| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;108:532-539
© 1994 Mosby, Inc.
SURGERY FOR ACQUIRED HEART DISEASE |
Tokyo, Japan
Address for reprints: Hiroshi Nishida, MD, Department of Cardiovascular Surgery, The Heart Institute of Japan, Tokyo Women's Medical College, 8-1 Kawadacho, Shinjuku-ku, Tokyo 162, Japan.
Abstract
From December 1988 to December 1992, 174 patients (160 men, 14 women, mean age 59.7 years, range 38 to 79 years) underwent coronary artery bypass grafting with the right gastroepiploic artery. The graft was anastomosed to the right coronary artery (n = 137), the circumflex artery (n = 18), the left anterior descending artery (n = 23), and the diagonal artery (n = 1). Three early deaths (1.7%) and one late death (0.6%) occurred. Graft patency and flow were evaluated noninvasively in 44 of the patients, selected at random between 1990 and 1993. They underwent transcutaneous Doppler echocardiographically to detect postoperative gastroepiploic artery flow. The patients were divided into two groups on the basis of the angiographic study: group I, good patency (n = 38); group II, poor flow in the graft or more than 75% stenosis of the anastomosis (n = 6). Biphasic Doppler flow signals were identified in 39 patients (88.6%) (group I, 35/38; group II, 4/6). The ratio of diastolic to total flow measured by time-velocity integral was 0.68±0.07 in group I and 0.32±0.09 in group II (p < 0.001). We conclude that the right gastroepiploic artery is an effective graft and that Doppler echocardiography may be a useful tool to noninvasively evaluate the patency and flow of the gastroepiploic artery graft as a coronary artery graft. (J THORAC CARDIOVASC SURG 1994;108:532-9)
In recent years, the trend in coronary artery bypass grafting (CABG) has been toward a gradual transition to the routine use of a multiple complex arterial grafting technique and new arterial conduits because of their superior long-term patency. 
1 The recently introduced right gastroepiploic artery (RGEA) 2-4 seems to be a promising alternative arterial conduit in terms of satisfactory midterm results. 5 Since we started using the RGEA in December 1988, we have expanded its indications to include routine use. In this report, surgical results, angiographic findings, and postoperative noninvasive flow evaluation with transcutaneous Doppler echocardiograms of the RGEA graft are reported and evaluated.
PATIENTS AND METHODS
From December 1988 to December 1992, 174 patients underwent CABG with the RGEA. There were 160 male patients and 14 female patients with a mean age of 59.7 years and a range of 38 to 79 years. Postoperative follow-up ranged from 4 to 52 months with a mean of 24 months. Eighty-five percent of patients had either triple vessel disease or left main disease. Seventeen patients had a previous CABG (10%). The preoperative left ventricular ejection fraction measured by biplane ventriculography ranged from 11% to 82% with a mean of 49% (Table I).
|
Transcutaneous Doppler echocardiography.
Transcutaneous Doppler ultrasonography was used to evaluate the flow in the RGEA graft to the right coronary artery (RCA) in 44 randomly selected patients since October 1990. The instrument used was the Acuson 128XP/10 computed sonography system (Acuson, Inc., Mountain View, Calif.). By means of a 7.5 MHz linear scan method (Acuson linear transducer L558), the RGEA graft flow was detected by color Doppler echocardiography. The transducer was positioned below the right chondral margin 1 to 2 cm lateral to midline with the ultrasonic beam angled upward at a 5-degree angle tilted toward the left shoulder. A tubular structure containing an upward biphasic Doppler flow signal posterior to the abdominal wall yet anterior to the liver was judged to be the patent RGEA graft (Fig. 1). After the graft flow was detected, the flow-velocity curve was recorded with the pulsed Doppler method. Peak velocity and time-velocity integral were measured separately during systole and diastole (Fig. 1). Time-velocity integral is an area under the flow-velocity curve and means flow volume.
|
We analyzed the relationship between the preoperative degree of proximal stenosis and postoperative pattern of flow dependency in the RCA distal to the RGEA anastomosis in 89 patients with selective angiography. The degree of proximal stenosis was defined according to the American Heart Association classification. 6 A 75% stenosis was present in 18 patients, 90% stenosis in 15 patients, and 99% to 100% occlusion in 56 patients. The pattern of flow dependency distal to the RGEA anastomosis was classified into four categories from the angiographic findings obtained from both native and RGEA injections.
