low for an arterial conduit grafted to the coronary artery.
Flow reversal in early systole was observed in the arterial conduits to the coronary artery with lower grade stenosis. The distance from the aortic root to the graft anastomosis causes a delay in the pressure wave of the graft, with the wave reaching the ITA or GEA after the coronary artery, which is much closer to the aorta. This is one of the reasons why the ITA graft shows the so called to and fro pattern [19] or swinging flow pattern [10] under high-grade competitive flow conditions. This delay might affect graft patency, and it is unlikely that such an effect would be beneficial under competitive flow conditions. Compared with the ITA, flow reversal seemed to be more apparent in the GEA, which is farther from the aorta. However, the long-term patency rate for coronary arteries with intermediate stenosis is likely to be better in ITA grafts than GEA grafts.
It has been speculated that increased shear stress might be one of the major factors affecting the patency of the saphenous vein after implantation into arterial circulation. Porter et al. [20] found that the arterial shear stress (9 dyn/cm2) inhibited the development of saphenous vein intimal hyperplasia more than venous shear stress (1 dyn/cm2). Zhu et al. [21] compared the expression levels of endothelin-1 and constitutional nitric oxide synthesis by the saphenous vein endothelial cells exposed to various shear stress (30, 6, 2, and 0 dyn/cm2) and found that the saphenous vein under 6 dyn/cm2 had the best endothelial function. Whether the calculated value (5 dyn/cm2) of SVG shear stress obtained in the present study can be estimated to be increased or decreased as a venous conduit grafted into coronary circulation remains unclear.
The present study has several limitations. First of all, shear stress was estimated from the graft diameter and velocity on the hypothesis that blood viscosity was constant. Although blood is a non-Newtonian fluid, blood viscosity has a relatively constant value for share rates above 100 s-1 [22]. Therefore, many investigators have estimated shear stress on this hypothesis in previous studies [23]. Secondly, percent diameter stenosis is not always the best predictor for native coronary flow. Therefore, it is difficult to define the degree of coronary stenosis required to keep the GEA graft patent.
Thirdly, the radial artery is a common arterial conduit for coronary revascularization, however, it had not been used in our institutes yet during the study period. Moreover, the radial artery, which might also be sensitive to competitive flow [24], seems to be unsuitable for this study because it is at high risk of spasm induced by a DGW. Fourthly, because the present study only determined the hemodynamic differences among the three kinds of bypass conduits, long-tem graft patency rates for lower grade stenosis should be considered to decide which conduit should be used to each target coronary artery. Our data suggest that the GEA is not necessarily feasible for lower grade coronary stenosis. When the right coronary artery system is a target vessel with lower grade coronary stenosis, the radial artery is also at high risk of failure [24]. Results of the right ITA to the posterior descending artery are not always acceptable [25]. The SVG to the right coronary artery system may be the best choice for coronary revascularization with native flow competition. When the circumflex branch is a target, the right ITA is often available via the transverse sinus while the left ITA to the circumflex and the right ITA to the left anterior descending artery are other options. In addition, prophylactic ITA anastomosis to the LAD without stenosis is still not recommended, although the present study showed superior shear stress of the ITA.
In conclusion, shear stress of the ITA was superior and maintained despite the flow volume being reduced by flow competition. These results suggest that the ITA has flow adaptability to modulate arterial diameter so as to maintain flow velocity, shear stress and graft patency. Lower shear stress of the GEA for less severe coronary stenosis may be associated with the development of conduit failure. Therefore, GEA grafting for intermediate stenosis should be avoided when the other conduits are appropriate.
| Footnotes |
|---|
| Appendix A. Conference discussion |
|---|
|
|
|---|
Dr Shimizu: Yes, it is occasionally correct. The diameter is very important to assess the shear stress. If the bypass conduits can provide the same flow volume, the flow velocity is lower when the diameter is greater. This is the reason why the gastroepiploic artery has low shear stress.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
N. D. Desai, C. D. Naylor, A. Kiss, E. A. Cohen, R. Feder-Elituv, S. Miwa, S. Radhakrishnan, J. Dubbin, L. Schwartz, S. E. Fremes, et al. Impact of Patient and Target-Vessel Characteristics on Arterial and Venous Bypass Graft Patency: Insight From a Randomized Trial Circulation, February 13, 2007; 115(6): 684 - 691. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Miwa, N. Desai, T. Koyama, E. Chan, E. A. Cohen, S. E. Fremes, and Radial Artery Patency Study Investigators Radial Artery Angiographic String Sign: Clinical Consequences and the Role of Pharmacologic Therapy Ann. Thorac. Surg., January 1, 2006; 81(1): 112 - 119. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Deja, K. S. Golba, M. Malinowski, S. Wos, M. Kolowca, J. Biernat, M. Kajor, and T. J. Spyt Skeletonization of internal thoracic artery affects its innervation and reactivity Eur. J. Cardiothorac. Surg., October 1, 2005; 28(4): 551 - 557. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Shimizu, H. Suesada, M. Cho, S. Ito, K. Ikeda, and S. Ishimaru Flow Capacity of Gastroepiploic Artery Versus Vein Grafts for Intermediate Coronary Artery Stenosis Ann. Thorac. Surg., July 1, 2005; 80(1): 124 - 130. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |