| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2006;132:468-474
© 2006 The American Association for Thoracic Surgery
Cardiopulmonary Support and Physiology |
Division of Cardiac Surgery, University of Chieti-Pescara, Chieti, Italy.
Received for publication November 1, 2005; revisions received December 21, 2005; accepted for publication February 6, 2006. * Address for reprints: Gabriele Di Giammarco, MD, Division of Cardiac Surgery, "S Camillo de Lellis" Hospital, Via Forlanini 50, University of Chieti-Pescara, Chieti, Italy (Email: gabriele.digiammarco1{at}tin.it).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
METHODS: Of 3567 patients submitted to isolated myocardial revascularization from June 1997 through June 2003, 157 (4.4%) underwent both intraoperative transit-time flow measurement and angiography at follow-up. Thirty-six have been revascularized on a beating heart. Three hundred four grafts, 227 arterial conduits, and 77 saphenous vein grafts were checked.
RESULTS: No patients died, and none of them had an acute myocardial infarction within 12 months after the operation. After a mean of 6.7 ± 4.8 months from the operation, 266 grafts (group A) were completely functioning, whereas 38 grafts (group B) had failed. The transit-time flow parameters recorded in the latter group had significantly lower mean flow and higher pulsatility index and percentage of backward flow values at both univariate and multivariate analysis. Moreover, mean flow values of 15 mL/min or less, pulsatility index values of 3.0 or greater, and percentage of backward flow values of 3.0% or greater were found to be independent variables for higher incidence of graft failure.
CONCLUSIONS: Transit-time flow measurement represents a quick, easy, and reproducible method for intraoperative evaluation of graft function. The combination of the 3 major parameters (mean flow, pulsatility index, and percentage of backward flow) results in the chance to predict a graft failure (either anatomic or functional) within the first postoperative year.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
|
The incidence of perioperative graft failure has been estimated to be from 5% to 11%.1-5 Early graft patency can influence either early or late outcome after coronary artery bypass grafting (CABG).6 Even if angiography still represents the gold standard for anatomic evaluation, the chance of intraoperative assessment of the quality of the anastomosis has been an unachieved goal since CABG was started.
Several methods have been introduced over the last decade,7-9 with the aim to check graft viability. Transit-time flow (TTF) measurement has been reported to be a suitable method for easy and quick intraoperative functional graft assessment,10-13 independent of vessel size and shape.8
Many authors have already validated this method, comparing TTF with intraoperative14 or postoperative15 angiography. The aim of this study was to evaluate to what extent TTF parameters can be considered as predictors of graft failure in the early follow-up period.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Of 3567 patients submitted to isolated CABG from June 1997 through June 2003, 157 (4.4%) underwent both quality control methods. Mean age was 63 ± 9 years. Twenty-eight (17.8%) patients were women. The mean value of the logistic EuroSCORE across the total population was 5.5%.
Of 304 conduits checked by means of either intraoperative TTF measurement or postoperative angiography, 227 were arterial conduits, and 77 were saphenous vein grafts. Among the arterial conduits, 157 were left internal thoracic artery, 62 were right internal thoracic artery, 3 were right gastroepiploic artery, 2 were radial artery, and 3 were inferior epigastric artery conduits. One hundred eighty anastomoses were grafted on the left descending artery, 77 on the circumflex system, and 47 on the right coronary artery system.
Patient Selection
Anatomic selection criteria to off-pump surgery were vessel size (>1.2 mm) and freedom from diffuse coronary calcifications. Clinical selection criteria were the absence of mechanical instability, electrical instability, or both in the operating theater, with the final decision depending on the expertise of the surgical team responsible for the operation.
Surgical Technique
On-pump
Cardiopulmonary bypass (CPB) was instituted by means of cannulation of the ascending aorta and right atrium. A standard circuit with a hollow-fiber membrane oxygenator and a roller or centrifugal pump was used. Body temperature during CPB ranged between 34°C and 37°C. Myocardial protection was achieved by means of intermittent antegrade warm blood cardioplegia.
Off-pump
One hundred twenty-one patients were revascularized on a beating heart. The technique of exposure and stabilization of target coronary vessels changed over time. In the first period we used a method previously reported by our group.16 In the most recent years, apical suction was used to expose the target vessels (Xpose; Guidant Corp, Cupertino, Calif); stabilization was achieved by means of pressure (Access Ultima System, Guidant Corp) or suction (Axius Vacuum 2 System, Guidant Corp). The target vessel was occluded with different methods over time. In the first part of our experience, a 5-0 Prolene suture that was passed on a small piece of silicon tubing and then gently snared was used. Recently, the shunting procedure (Chase Medical, Dallas, Tex) became the method of choice.
