|
|
||||||||
Ann Thorac Surg 1998;66:1282-1287
© 1998 The Society of Thoracic Surgeons
a Cardiovascular and Thoracic Surgery,University Hospital of Mont-Godinne, Catholic University of Louvain, Yvoir, Belgium
b Cardiology,University Hospital of Mont-Godinne, Catholic University of Louvain, Yvoir, Belgium
c Biostatistics, University Hospital of Mont-Godinne, Catholic University of Louvain, Yvoir, Belgium
Accepted for publication May 4, 1998.
Address reprint requests to Dr Louagie, Cardiovascular and Thoracic Surgery, University Hospital of Mont-Godinne, 1 av Therasse, B5530 Yvoir, Belgium
| Abstract |
|---|
|
|
|---|
Methods. A pulsed Doppler flowmeter was routinely used during operation to determine the hemodynamic parameters of coronary bypass grafts. During a 7-year period, 85 patients underwent angiographic evaluation. As a result, a thorough hemodynamic assessment of 214 grafts (89 arterial and 125 venous) at initial operation was available for analysis.
Results. The overall patency rate was 88.3%. The mean flow measured intraoperatively in 168 intact grafts was 60 ± 3 mL/min (range, 9 to 230 mL/min), and the resistance was 1.8 ± 0.1 peripheral resistance units (range, 0.3 to 9.0 peripheral resistance units). The mean flow was 36 ± 5 mL/min (range, 2 to 107 mL/min), and the resistance was 5.9 ± 2.0 peripheral resistance units (range, 0.6 to 46.0 peripheral resistance units) in 25 grafts found occluded at angiographic evaluation. Multivariate analysis identified three independent variables associated with a reduced patency rate: increased resistance as measured in the graft (p = 0.012), increasing interval of control angiography (p = 0.006), and preoperative cardiogenic shock (p = 0.040).
Conclusions. The prognosis for midterm patency of aortocoronary bypass grafts depends on the intraoperative hemodynamic status.
| Introduction |
|---|
|
|
|---|
The purpose of our study was to focus on the characteristics of intraoperative graft hemodynamic assessment, the appearance (angiographic and macroscopic) of the host coronary vessels, and the relationship to graft patency. Data were obtained by using a pulsed Doppler flowmeter that has been adapted to the assessment of both arterial and venous coronary bypass grafts [68].
| Material and methods |
|---|
|
|
|---|
The mean age at operation was 61.2 ± 1.0 years. There were 71 men and 14 women. The indications for bypass grafting included unstable angina in 30 patients (35.3%) and angina after recent (<6 weeks) myocardial infarction in 11 patients (12.9%). The extent of coronary artery disease at angiography was as follows: left main disease (isolated or associated), 10 patients; one-vessel disease, 2 patients; two-vessel disease, 30 patients; and three-vessel disease, 50 patients. Left ventricular ejection fraction was obtained angiographically in 45 patients and averaged 0.62 ± 0.02. It was measured by radioisotope in less favorable cases in 14 patients and averaged 0.45± 0.03.
Repeat catheterization was performed within 1 month of operation in 16 patients and later (median, 17 months; range, 1 to 83 months) in 69 patients. Grafts were classified angiographically as intact (meaning widely patent with no perceptible narrowing), stenotic, or occluded. Nonoccluded grafts, including stenotic grafts, were classified as patent. The stenotic category included grafts with subcritical stenoses (including new angiographic irregularities) as well as grafts severely stenotic and hemodynamically compromised. The entire length of the grafts and the anastomoses were considered. If multiple stenoses were present, the graft was classified according to the most severe narrowing.
Surgical techniques
All distal anastomoses were done during a single interval of aortic cross-clamping, and the aortic anastomoses were done with tangential aortic cross-clamping while the heart was kept in the beating empty state. The luminal diameter of the recipient coronary arteries was measured with malleable probes in 0.5-mm gradations. A total of 245 distal anastomoses were constructed (2.9 per patient; range, 1 to 4). One hundred five anastomoses (43%) were completed with arterial grafts, whereas 140 anastomoses (57%) were done with venous grafts. Sixty-nine patients (81%) underwent the implantation of a single internal thoracic artery and 12 patients (14%) had bilateral internal thoracic artery grafting. Sequential bypass grafting was performed in 7 patients. An epigastric artery was implanted in 5 patients and a gastroepiploic artery was grafted to the posterior descending artery in 6 patients. Redo procedures were performed in 6 patients.
Hemodynamic measurements and velocity patterns
The flow in the grafts was measured on bypass immediately after completion of the proximal anastomoses, while the patient was rewarmed to 36°C. An 8-MHz pulsed-wave Doppler ultrasound flowmeter (OPDOP 130; Scimed, Bristol, UK) was used. Recording sites were in the proximal vein graft segment, the midportion of the internal thoracic artery pedicle, or the internal epigastric artery. In the case of sequential bypass grafts, blood flow velocities were measured in the region proximal to the side-to-side anastomosis. Flow measurements were obtained by constraining the vessel in an acrylic cuff, the two halves of which are clipped around the vessel. The ultrasound pencil probe is slotted into the cuff and acoustically coupled to the vessel by a small amount of sterile gel. A detailed description of the method used and results of a validation study were published previously [8]. Sharp dissection was not necessary to measure flow in the internal thoracic artery. Regarding the gastroepiploic artery, the surrounding fat was incised at a level 2 to 3 cm distal from the origin of the gastroduodenal artery.
Intraoperative Doppler measurements and selective graft injection at control angiography were obtained for 214 bypass grafts over a total of 238 implanted grafts (90%).
The parameters determined by the pulsed Doppler were flow (mL/min), velocity (cm/s), and internal diameter (mm) of the vessel. Furthermore, total resistance of the graft and coronary bed was calculated by the OPDOP 130 software from mean radial artery pressure divided by mean flow measured after completion of the coronary anastomoses and is expressed in peripheral resistance units (PRU). A pulsatility index (PI) was calculated to describe the shape of the curves. It is a dimensionless variable, independent of probe to vessel angle. This variable was defined as follows:
.
Data analysis
Perioperative data, including hemodynamic graft measurements, were collected and entered prospectively into a clinical research data base. Values are presented as mean ± standard error of the mean. Statistical analysis was performed with the SPSS (SPSS Inc, Chicago, IL) software package, unless otherwise specified. Clinical categoric data were compared between two populations by
2 test. Intergroup comparisons (intact graft, stenosis, occlusion) were done by Kruskal-Wallis exact (or Monte Carlo) permutation test for categoric variables and Jonckheere-Tepstra exact permutation test for ordered variables, both performed with StatXact (Cytel Software Corp, Cambridge, MA). Graft hemodynamic parameters expressed as continuous variables were compared between the aforementioned three groups by one-way analysis of variance followed by Scheffé tests for 2 by 2 comparisons.
Graft patency curves were constructed by the Kaplan-Meier method and compared by log rank test, assuming independence of the grafts in the same patient.
Graft patency rates were compared by regression analysis of repeated measures using generalized estimating equations as described by Liang and Zeger [9], the occlusion of the graft being considered as the dependent variable. The latter analysis allows to take into account simultaneously patient-related and graft-related variables, and was performed by RMGEE program [10]. Backward selection with a p value less than 0.05 limit was used. In the first analysis, only graft hemodynamic data were considered. These variables included flow, velocity, resistance, pulsatility index, and internal diameter. Both graft- and patient-specific characteristics were included in the second analysis (Appendix 1).
| Results |
|---|
|
|
|---|
Nature of the bypass conduit
The nature of the graft had a marked influence on patency because the overall difference between grafts was significant (p = 0.008) (Table 1 ). The arterial grafts had a global patency rate of 94% (84 of 89 patients) contrasting with a patency rate of 84% (105 of 125 patients) for saphenous vein bypass grafts (p = 0.020). In addition, the left internal thoracic artery graft had a patency rate superior to the right internal thoracic artery, the internal epigastric artery, and the saphenous vein graft. Nevertheless, these data need to be interpreted cautiously as a given bypass graft is often associated with a specific runoff bed. For example, in the majority of patients the left internal thoracic artery was grafted to the widest runoff bed, the left anterior descending artery.
|
|
|
|
3 PRU (Fig 2 ).
|
Multivariate analysis of factors likely to influence patency
Patient- and graft-specific variables were simultaneously assessed by generalized estimating equations. Three independent variables were associated with a reduced patency rate: increased resistance as measured in the graft (p = 0.012), increasing interval of control angiography (p = 0.006), and preoperative cardiogenic shock (p = 0.040). The nature of the bypass conduit was not selected by the multivariate analysis as a variable independently influencing graft outcome.
| Comment |
|---|
|
|
|---|
The present study differs from those studies by the use of a pulsed Doppler flowmeter and the assessment of both arterial and venous grafts midterm and late patency. Thus, the influence of the nature of the bypass conduit, the morphologic features of the recipient vessel, the graft hemodynamic variables, and patient-specific variables could be studied individually and included in a multivariate analysis.
