Total arterial revascularization with an internal thoracic artery and radial artery T graft

Total Arterial Revascularization With an Internal Thoracic Artery and Radial Artery T Graft Thoralf M. Sundt III, MD, Hendrick B. Barner, MD, Cynthia J. Camillo, RN, and William A. Gay, Jr, MD Division of Cardiothoracic Surgery, Department of Surgery, Washington University School of Medicine, St. Louis, Missouri

Background. Proximal anastomosis of the radial artery to the side of the internal thoracic artery (ITA) permits complete arterial revascularization in most patients, with the aim of improving long-term results of coronary artery bypass through greater long-term graft patency. The short-term results, however, have yet to be defined. We therefore reviewed our early experience with this grafting strategy. Methods. Between October 1, 1993, and September 1, 1998, 649 patients aged 30 to 85 years (mean, 60 ⴞ 10 years) had primary coronary artery bypass using an ITA and radial artery in a T-graft configuration. Left ventricular function was severely depressed (ejection fraction <35%) in 12%, and left main stenosis was present in 14%. Results. A total of 937 distal anastomoses were performed with the left ITA (1.4 per patient) and 1,452 with

the radial artery (2.2 per patient). There was one perioperative death (0.2%). There were 32 (5%) q-wave myocardial infarctions, and 14 patients (2%) had transient low output syndrome. There was one episode of hypoperfusion corrected by lengthening the left ITA. Angiography for clinical indications in 27 patients 1 to 35 months postoperatively (mean, 9.5 ⴞ 8.3 months) demonstrated a distal anastomotic patency of 100% for ITA and 82% for radial artery grafts. Conclusions. Complete arterial revascularization can be achieved with an ITA and radial artery T-graft with low operative risk and acceptable early patency. These results support the continued investigation of this grafting strategy. (Ann Thorac Surg 1999;68:399 – 405) © 1999 by The Society of Thoracic Surgeons

D

proximal ITA. Although the aim of this procedure is to improve long-term results, the short-term risks must be defined. We reviewed our entire first 5 years’ experience with this grafting strategy.

espite the excellent results achievable with coronary artery bypass (CAB) operations using a single internal thoracic artery (ITA) graft to the left anterior descending coronary artery [1, 2] and additional saphenous vein grafts, late graft failure and its sequelae remain significant limitations. Attempts to prolong event-free survival after CAB by using additional arterial conduits have been based on the observed superior late patency of the ITA as a bypass conduit. The benefit of additional arterial conduits has been supported by some [3, 4] and disputed by others [5, 6]. Recently, Lytle and associates [7] confirmed the impact of bilateral ITA grafts on the hard end points of reintervention and death. Critical to such discussions is the target area to which the additional arterial grafts are placed and the number of territories revascularized with nonarterial grafts. The optimal revascularization strategy would provide durable bypasses to all target vessels. Complete arterial revascularization can be achieved in most patients by proximal anastomosis of the radial artery (RA) to the side of the ITA in a T-graft configuration. This procedure is technically complex, however, and makes the entire revascularization dependent on the Presented at the Forty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 12–14, 1998. Address reprint requests to Dr Sundt, Division of Cardiothoracic Surgery, Department of Surgery, Washington University School of Medicine, Suite 3106 Queeny Tower, One Barnes Hospital Plaza, St. Louis, MO 63110; e-mail: [email protected].

© 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc

Patients and Methods The study group for this analysis includes all patients who had primary CAB with an ITA and a RA in a T-graft configuration from the beginning of our experience with this grafting strategy in October 1993 to September 1, 1998. These results therefore encompass the learning curve for all surgeons involved. Demographic information, operative characteristics, and surgical (30-day) outcomes were collected prospectively.

Patient Selection Initially, this grafting strategy was applied in younger patients and in those for whom conduit was limited in availability. With experience, however, T-grafts have become our grafting strategy of choice for patients younger than 75 years. In the senior author’s practice, 116 of 153 (76%) of patients who had isolated primary CAB in the last 12 months of the study received an ITA and RA T-graft. No attempt is currently made to reserve this procedure for low-risk patients. As shown in Table 1, the incidence of female sex, diabetes mellitus, hypertension and hypercholesterolemia, left main coronary artery stenosis, and severely depressed left ventricular 0003-4975/99/$20.00 PII S0003-4975(99)00563-9

400

SUNDT ET AL RADIAL ARTERY T GRAFTS

Ann Thorac Surg 1999;68:399 – 405

Table 1. Preoperative Patient Characteristics

Table 3. Distal Anastomoses

Demographic Characteristic

No. (%)

Age (y) Mean Range Female sex Hypertension Diabetes mellitus Oral medication Insulin dependent Tobacco abuse Active Inactive Chronic obstructive pulmonary disease Chronic renal insufficiencya Peripheral vascular disease Previous stroke Previous myocardial infarction ⬍1 month ⬎1 month Previous percutaneous transluminal coronary angioplasty Left main disease Ejection fraction ⬍35% Canadian Cardiovascular Society angina classc 0 1 2 3 4 New York Heart Association classd I II III IV

60 ⫾ 10 30 – 85 180 (28) 322 (50) 81 (13) 89 (14)

LAD Diagonal Ramus intermedius Circumflex branch RPLV PDA Acute marginal RCA Total

600 293 24 20 0 0 0 0 937

3 2 1 1 1 4 7 0 19

RA

SV GEA IEA

27 1 85 2 76 0 667 4 88 1 474 8 5 0 30 4 1,452 20

0 1 0 2 0 3 0 2 8

0 0 0 1 0 0 0 0 1

GEA ⫽ gastroepiploic artery; IEA ⫽ inferior epigastric artery; ITA ⫽ internal thoracic artery; LAD ⫽ left anterior descending artery; PDA ⫽ posterior descending artery; RA ⫽ radial artery; RCA ⫽ right coronary artery; RPLV ⫽ posterior left ventricular branch of the right coronary artery; SV ⫽ saphenous vein.

