|
|
||||||||
Eur J Cardiothorac Surg 2006;30:341-346
© 2006 Elsevier Science NL
Review |
ko G. Ne
i
*
evi

ko P.
ukanovi
. Jovi

kovi
Dedinje Cardiovascular Institute, M. Tepi
a 1, 11040 Belgrade, Serbia and Montenegro
Received 23 March 2006; received in revised form 27 April 2006; accepted 15 May 2006.
* Corresponding author. Address: Department of Cardiac Surgery, Dedinje Cardiovascular Institute, M. Tepi
a 1, 11040 Belgrade, Serbia and Montenegro. Tel.: +381 11 3601647/9/724; fax: +381 11 2666392. (Email: nezic{at}EUnet.yu).
| Abstract |
|---|
|
|
|---|
Key Words: CABG Surgery Arterial grafts
| 1. Introduction |
|---|
|
|
|---|
Ten years after CABG, 90% of internal thoracic artery (ITA) grafts are patent and disease free, while about 50% of the saphenous vein grafts (SVG) are occluded, and 25% have been severely stenosed [2]. Impressive long-term patency of the left internal thoracic arteryleft anterior descending coronary artery (LITALAD) graft, coupled with proven long-term survival benefit, has led to its becoming a golden standard of CABG. Significantly better CABG survival when both ITA conduits are used concomitantly compared with a single ITA has been reported [3].
Thus, the use of arterial conduits has expanded beyond the ITAs to include the right gastroepiploic artery, the inferior epigastric artery, and the radial artery (RA). Acar and colleagues [4] strongly recommended revival of the RA conduit (almost completely abandoned due to high occlusion rate reported by Carpantier and colleagues [5] in 1971) having detected several patent RA grafts up to 18 years after the initial myocardial revascularization. Several studies [68] have documented excellent clinical results and satisfactory short- and mid-term patency rates at restudy angiography of radial artery conduits used for CABG. One long-term angiographic study has demonstrated disease-free patency rate of the radial artery graft of 88%, 9 years after surgery [9]. These reports led many other groups to reassess the role of this conduit in coronary bypass procedures.
| 2. Advantages of the radial artery conduit |
|---|
|
|
|---|
| 3. Potential disadvantages of the radial artery conduit |
|---|
|
|
|---|
If there are no other contraindications before harvesting (RA plaque on ultrasound, damaged RA due to trauma of previous cannulations, the presence of an arterio-venous fistula for hemodialysis, vasculitis, carpal tunnel syndromes or Raynaud's disease), removal of the RA does not result in any symptoms of hand ischemia or motor dysfunction. Mid-term (few to 24 months) follow-up studies [12,14] have demonstrated chronic enlargement of the remaining artery (ulnar artery [12] or rather interosseous [14] artery) to compensate blood supply of the forearm.
Acute hand ischemia with serious motor deficiency was reported in two articles [15,16], presenting the same patient with identical angiographic material. In that single case angiographic examination confirmed ulnar artery agenesis although perioperative tests had established adequate collateral circulation. Circulation was restored by performing brachioradial bypass grafting using reversed cephalic vein [15,16].
In a prospective, randomized study, Hata and colleagues [17] have suggested that sharp dissection with scissors and clips may be better for early postharvest forearm circulation and can decrease the incidence of hand numbness. However, there were no differences among the three methods (traditional dissection, electrocautery, or ultrasonic scalpel) with respect to forearm circulation 12 months after radial artery harvesting.
The most common complications noted, occurring in 2.6% to 15.2% of patients [12,18], are sensory abnormalities, especially cutaneous paresthesis and/or numbness in the radial distribution of sensory nerves (lateral antebrachial cutaneous nerve, superficial branch of the radial nerve) due to nerve injury from direct trauma, edema or carpal tunnel hematoma. Long-term persistence of the symptoms were recorded in 10% of the patients, but were considered a constant and significant source of discomfort in only 1% of patients (without a significant limitation in hand activities) [18].