We demonstrated by univariate analysis a striking influence of the nature of bypass conduit on patency, the patency rate of arterial grafts being by far superior to venous grafts. This has been a well-established concept since the works of Lytle [12] and Zeff [13] and their colleagues. However, the nature of the bypass conduit was not selected in the multivariate analysis as a variable influencing independently graft outcome, the resistance of the run-off bed being dominant. This can be explained by the fact that our study involves a majority of midterm (<5-year interval) angiographic controls and that the attrition rate of vein grafts starts to markedly exceed that of mammary artery grafts at 5 years after the operation [12]. The same observations were made by van der Meer and associates [14] who demonstrated the absence of difference in adjusted risk of 1-year occlusion rates between internal thoracic artery grafts and vein grafts. Rather, the location of distal anastomosis and lumen diameter of the grafted coronary artery were shown to be predictors of occlusion.
Regarding the morphologic features of the recipient vessels, there was no significant difference in the patency rate with varying degrees of coronary arterial stenosis or total occlusion. This corroborates the findings of Crosby and associates [15], who demonstrated that the patency rate of saphenous vein grafts was not significantly lower even in vessels with less than 60% stenosis, despite the concept that competitive flow predisposes to graft occlusion. In contrast, the coronary artery diameter measured at the level of the anastomosis was a good predictor of graft patency as the occlusion rate increased strikingly in coronary vessels with an internal diameter less than 1.5 mm. Indeed, not only is a larger coronary artery easier to graft but its distal bed is also better developed, as it was demonstrated that a maximally dilated coronary artery luminal cross-sectional area is linearly related to the volume of muscle it perfuses [16].
Several intraoperative hemodynamic parameters were found to be associated with postoperative graft occlusion by univariate analysis: reduced flow, reduced velocity, increased resistance, and increased pulsatility index. In multivariate analysis, increased resistance was the only predictor among flow-derived data. Accordingly, Moran and associates [2] measured a high resistance, ranging from 2.5 to 8.0 PRU, in four instances of graft occlusion.
A combination of patient- and graft-specific parameters was assessed in a multivariate analysis. Again, increased resistance was demonstrated to strongly affect graft patency. The influence of increased interval of the control angiography from the initial procedure is obvious because the risk of graft occlusion increases with time, which confirms previous studies [12].
Although a fairly satisfactory predictive value could be attributed to intraoperative graft hemodynamic assessment regarding the risk of midterm occlusion, these measurements were useless for prediction of midterm stenosis. Intraoperative hemodynamic parameters did not differ between intact and stenosed grafts (see Table 3; Fig 1). Stenosis development is related to scar tissue or to degeneration of a graft, a phenomenon that cannot be detected at the time of the procedure and is predominantly patient related. In addition, the localization and the degree of stenosis is very variable. Fortunately, the prediction of late stenosis development is much less crucial than the prediction of late graft occlusion. Lytle and colleagues [17] demonstrated that patients with circumflex or right coronary artery vein graft stenosis had a survival equivalent to that of their control group. In contrast, late stenoses in saphenous vein grafts to the left anterior descending coronary artery predicted a high rate of death and cardiac events and were an indication for operation. However, at present saphenous veins are rarely used as conduits to bypass that territory.
This study has several limitations. First, it must be emphasized that our series is selected because a majority of our patients were restudied for recurrent symptoms, and that the conclusions drawn may not be applied generally to all patients after coronary revascularization. Yet, for the purpose of the study this selection constitutes an advantage. Second, measurements taken at operation may not truly reflect the capacity of the graft to carry flow because the heart may not have fully recovered from the consequences of ischemic arrest. However, consecutive flow measurements made from aortic unclamping until the end of the procedure show that the highest flow was obtained immediately after weaning from cardiopulmonary bypass [18]. Furthermore, induction of maximal blood flow by papaverine injection directly into the graft did not improve significantly the predictive value of previous flow studies [1, 2]. Third, the analysis of the results poses an important methodologic problem attributable to the characteristics of 1 patient (ie, age, diabetes, hypercholesterolemia); these are correlated with the various features of several bypass grafts and recipient arteries (ie, flow, coronary diameter). Indeed, for the purpose of the analysis, each bypass graft was considered individually. Thus, patient-related factors, such as diabetes, will influence the fate of all the grafts of a given patient and thus will be taken into account several times (equivalent to the amount of distal anastomoses) in logistic regression analysis. Therefore, the generalized estimating equation approach, which takes into account and corrects the latter bias, was preferred to the more usual multiple regression analysis. Conversely, to realize Kaplan-Meier analysis, we had to admit that the fate of all grafts was independent, although clusters of grafts from the same patient were obviously exposed to an unique risk factors set. Finally, the present study is limited to early and midterm angiographic follow-up. Although the operation-to-catheterization interval was longer than in the majority of the previous studies, it rarely exceeded 5 years. On the other hand, determinants of early graft closure differ from the determinants of late graft closure [12, 19]. Thus, after a longer follow-up, the marked influence of intraoperative hemodynamic measurements may be overshadowed by risk factors for atherosclerosis and by the nature of the bypass conduits.