TTF Measurement Technique
The theoretic basis and procedure of TTF measurement have already been reported.13-15,17,18 The TTF measurement device used in this study was the CardioMed Flowmeter (Medi-Stim AS, Oslo, Norway). In case of off-pump surgery, TTF trace was recorded after each graft had been completed; in case of on-pump procedure, TTF was evaluated after weaning from CPB, after protamine, and immediately before chest closure. We always considered the trace recorded immediately before the chest closure for the analysis. The size of the probe was chosen, taking care to fit the conduit caliber.
In case of harvesting of internal thoracic arteries as skeletonized vessels (the vast majority of cases), a good coupling with the probe was easily achieved. In 8 patients a pedicled graft (radial artery, gastroepiploic artery, or inferior epigastric artery) was used. In this case a small segment of the graft was skeletonized to improve the contact with the probe.
In all cases TTF measurement was recorded at a systemic systolic pressure of 100 mm Hg.
Definition of Terms
Failed graft
Every graft showing a type B or 0 lesion at angiography, according to the Fitzgibbon classification,19 was considered a failing graft. Twelve-month follow-up was chosen as a time span cutoff point within graft failure that could be considered a result of technical problems.
TTF analysis
TTF analysis can provide the following parameters: mean flow calculated across 5 cardiac cycles(Qmean), maximum flow recorded in 1 cardiac cycle (Qmax), minimum flow recorded through 1 cardiac cycle (Qmin), pulsatility index (PI) as the ratio between the difference ((Qmax–Qmin)/Qmean), and percentage of backward flow (%BF) as the amount of flow through the graft directed backward across the anastomotic site (Figure 1). The latter is defined as that percentage of the area of 1 cardiac cycle drawn by the trace below the zero flow line compared with the total flow area of the same cycle. In case of nonoccluded or well-collateralized left anterior descending coronary artery grafted by left internal thoracic artery, the BF is produced by the time delay running between the systolic wave onset through the native coronary system and the systolic wave through the graft. This delay could be assigned to the anatomic distance from the coronary ostia and the origin of internal thoracic arteries from subclavian arteries. The PI and %BF values are positively correlated (r = 0.82).17
|
Statistical Analysis
The 2 groups were compared by using unpaired 2-tailed t tests for continuous variables or the 
2 test for categoric variables. Logistic regression was used to obtain risk factors for graft failure within the first postoperative year. Optimal cutoff values of Qmean, PI, and %BF to predict graft failure were determined by means of receiver operating characteristic (ROC) curve analysis. The area under the curve, 95% confidence limit, and P value are also reported. The optimal cutoff value was defined as providing maximal sensitivity and specificity. SPSS software (Chicago, Ill) was used for analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Angiographic results
After a mean time interval of 6.7 ± 4.8 months after the operation, 266 grafts were fully patent (group A), and 38 grafts (group B) were failing, with the overall perfect patency rate19 being 87.5%. According to coronary distribution, the patency rate was 90% on the left anterior descending coronary artery territory, 87% on the circumflex system, and 78.7% on the right coronary artery system (P = 0.561). The patency rate was 86.1% in off-pump surgery and 88.3% in on-pump surgery (P = 0.561).
TTF analysis
The Qmean of the total group was 28 ± 21 mL/min. One hundred five grafts had a Qmean value of equal to or less than 15 mL/min. The mean PI value was 3.2 ± 5.9. The %BF value was 3.6% ± 9.0%. Failed grafts had significantly lower Qmean values along with higher PI and %BF values compared with those of the patent grafts at univariate and multivariate analysis (Table 1). This finding was confirmed either in case of recurrence of angina or in asymptomatic patients. Moreover, Qmean values of 15 mL/min or less, PI values of 3.0 or greater, %BF values of 3.0% or greater, and the absence of BF represent independent variables for higher incidence of graft failure at follow-up. The ROC analysis is reported in Table 2.
|
|
Qmean (OR, 0.86; P = .002), PI (OR, 1.3; P = .031), and %BF (OR, 1.1; P = .041) values were confirmed to be predictive parameters of graft failure, even in case of venous grafts.