Gaudino and colleagues [19,20] have recently brought to our attention a new and alarming perspective on the possible chronic consequences of radial artery removal on the forearm circulation. Twenty-five [19] and 39 patients [20] submitted to radial artery removal (nondominant arm) for CABG underwent to serial echo-Doppler evaluation of the flow and morphology of the forearm arteries until 10 years follow-up. The peak systolic velocity as well as the intima-media thickness of the ulnar artery has always been higher on the operated side, and this difference reached statistical significance at 10 years follow-up. There was a significantly higher prevalence of atherosclerotic plaques in the ulnar artery of the operated versus control arm (7/25 vs 0/25; p = 0.03, and 11/39 vs 0/39; p = 0.005, respectively). Thus, the authors concluded that radial artery removal for CABG surgery leads to a chronic increase in ulnar flow accompanied by increased intima-media thickness and accelerated atherosclerotic disease. These findings may have potentially important implications for surgical indications and patient management, especially in young patients with a long life expectancy.
Patients who had undergone endoscopic RA harvesting have been reported to have significantly fewer major complications (hematomas, wound infections, neuralgias) than patients who underwent the open RA harvesting technique [21].
In patients with coronary artery disease, RA atherosclerotic involvement is more frequent (up to 6.9%) than that of the gold standard ITA [22]. The RA that is heavily calcified poses technical difficulties intraoperatively apart from indicating a doubtful long-term patency.
It is commonly agreed upon that RA harvest must be performed with gentle mobilization of the RA, with light tractions, minimal touch and minimal diathermy or ultrasonic dissection, thus diminishing endothelial trauma related to harvesting process. The RA is expected to be more vasospastic (reported rate of 410%) compared to other arterial grafts [23] due to its characteristics of being a type III artery (He's [24] classification) and of having predominantly
-adrenergic receptors. The propensity for the RA to go into spasm is likely due to the higher density of muscle cells in the media of this vessel resulting in a significantly higher maximal contractile force in response to vasoconstricting agents (norepinephrine, serotonin, endothelin I, and angiotensin II) generated in response to endothelial damage and dysfunction due to inflammatory response to cardiopulmonary bypass or surgical stress, as well as exogenously administered inotropes or vasoconstrictors [23]. Systemic vasodilators were suggested and empirically used with the beginning of the revival of the interest of the RA, but were found to be ineffective and unnecessary by later reports [25]. Currently, the propensity of the RA to spasm has greatly been reduced using topically different categories of vasodilators, including calcium channel blockers, the phosphodiesterase inhibitors,
-adrenergic antagonists as well as drug mixtures. Clinically useful agents should ideally be effective against a wide variety of endo and exogenously delivered substances causing receptor-mediated as well as receptor-independent vasoconstriction of the RA graft [23]. Currently, no single topical antispasmodic agent fulfils all the criteria necessary to prevent RA vasospasm in perioperative period. Verapamilnitroglycerin solution is effective against a broad range of vasoconstricting agents [23], but its relatively short duration of effects (up to 5 h) limits its usefulness in the clinical practice. Phenoxybenzamine is a more effective agent to prevent
-adrenergic spasm, but has little effect on vasoconstriction caused by noncatecholamine vasoactive mediators. Its prolonged duration [23] of action (up to 18 h in vitro) as well as the ability to prevent catecholamine-mediated vasoconstriction, recommends it as a useful agent in the immediate postoperative period, especially in CABG patients requiring inotropic support. Finally, a combination of agents may be appropriate to combat vasospasm of the RA graft and thereby reduce the likelihood of early graft failure.
It has recently been shown that adenoviral-mediated gene transfer (suggesting a possible future clinical application), causing overexpression of nitric oxide synthase in human RA, can minimize RA vasospasm through the inhibition of voltage-dependent as well as receptor-dependent pathways [26].
In order to prevent delayed vasospasm, oral calcium channel antagonists have been recommended for as long as 1 year after surgery for patients receiving RA conduits. Gaudino and colleagues in prospective, randomized studies have confirmed that calcium channel blocking therapy started immediately after surgery and continued for the first postoperative year [27] as well as beyond 1 year [28] did not affect either radial artery conduit patency or clinical outcomes.
However, the role of calcium channel blocking therapy in the early postoperative period, when radial artery tendency to spasm is thought to be maximal, has yet to be clarified, and its use remains routine in many centers.
| 4. Optimal grafting strategy |
|---|
|
|
|---|
90%, as compared with those coronary-vessel lesions with stenosis of 7089%, were associated with a lower rate of occlusion of the radial artery conduit (5.9% vs 11.8%, p
= 0.03). The concept of competitive flow [23] suggests that graft flow is influenced by native coronary flow. Grafted conduits may, therefore, fare better in conditions of poor native coronary flow, i.e. in the case of high-grade stenosis or occlusion of targeted coronary arteries. Some studies [29,30] have identified target location as predictor for conduit failure, thus lower angiographic patency rates for RA grafts to both the circumflex and right (statistically significant) coronary arteries were reported. These findings remain controversial because there are many studies confirming that target vessel location does not influence RA conduit patency [8,9,31].