This study has several clinical implications. Intraoperative hemodynamic graft assessment and particularly resistance measured in the grafts may provide a good prognostic index and also be of value in the selection of patients for postoperative angiographic evaluation. In addition, graft resistance as determined at the initial operation may represent a valuable guide to the possible success of repeat operation, should this become necessary. Arterial conduits had a patency rate superior to that of saphenous vein grafts. However, the nature of the conduit, like patient-specific parameters such as sex, smoking habits, hypertension, lipid abnormalities, and diabetes, had no bearing on midterm graft patency by multivariate analysis.
| Appendix 1. Variables used in the multivariate analysis |
|---|
|
|
|---|
|
a 1 = normal; 2 = atheromatous plaque at the anastomosis; 3 = diffusely atheromatous; 4 = very atheromatous; 5 = need for endarterectomy.
b 1 = internal thoracic artery; 2 = epigastric artery; 3 = gastroepiploic artery; 4 = saphenous vein.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Di Giammarco, M. Pano, S. Cirmeni, P. Pelini, G. Vitolla, and M. Di Mauro Predictive value of intraoperative transit-time flow measurement for short-term graft patency in coronary surgery. J. Thorac. Cardiovasc. Surg., September 1, 2006; 132(3): 468 - 474. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Balacumaraswami, Y. Abu-Omar, B. Choudhary, D. Pigott, and D. P. Taggart A comparison of transit-time flowmetry and intraoperative fluorescence imaging for assessing coronary artery bypass graft patency J. Thorac. Cardiovasc. Surg., August 1, 2005; 130(2): 315 - 320. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Balacumaraswami, Y. Abu-Omar, K. Anastasiadis, B. Choudhary, D. Pigott, S.-K. Yeong, and D. P. Taggart Does off-pump total arterial grafting increase the incidence of intraoperative graft failure? J. Thorac. Cardiovasc. Surg., August 1, 2004; 128(2): 238 - 244. [Abstract] [Full Text] [PDF] |
||||
![]() |
P.-G. Chassot, P. van der Linden, M. Zaugg, X. M. Mueller, and D. R. Spahn Off-pump coronary artery bypass surgery: physiology and anaesthetic management{dagger} Br. J. Anaesth., March 1, 2004; 92(3): 400 - 413. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Bardell, G. Hamilton, D. H. Bergstrome, and M. J. Ali Anastomotic quality in CABG: when life depends on millimetres Can. Med. Assoc. J., October 1, 2002; 167(7): 783 - 783. [Full Text] [PDF] |
||||
![]() |
J. A. Wolfe The coronary artery bypass conduit: II. Assessment of the quality of the distal anastomosis Ann. Thorac. Surg., December 1, 2001; 72(6): S2253 - 2258. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Shin, R. Yozu, A. Mitsumaru, Y. Iino, K. Hashizume, T. Matayoshi, and S. Kawada Intraoperative assessment of coronary artery bypass graft: transit-time flowmetry versus angiography Ann. Thorac. Surg., November 1, 2001; 72(5): 1562 - 1565. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. D'Ancona, M. Ricci, J. Bergsland, T. A. Salerno, and H. L. Karamanoukian Graft Patency Verification in Coronary Artery Bypass Grafting: Principles and Clinical Applications of Transit Time Flow Measurement Angiology, September 1, 2000; 51(9): 725 - 731. [Abstract] [PDF] |
||||
![]() |
Y. A.G. Louagie, J. Jamart, M. Buche, P. M. Eucher, and J.-C. Schoevaerdts Reply Ann. Thorac. Surg., April 1, 2000; 69(4): 1301 - 1301. [Full Text] [PDF] |
||||
![]() |
G. D'Ancona, H. L. Karamanoukian, T. A. Salerno, M. Ricci, and J. Bergsland Flow measurement in coronary artery surgery Ann. Thorac. Surg., April 1, 2000; 69(4): 1300 - 1301. [Full Text] [PDF] |
||||
![]() |
G. D'Ancona, H. L. Karamanoukian, M. Ricci, S. Schmid, J. Bergsland, and T. A. Salerno Graft revision after transit time flow measurement in off-pump coronary artery bypass grafting Eur. J. Cardiothorac. Surg., March 1, 2000; 17(3): 287 - 293. [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 |