The prediction model and the cutoff values based on ROC analysis according to the type of graft are reported in Table 2.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
is one of the most important risk factors for perioperative myocardial infarction and hemodynamic instability.3,20 According to our database, early mortality dramatically increases in case of either perioperative myocardial infarction (from 1.4% to 66.7%, P < .001) or hemodynamic instability (from 2.3% to 17.0%, P < .001). Therefore we strongly believe that the possibility of intraoperative assessment of graft function is crucial in determining the results of CABG, independent from the surgical strategy chosen. An intraoperative control method should be easy to handle, not time consuming, minimally invasive, easily meaningful, and relatively cheap; in addition, it should offer objective parameters more than qualitative criteria.
Although intraoperative angiography is still considered the gold standard, it presents many shortcomings: manpower, large instrumentation, significant invasiveness because of the need of contrast dye injection, long operating time, and high cost.14,21 Epicardial ultrasonography uses a 13-MHz probe to visualize the geometry of the anastomotic site.22,23 Added value is given by color Doppler technology that can lead to more complete intraoperative assessment of the anastomosis.23 Epicardial ultrasonography, in the best technologic setup, is able to provide a nice view of the anastomotic site, but functional data regarding runoff are missing.22
Recently, a novel intraoperative imaging technique was reported to evaluate graft patency.21 The SPY imaging system (Novadaq Technologies, Inc, Toronto, Canada) is based on the properties of indocyanine green to become fluorescent when illuminated by laser energy (806 nm). The fluorescence is captured by a charge-coupled device video camera at a rate of 30 images per second. Although this system is relatively noninvasive and safe and can be reproduced more easily than standard angiography, it shows some limitations, as reported by the same author.21 The technique cannot quantify the amount of flow running through the graft; in addition, there is quite often a lack of objective information about the quality of the distal anastomosis because of multiple influences of surrounding soft tissues. On the other hand, it can definitely help in assessing the state of proximal saphenous vein graft anastomosis and graft body, provided the conduit is being harvested without surrounding tissues, as in the case of vein grafts or skeletonized arteries. Concerning the financial aspects, TTF measurement provides lower costs per procedure compared with dye imaging methods, even for the chance to take additional measurements at the same cost.
The possibility to detect graft flow intraoperatively has been evaluated by many surgeons since the early 1970s by using an electromagnetic flowmeter (EMF).24 However, the ability of EMFs in predicting postoperative angiographic patency has not been proved. Louagie and colleagues,7 analyzing more than 600 samples, stated that EMFs can be altered by probe fitting, nonconstant zeroing, and hematocrit value.
Louagie and coworkers25 reported more than 900 graft assessments using pulsed Doppler flowmetry. Seven (2%) of them were identified as failing by means of abnormal waveform configuration, low diastolic flow, and high PI. After the revision, all these parameters improved. In a further study26 the same authors reported the angiographic results of 214 grafts checked by means of pulsed-wave velocity flow Doppler scanning at the time of the operation. Reduced Doppler flow velocity was not found to be a risk factor for lower patency rate at multivariate analysis. This method presents many limitations: probe positioning (angle), motion artifacts, sample size, flow velocity profile, and vessel diameter.27 In other words, pulsed-wave Doppler scanning provides nice information about flow velocity but no data concerning the volume of flow.
Use of a transit-time flowmeter is reported to be a noninvasive, relatively cheap, and time-saving procedure,8,10-13 working independently from vessel size, shape, and Doppler angle.8 Many authors have already validated the method, comparing TTF with intraoperative14 or postoperative15 angiography. Its main ability is to provide the amount of flow running through the graft. How much this parameter is exact and reliable was demonstrated by Beldi and associates,18 who compared the flow measured by means of TTF with the flow measured by means of volume sampling (true flow) in arteries and veins. They found a high correlation (r = 0.93-0.95) between the 2 measurements.
The main limitation until now toward a wider use of TTF technology has been the lack of objective parameters and clear-cut values to be considered predictive of failure beyond that noisy and almost flat trace, very close to the zero flow line, which is striking evidence of a no-flow situation like that occurring in graft occlusion. Jaber and coworkers15 reported the results of an important survey aimed to test the ability of 19 surgeons to detect anastomotic errors using TTF; the benchmark parameters were the analysis of Qmean values and the flow waveform. More than 70% of them were unable to diagnose critical stenosis at the anastomotic site. In fact, although Walpoth and colleagues11 stated that a Qmean value of 20 mL/min is the cutoff value for intraoperative graft patency, others28 report some cases of patent grafts showing even lower intraoperative Qmean values.
Indeed, Takami and Ina28 analyzed a series of 82 grafts intraoperatively checked by means of TTF and submitted to angiography 2 weeks after surgical intervention. At univariate analysis, they found significant difference in some TTF parameters. The analysis demonstrated that in nonpatent grafts, Qmean values were lower, along with significantly higher PI and %BF values, compared with those seen in patent grafts, even if they did not report any cutoff value.