There are few options to perform the proximal RA anastomosis. The most commonly used are to the aorta as an aortocoronary conduit or to another graft (usually ITAs) as a composite conduit (in a T or Y fashion). It has been speculated that the histologic and structural characteristics of the RA can render this conduit particularly prone to vessel wall ischemia and consequent intimal proliferation, especially when exposed to the hemodynamic stress due to the sharp increase in dp/dt present in the initial part of the ascending aorta. For this reason some authors have advocated performing the proximal anastomosis of RA grafts to a vascular region with a lower dp/dt, such as an ITA graft [32].
Maniar and colleagues [30] have reported that the site of proximal anastomosis does not appear to influence RA conduit patency. In their study, angiographic (average period to reangiogram 26.1 months) RA conduit patency in aortocoronary versus composite configuration demonstrated no significant difference (72.0% vs 70.5%). Other authors [7,33] have reported excellent midterm (average period to reangiogram 16.2 and 36.8 months, respectively) RA conduit patency (96.8% and 91.9%, respectively) in aortocoronary configuration. Matchless results (patency rate of 99%) were reported [32] for composite configuration (average period to reangiogram 35 months). However, composite configuration might be considered as a form of sequential grafting, as inflow from one conduit is derived to at least two coronary beds.
| 5. Follow-up studies |
|---|
|
|
|---|
The most essential data from mid-term and long-term studies radial artery patency not superior as expected" (the RA conduit patency equivalent or even worse than SVG patency) are presented in Table 1 . Data in Table 2 represent results radial artery patency superior as expected (the RA conduit patency better then SVG, not worse than right internal thoracic arteryRITA conduit patency). It should be noticed that RA conduit was used in sequential fashion by some authors (marked with a in Tables 1 and 2, greater number of reassessed anastomoses than used graftswhich means more distal anastomoses per graft, up to one-third of RA grafts in the report by Shah and colleagues [33]). Excellent long-term patency rate have been reported for saphenous vein grafts in such construction (85.4% at 10 years) [35].
|
|
In a retrospective study by Cameron and colleagues [38] (only 50 asymptomatic patients were included) RA graft patency was 88.7% versus 91.7% patency rate of SVG (average period to reangiogram 5.2 years). In addition, in a group of 21 patients who underwent restudy coronary angiography due to recurrent angina the same patency rate for RA conduit (88%) has been found. Therefore, authors concluded that the patency rate of 88% for RA conduit might be a true reflection of patency rate for entire cohort. Unfortunately, authors have overlooked the fact that average period to restudy angiogram was only 2.7 years in symptomatic patients (thus, in all appearance, additional graft failure can be expected in next 2.5 years).
Unfortunately, propensity score [39] statistical technique (used to overcome most of the bias sources of nonmatched studies) was employed only in the manuscript of Zacharias and colleagues [40]. Although cumulative 0- to 6-year survival was better for patients with radial artery conduit (vs patients with SVG as second conduit), the angiographic RA conduits patency (1.8 ± 1.4 years to reangiogram) were 68.2%, statistically not better (p = 0.11) than SVG patency rate (63.3%). Low patency rates may stem in part from the fact that the patency data were derived from symptomatic patients exclusively, who reflected a worst-case scenario. Due to possibility of silent graft occlusion, the assumption that symptom-free patients are likely to have more patent grafts does not justify extrapolation of better patency rates to entire cohort. This problem has been discussed only in one manuscript so far. The purpose of that study (Buxton and colleagues [41]) was to compare protocol-directed angiographic patency data from a randomized clinical trial (Radial Artery Patency and Clinical Outcome (RAPCO) study) [42] with symptom-directed angiography patency data in patients who were not included in the randomized controlled trial. Five-year estimates for graft patency were made using survival analyses accounting for interval censoring. The odds ratio for graft failure for nontrial (symptom-directed) patient grafts was 2.6 (95% confidence interval, 1.64.3; p < 0.001). Five-year estimates for individual conduits were 99% versus 92%, p = 0.002 for LITA, 86% versus 56%, p = 0.003 for saphenous vein, but not significant, 86% versus 87%, p = 0.8 for RITA and 87% versus 86%, p = 0.6 for radial artery.