2 test. Results are expressed as mean ± standard deviation. RESULTS
The RGEA was used as an in situ graft in 173 patients, and in one patient it was used as a free RGEA graft. Sequential RGEA grafting was performed in nine patients: a combination of proximal RCA and posterior descending artery in two patients, the atrioventricular artery and the posterior descending artery in three patients, and the RCA and posterolateral branch of the circumflex artery in four patients. The total number of distal coronary anastomoses with the RGEA graft was 179. Arteries grafted with the RGEA are listed in Table II, with the RCA being the most frequent target. Among 137 RCA bypass grafts with the RGEA, 98 anastomoses (72%) were to the distal RCA branches such as the posterior descending artery and atrioventricular branch. Eight patients with extremely poor distal runoff underwent concomitant endarterectomy of the distal RCA.
|
Clinical outcome.
Three patients (1.7%) died within 30 days after the operation. One patient died of postoperative bleeding, one patient died of mediastinitis and septic shock on postoperative day 19, and one patient died of apnea from cardiac arrest on postoper ative day 11. Four patients (2.3%) had a perioperative myocardial infarction. Six patients had pancreatitis (3.4%). Although four patients recovered within 2 weeks, an abnormal pancreas-specific amylase level persisted more than 1 month in two patients. Early, late, and angiographic results are summarized in Table III.
|
Transcutaneous Doppler echocardiography.
Flow in the RGEA graft was detected in 39 of 44 patients (89%). In group I (n = 38) of the angiographic study the RGEA graft was identified by Doppler echocardiography in 35 patients (92%), and in group II (n = 6) the RGEA was identified in four patients (67%). Anatomically, no other arteries of the same size, flow direction, and flow pattern exist in this region; therefore, there were no cases of false-positive visualization of the RGEA graft.
Peak velocity.
Systolic peak velocity was 48.0 ± 18.1 cm/sec in group I and 35.0 ± 22.0 cm/sec in group II, with no statistically significant difference (Fig. 2). Diastolic peak velocity was 34.9 ± 14.0 cm/sec in group I and 15.3 ± 10.0 cm/sec in group II (p < 0.05). The ratio of systolic to diastolic peak velocity was 1.42 ± 0.41 in group I and 2.62 ± 0.60 in group II (p < 0.01).
|
|
The relationship between the preoperative degree of proximal RCA stenosis and postoperative flow dependency of the distal RCA (Fig. 4) was as follows: With a proximal stenosis of 75% (n = 18), RGEA occlusion was seen in two patients (11%), native-dependent flow in four (22%), balanced flow in nine (50%), and RGEA-dependent flow in three (17%). A proximal stenosis of 90% (n = 15) demonstrated native-dependent flow in one (7%), balanced flow in five (33%), and RGEA-dependent flow in nine (60%). When the proximal stenosis was 99% to 100% and severe (n = 56), RGEA occlusion was seen in three patients (5%). The other 53 patients (95%) had significant RGEA-dependent flow (p < 0.001).
|
As encouraging reports 5,7-12 and our own clinical experiences with the RGEA graft used for CABG have accumulated, we have gradually expanded the indications. Without setting any particular absolute contraindications, we do believe that urgent operations with severely impaired hemodynamics and significant RGEA or abdominal aortoiliac disease are not clinical situations in which to use the RGEA. Other factors such as increased age, diabetes mellitus, poor left ventricular function, and chronic renal failure are not considered to be contraindications for RGEA grafting. Seven patients had chronic renal failure and required dialysis. These patients were temporarily managed with peritoneal dialysis after CABG until the hemodynamics stabilized enough for hemodialysis to be resumed. Because the incision in the diaphragm was sealed with fibrin glue, no fluid leaked into the pericardial cavity. One patient returned to continuous ambulatory peritoneal dialysis without any problems, and another patient underwent concomitant total gastrectomy for gastric cancer. 13
Suma and associates 5,10 reported that perioperative mortality and morbidity did not increase with the use of the RGEA. In their 200 consecutive patients, operative mortality was 3% with excellent midterm results. Operative mortality in our series was 1.7% with no life-threatening complications as a direct result of RGEA grafting. Although clinical data show that the RGEA is a safe and useful alternative arterial graft second only to the ITA, flow reserve of the RGEA should be carefully investigated as in the ITA studies by Flemma and associates, 14 who demonstrated 2.7 times higher saphenous vein graft flow than ITA graft flow to the same coronary artery bed. This flow discrepancy also may be true for the RGEA graft.
After this series, we recently encountered inadequate RGEA graft flow in a patient who had persistent right ventricular failure during weaning from cardiopulmonary bypass. This patient had received a small RGEA graft, 1.8 mm in diameter, to a dominant RCA with a mild proximal stenosis, together with ITA grafting to the left anterior descending artery. No technical problems were identified with the RGEA graft. This patient was successfully treated for right ventricular failure by an additional saphenous vein bypass graft to the RCA.