The ability of TTF measurement to provide on-off information concerning the diagnosis of intraoperative graft occlusion has been well assessed. Many authors13,29 have reported how TTF parameters could vary in failing grafts before and after surgical revision. D'Ancona and associates13 analyzed the TTF findings in 41 revised grafts because of low flow or signal demodulation. After surgical revision, Qmean values increased from 6.6 to 36.3 mL/min (P < .0001), and PI values decreased from 24.4 to 2.8 (P < .0001). Moreover, all but 3 presented a systolic flow pattern without an evident diastolic modulation before revision; after surgical correction, the pattern changed into a different modulation with a diastolic peak of flow higher than the systolic peak, coming back to the same modulation found in normal coronary circulation.17 In a recent study29 TTF was used to check 323 grafts. Seven of them were judged to be failing because of either low flow (Qmean of 5.4 mL/min) or high PI value (11.3). After surgical revision, the Qmean value increased significantly to 26.4 mL/min, and the PI value decreased to 3.1.
A careful analysis of the above-mentioned reports could be helpful in drawing some guidelines for surgeons who want to use TTF for detecting an intraoperative graft failure: a combination of high Qmean values and low PI and %BF values, along with a diastolic waveform pattern, should be considered as a marker of good graft performance with a fairly good probability of a postoperative course free from ischemic episodes. Moreover, a prompt detection of intraoperative graft failure can reduce the cost of the CABG procedure, avoiding return to the operating room for hypoperfusion syndrome.
Nevertheless, none of the above-cited authors was able to define cutoff values for the given parameters.
Our retrospective study was carried out on 304 grafts whose TTF parameters were recorded at the time of the operation. All the grafts underwent angiographic control within the first postoperative year. In the aim to assess the predictive power of TTF parameters for graft patency in the first postoperative year, we analyzed those 38 grafts failing at the angiographic control. Stepwise logistic regression was able to prove that a Qmean value of 15 mL/min or less (OR, 21.2), a PI value of 3 or greater (OR, 3.5), and a %BF value of 0 (OR, 2.1) or 3.0% or greater (OR, 3.5) can be considered cutoff values predictive for early graft failure.
Of 38 failing grafts, 5 (13.2%) showed low Qmean values, high PI values, and lack of BF along with a spiky systolic pattern (1 arterial and 4 venous grafts).13 This is a clear picture of a failure caused by anastomotic trouble (intimal flap, purse-string effect, heel or toe tapering, and acute thrombosis); all of them belong to the beginning of our experience with flowmetry, when we were somehow skeptical in revising the anastomosis on the basis of this method. We progressively increased our confidence with TTF measurement so that we sooner became strong believers, considering TTF measurement as mandatory in all grafts, irrespective of the surgical strategy used.
Beyond the chance to intraoperatively detect a graft failure and taking into account that notwithstanding the intraoperative revision we continue to observe a given percentage of angiographic evidence of early occlusion, we aimed to highlight other pictures that could predict an early failure.
Among the remaining 33 failing grafts, 20 (11 arterial and 9 venous grafts) showed a Qmean value of 15 mL/min or less with a PI value of 3 or greater and a %BF value of 3.0% or greater compared with 6.8% of functioning grafts (P < .001). Logistic regression confirmed that a combination of these 3 parameters is strongly associated with graft failure (OR, 15.3; P < .001), either in arterial (OR, 24.6; P < .001) or venous (OR, 6.6; P = .002) grafts.
The combination of low Qmean and high PI values in the presence of high %BF values needs some speculation. The evidence of some degree of %BF, as reported in the literature,17 is per se the proof of the anatomic patency of the anastomosis. In fact, the formula for PI calculation clearly demonstrates that it is directly proportional to the difference between Qmax and Qmin and inversely proportional to Qmean. Therefore in case of very low Qmean with a wide difference between maximum flow and minimum flow recorded (as happens in presence of a noticeable %BF), the PI recorded is fairly high (Figure 2). In similar situations the %BF value proves the patency of the anastomosis, but the high PI value is a strong predictor of late failure because the higher the BF through the graft, the lower its downward runoff. In our clinical practice we used to describe this situation as functional occlusion. The situation can occur in case of arterial Y grafting directed to different coronary territories with unbalanced stenoses in a feature known as competition of flow. The TTF measurement makes this condition immediately evident, showing the normalization of the waveform during temporary occlusion of the stealing branch. In case the trace modulation is totally below the zero flow line with a reversal after occlusion of the stealing branch, there is a strong indication to disconnect the branch from the Y conduit, performing a new proximal anastomosis on the ascending aorta. In case of modulation across the zero flow line, a pharmacologic test could be helpful to simulate a hemodynamic condition similar to a standard lifestyle. In our institution we used to perform 2 different pharmacologic tests addressed to different issues: the norepinephrine test, involving injection of a 1-mL (0.02 mg/mL) bolus of the drug (Noradrenalina tartratol SALF Spa, Bergamo, Italy), allows us to increase systolic pressure, and the dobutamine test, involving an injection of a 20 µg/kg bolus of dobutamine (Dobutrex; Eli Lilly Italia, Sesto Fiorentino, Italy), increases myocardial oxygen demand.30 If the Qmean value improves in these particular conditions, the probability of an anastomotic failure is very low.