In the article by Shah and colleagues [33] (average period to reangiogram, again in symptomatic patients, 3.0 ± 2.0 years), the RA conduit did not achieve better patency rate over SVG (90.6% vs 88.9%); thus they stated, When additional grafts are required, there is no evidence to suggest that either the radial artery or saphenous vein is superior.
The RAPCO study (prospective, randomized, single-center trial) compared the RA conduit angiographic patency with the free right internal thoracic artery (FRITA) patency in the first group of patients, and with SVG in the second group [42]. The trial conduit was grafted to the largest available coronary artery (diameter at least 1.5 mm, proximal stenosis of at least 70%) other than the LAD. Protocol-directed angiography at 5 years did not support the hypothesis that the RA has superior patency (RA vs FRITA97.4% vs 100%: RA vs SVG87.5% vs 95.5%) to or is associated with fewer clinical events than free right internal thoracic artery or saphenous vein grafts. It has to be underlined that SVG patency appears superior to that of recorded previously, suggesting that the choice of a large target artery and factors used to protect grafts (statin agents, vasodilators and 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors), may have improved the outcomes of both, the control and study grafts.
Possati and colleagues have reported excellent long-term patency rate of radial artery conduit in two prospective trials. In the first one [43], the RA graft patency (
5 years to reangiogram) was better than SVG patency (88.6% vs 74.0%, respectively, p
= 0.4). In the second one [9], the RA conduit patency rate (
9 years to reangiogram) was excellent, significantly better than SVG patency (90.5% vs 67.1%, respectively, p
< 0.0001), and not worse compared to LITA conduit patency (90.5% vs 96.3%, respectively, p
= 0.23).
The well-designed prospective, randomized trial (Radial Artery Patency Study (RAPS)) comparing the patency of saphenous vein grafts with that of RA grafts has recently been reported by Desai and colleagues [31]. The radial artery was randomly assigned to bypass the major vessel in either the right coronary territory or the circumflex territory, with the saphenous vein graft used for the opposing territory (control). The target coronary vessels had to be at least 1.5 mm in diameter, with proximal lesions causing narrowing of at least 70% of the diameter. The primary end point was graft occlusion, determined by angiography 812 months postoperatively. The angiographic study performed 1 year after surgery (average period to reangiogram 0.9 ± 0.4 years) in this trial demonstrated an occlusion rate of 8.2% for RA conduits and 13.6% for saphenous vein grafts (p
= 0.009). However, 7% of RA grafts (vs only 0.9% of SVG) had diffuse angiographic narrowing (the string sign), resulting in a total bad graft rate of 15.2% for RA conduit (vs 14.5% for SVG). Furthermore, perfect graft patency, defined as grafts with thrombolysis in myocardial infarction (TIMI) flow grade 3, was similar (87.7% for RA graft vs 85.7% for SVG, p
= 0.37). Multivariable analysis revealed that presence of radial artery string sign was closely related to the perioperative use of
-adrenergic agonists and target vessel stenosis less than 90% [44].
| 6. The future of the radial artery conduit |
|---|
|
|
|---|
The most important data are still to come. Prospective, randomized angiographic studies and/or meta-analyses between 10 and 15 years after CABG surgery should determine the position of the radial artery conduit in the future of CABG surgery.
| References |
|---|
|
|
|---|
lu D, Ömero
lu S, Göksedef D,
zgi A, Kirali K,
pek G, Yakut C. Does radial artery harvesting cause any changes in the forearm circulation during the postoperative period? An angiographic studyoriginal investigation. Anadolu Kardiyol Derg 2004;4:149-152.[Medline]This article has been cited by other articles:
![]() |
S. Bleiziffer, I. Hettich, B. Eisenhauer, D. Ruzicka, M. Wottke, J. Hausleiter, S. Martinoff, M. Morgenstern, and R. Lange Patency rates of endoscopically harvested radial arteries one year after coronary artery bypass grafting J. Thorac. Cardiovasc. Surg., September 1, 2007; 134(3): 649 - 656. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. B. Savage, J. D. Grab, S. M. O'Brien, A. Ali, E. J. Okum, R. A. Perez-Tamayo, D. S. Eiferman, E. D. Peterson, F. H. Edwards, and R. S.D. Higgins Use of Both Internal Thoracic Arteries in Diabetic Patients Increases Deep Sternal Wound Infection Ann. Thorac. Surg., March 1, 2007; 83(3): 1002 - 1006. [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 |