Physiologically, the flow reserve of the RGEA can be no more than that of the ITA. Tedoriya and associates 15 measured the pressure in the ITA and RGEA at their cut ends simultaneously with the pressure in the ascending aorta during 25 clinical CABGs. Systolic pressure was not different, but both diastolic (ascending aorta, 63 ± 2 mm Hg; ITA, 55 ± 2 mm Hg; RGEA, 50 ± 2 mm Hg) and mean (ascending aorta, 75 ± 2 mm Hg; ITA, 65 ± 2 mm Hg; RGEA, 60 ± 2 mm Hg) pressures were significantly (p < 0.01) lower in the RGEA than in the ascending aorta and the ITA. Second, the RGEA graft is longer than the ITA graft, and thus the resistance to flow may be higher in the RGEA than in the ITA. Mills and Everson 9 reported that the pedicled RGEA graft was 18 to 30 cm long (average 23.7 cm). Saito and associates 16 reported the required length of an RGEA to bypass the distal RCA to be about 17 cm, 22 cm for the posterolateral branch and 21 cm for the anterior descending artery. ITA pedicles measured 12 to 18 cm (average 14.5 cm) in our series.
Seki and his associates 17 pointed out that the ITA diameter and the degree of stenosis of the left anterior descending coronary artery influenced graft flow the most. Similarly, findings of our angiographic study revealed that the RGEA graft may not provide additional flow in patients with mild proximal stenosis. Because of this finding, when deciding on the usage of the RGEA graft, one must consider both the size of the RGEA and the degree of proximal stenosis and distal runoff. A small RGEA, less than 2 mm in diameter, should not be used to revascularize the relatively dominant coronary artery with mild proximal stenosis. If ischemia or ventricular dysfunction persists, immediate countermeasures such as additional saphenous vein grafting or intraaortic balloon pumping should be undertaken after the possibility of an anastomotic failure or graft spasm has been excluded.
We have observed no cases of gastric complications, but pancreatitis is developed in six patients (3.4%). In these patients the RGEA was larger than 3 mm. Lytle and his associates 7 reported three cases of transient hyperamilasemia in 15 patients (20%) but clinical pancreatitis was not observed. Mills and Everson 9 have also reported three cases of transient hyperamilasemia in 39 patients (7.7%). Although the exact cause of pancreatitis and hyperamilasemia is still unknown, some ischemic or mechanical damage related to RGEA grafting or the harvesting procedure may play a role in the development of pancreatitis.
Selective angiography of the RGEA is not always easy. Isshiki and associates 18 reported that the rate of successful catheterization of the RGEA was 78% (29/37). Transient vasospasm occurred in three patients (8%). Transcutaneous echocardiography has been sporadically used to detect saphenous vein bypass flow 19,20 and ITA flow. 21 The RGEA can be evaluated more readily by transcutaneous echocardiography than a saphenous vein graft or an ITA graft. There is no interference by lung or bone. The location is motionless and not affected by the beating heart. No matter to what coronary artery the RGEA is anastomosed, the course of the RGEA is always anterior to the liver and, anatomically, there are no other arteries from which it needs to be differentiated. Not only quantitatively, but also qualitatively, the time-velocity integral and its diastolic ratio obtained from a flow-velocity curve seem to correlate well with angiographic assessment of flow quality. Noninvasive echocardiography has also been especially useful in five patients in whom postoperative angiography was not possible because of severe atherosclerosis, marginal renal failure, or persistent pancreatitis.
In conclusion, the RGEA is an effective coronary artery graft and Doppler echocardiography may be a useful noninvasive tool to evaluate its patency and flow.
Appendix: DISCUSSION
Dr. Scot H. Merrick (San Francisco, Calif.).
This is a study on the use of the gastroepiploic artery (GEA) in coronary revascularization. Saphenous vein grafts have had limited durability, and the ITA has had significant advantages with regard to long-term patency and patient survival. Evolution toward the use of multiple arterial grafts constructed with the GEA, as well as the ITA, is logical, but long-term patency issues and patient survival with this approach is still in need of definition. This study impressively demonstrates that the GEA can be used liberally with a high degree of safety and with excellent short-term patency rates. I commend the authors for their excellent clinical results.
When compared with other conduits, the GEA has important differences in histologic composition and in endothelial properties that may influence baseline flow, reserve flow, and resistance to atherosclerosis. These issues need further study. A noninvasive method of interrogating this vessel would be most desirable because GEA arteriography is often difficult and may be hazardous.