|
|
In conclusion, TTF is a quick, easy, noninvasive, and reproducible method to evaluate graft functioning intraoperatively. It does not depend on vessel size, shape, and Doppler angle. Today its main limitation is the lack of standard interpretation of the TTF parameters to establish which graft can fail in the postoperative period. This retrospective study aimed to give surgeons some more indications to better analyze the graft functioning by mean of this technology. The Qmean, PI, and %BF values are the most important factors to be considered for the analysis. In more detail, a Qmean value of 15 mL/min or less should lead the surgeon to strongly suspect a graft failure. PI and %BF values can be helpful in distinguishing a functional failure because of incorrect indication to revascularization and/or incorrect surgical strategy from technical error either in graft harvesting or in performing the anastomosis. In this subset of patients, some pharmacologic tests could help to predict the fate of the graft.
| See related editorial on page 466.
|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  H. B. Ward, R. F. Kelly, and E. K. Weir Assessment of graft patency during coronary artery bypass graft surgery mitigating the risk. J. Am. Coll. Cardiol. Img., May 1, 2009; 2(5): 613 - 615. [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Takami and H. Masumoto Effects of Intra-Aortic Balloon Pumping on Graft Flow in Coronary Surgery: An Intraoperative Transit-Time Flowmetric Study Ann. Thorac. Surg., September 1, 2008; 86(3): 823 - 827. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Tokuda, M.-H. Song, H. Oshima, A. Usui, and Y. Ueda Predicting Midterm Coronary Artery Bypass Graft Failure by Intraoperative Transit Time Flow Measurement Ann. Thorac. Surg., August 1, 2008; 86(2): 532 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. D'Ancona, J. M. Hartman, F. Bartolozzi, E. van Deel, D. J.G.M. Duncker, A. J.J.C. Bogers, M. Pilato, and A. P. Kappetein Epicardial coronary artery Doppler: validation in the animal model Interactive CardioVascular and Thoracic Surgery, August 1, 2008; 7(4): 634 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Tokuda, M.-H. Song, Y. Ueda, A. Usui, and T. Akita Predicting Early Coronary Artery Bypass Graft Failure by Intraoperative Transit Time Flow Measurement Ann. Thorac. Surg., December 1, 2007; 84(6): 1928 - 1933. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Tokuda and M.-H. Song Internal Thoracic Artery Grafts and Upper Extremity Arteriovenous Fistula Ann. Thorac. Surg., December 1, 2007; 84(6): 2138 - 2138. [Full Text] [PDF]
|
|
|
|
 |
|
 N. Becit, B. Erkut, M. Ceviz, Y. Unlu, A. Colak, and H. Kocak The impact of intraoperative transit time flow measurement on the results of on-pump coronary surgery Eur. J. Cardiothorac. Surg., August 1, 2007; 32(2): 313 - 318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Desai and S. E. Fremes Reply to the Editor J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1396 - 1397. [Full Text] [PDF]
|
|
|
|
|
|
W. Hassanein, A. Albert, I. Florath, Y. Y. Hegazy, U. Rosendahl, S. Bauer, and J. Ennker Concomitant aortic valve replacement and coronary bypass: the effect of valve type on the blood flow in bypass grafts Eur. J. Cardiothorac. Surg., March 1, 2007; 31(3): 391 - 396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kulik, F. D. Rubens, and M. Ruel Intraoperative indocyanine green angiography: Ready for prime time? J. Thorac. Cardiovasc. Surg., February 1, 2007; 133(2): 592 - 593. [Full Text] [PDF]
|
|
|
|
|
|
F. Duhaylongsod Artificial artificial intelligence: Surgeon intuition and computers to predict graft patency. J. Thorac. Cardiovasc. Surg., September 1, 2006; 132(3): 466 - 467. [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 |