I have several questions. First, because vessel diameter and transducer angle may have a significant impact on derived flow data, do you think this introduced significant error in your two patient groups?
Dr. Nishida.
Thank you very much, Dr. Merrick.
As you suggested, the angle between the flow and Doppler signal is important and should be zero. The GEA goes in a dorsal direction after it passes the upper edge of the liver and the insertion in the diaphragm. Around that point, by using the very sophisticated sonography instruments now available, such as the Acuson system, we can launch the Doppler flow signal in any direction, and it is very easy to obtain coaxial relationship between the flow and Doppler echo.
The vessel diameter is an important factor to obtain and discuss the absolute flow volume. Of course we can measure the diameter precisely, but we put more importance on the ratio between systole and diastole than on absolute value. Also, time-velocity integral, which is a kind of second-dimension parameter, seems to be a better indicator than peak velocity.
Dr. Merrick.
It is unclear from your manuscript if your postoperative angiography and Doppler studies were simultaneous. Could you clarify that for me?
Dr. Nishida.
The angiographic study and Doppler examination were not simultaneous, but the Doppler study was performed within 2 days after the angiographic study. Therefore, I believe it is not a big problem.
Dr. Merrick.
Have you evaluated gastroepiploic flow velocity with exercise or with pharmacologic manipulation as a potential measure of flow reserve?
Dr. Nishida.
We have not systematically examined the flow reserve of GEA under exercise or pharmacological manipulation. However, we examined flow reserve of GEA before and after the meal. Although Dr. Suma, one of the pioneers and strong advocates of the GEA, reported an increase of GEA flow after the meal, we could not observe any change in graft flow. The flow was constant before and after the meal.
Dr. Merrick.
My final question relates to your comparison of the degree of native coronary stenosis and graft flow patterns. Many would advocate intraoperative measurement of absolute flow rates after transecting the ITA or GEA vessels as a way of assessing the suitabilities of these conduits. Did you measure intraoperative gastroepiploic flow, and do you think this may influence your graft flow patterns?
Dr. Nishida.
We did not measure free flow of the GEA from the cut end routinely because we cannot know the suitability of the graft until we finish the dissection. Therefore, we judge the suitability by preoperative angiography or the size on site after we release the spasm.
Dr. Louis Brunsting (Ann Arbor, Mich.).
Some concern has been expressed regarding use of the GEA in patients with poor ventricular function who have high end-diastolic pressures, for two reasons: (1) the decreased perfusion pressure and how that relates to early graft patency and (2) the need for inotropic stimulation, because most inotropic and pressor medications cause splanchnic vasoconstriction. Do you have any reservations? Is there an upper limit of end-diastolic pressure or a lower limit of left ventricular function in which you will not use the GEA? If you have to use inotropic drugs or pressors, do you have a preference for which ones to use and at what doses?
Dr. Nishida.
In our series, the lowest ejection fraction was 11%. Poor left ventricular function is not an absolute contraindication for GEA grafting. Essentially, the conditions are the same as for the ITA. However, physiologically, flow reserve of the GEA cannot be more than that of the ITA because the GEA is the third tributary of the aorta next to the celiac artery and the gastroduodenal artery, whereas the ITA is the second tributary of the aorta next to the subclavian artery. Also, the graft length of the GEA is 5 to 8 cm longer than that of the ITA, and it has been shown that mean pressure of the GEA at the cut end is approximately 5 mm Hg lower than that of the ITA. On the basis of these findings, a small GEA less than 2 mm in diameter should not be used to revascularize the relatively dominant coronary artery with mild proximal stenosis and large distal runoff, and it should not be used in patients with very poor left ventricular function. However, we believe a GEA of 3 mm or larger can be used in patients with poor ventricular function.
With regard to inotropic drugs, our first choice is dopamine, just as it is for ITA grafts. We do not compare the different responses to inotropic drugs from graft to graft very much. If the poor flow in the graft causes poor recovery of hemodynamics after cardiopulmonary bypass, we believe keeping perfusion pressure high enough by means of intraaortic balloon pumping or inotropic agents is most important.
Acknowledgments
We acknowledge Ronald K. Grooters, MD, Mid-Iowa Heart Institute, for his editorial assistance and revisions.
Footnotes
From the Departments of Cardiovascular Surgerya and Cardiology,b The Heart Institute of Japan, Tokyo Women's Medical College, Tokyo, Japan. 
Read at Nineteenth Annual Meeting of The Western Thoracic Surgical Association, Carlsbad, Calif., June 23-26, 1993.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