155 (24) 164 (25) 90 (14)

(Table 2), although most patients in this series had isolated CAB.

91 (14) 67 (12)b 73 (12) 28 (5) 28 (5) 322 (51) 170 (27) 30 (5) 368 (58) 144 (23) 92 (15)

b Chronic renal insufficiency defined as serum creatinine ⬎ 2.0. Data c for ejection fraction available for 572 patients. Data available for 621 d patients. Data for 634 patients.

function (ejection fraction ⱕ 35%) are similar to those in other series of primary CAB [2, 4]. This strategy has been used occasionally in association with other procedures Table 2. Operative Characteristics

Operative priority Emergent Urgent Elective Associated procedures Carotid endarterectomy Aortic valve replacement Mitral valve replacement Aortic and mitral valve replacement Cardiopulmonary bypass time (min) Aortic occlusion time (min)

Left ITA Right ITA

195 (30) 214 (33) 34 (5) 17 (3) 64 (10) 58 (9)

a

Variable

Conduit Vessel Grafted

Value 5 (5%) 44 (7%) 600 (92%) 9 2 1 2 118 ⫾ 26 (range, 53–244) 97 ⫾ 23 (range, 41–186)

Surgical Technique Surgical procedures were carried out as previously described [8] using cardiopulmonary bypass and cardioplegic cardiac arrest in all cases. Before cannulation, end-toside anastomosis of the RA to the ITA was done at the site where the ITA enters the pericardial space adjacent to the left atrial appendage. Radial artery grafts were filled with a solution of 60 mg papaverine diluted in 30 mL of heparinized blood after construction of the T graft and permitted to dilate under arterial pressure while preparations were made for cardiopulmonary bypass. All sideto-side sequential anastomoses were done in a longitudinal (parallel) manner. As shown in Table 3, our preference has been to place the ITA to the LAD, although the ITA was occasionally anastomosed to another target while the RA was used to graft the distal LAD and then taken over the acute margin to the right coronary system. Additional arterial or venous grafts were used occasionally to target vessels not grafted with the ITA or RA T graft, particularly early in our experience. More recently, however, additional conduits have been used only rarely. In only two cases in this series were saphenous vein grafts placed to target vessels in addition to a T graft as supplementary or back-up grafts. One was placed to an obtuse marginal branch of the circumflex artery early in our experience when the surgeon was unsure about the potential adequacy of flow through the T graft. A second saphenous vein was placed to the left anterior descending artery distal to an ITA anastomosis because of concern regarding local dissection of the distal ITA. This dissection appeared related to local trauma and did not originate at the site of the RA anastomosis. There was one instance of hypoperfusion recognized immediately after discontinuation of cardiopulmonary bypass. The ITA pedicle was noted to be tight and was lengthened by proximal mobilization. Bypass was again discontinued at

Ann Thorac Surg 1999;68:399 – 405

SUNDT ET AL RADIAL ARTERY T GRAFTS

Table 4. Thirty-Day Mortality and Morbidity Rates Event

No. (%)

Death Myocardial infarctiona Low output syndromeb Renal dialysis Stroke Respiratory failurec Ventricular arrhythmia Atrial arrhythmia Sternal infection Reoperation for bleeding

1 (0.2) 32 (5) 14 (2) 1 (0.2) 13 (2) 25 (4) 17 (3) 100 (16) 4 (0.6) 9 (1)

a

b Myocardial infarction defined by a new q wave. Low output is defined as cardiac index less than 2.0 or intraaortic balloon support. c Respiratory failure defined as intubation longer than 24 hours or reintubation.

which time the pedicle was slack and there were no sequelae. No additional or supplementary grafts were placed after the initial cross-clamp episode. Our technique of myocardial protection has evolved and now entails the use of tepid antegrade blood cardioplegia in almost all cases. Retrograde delivery is used selectively, in less than 5% of cases. The RA is harvested routinely by a surgical assistant simultaneously with preparation of the ITA. For expediency we prefer to use the left RA regardless of hand dominance. Recently we have used ultrasonic dissection for this purpose. Preoperative assessment of the palmar arch is done routinely with an Allen test alone. Early in our experience we insisted on ulnar filling of the entire hand in less than 5 seconds. As we have gained confidence and experience we have accepted delays up to 10 seconds without evidence of postoperative compromise. Digital plethysmography is done in questionable cases. Calcium-channel blocking agents are used intraoperatively and until the first postoperative day followed by oral therapy for 6 weeks (WAG, TMS) or not at all (HBB).

Follow-up Early (30-day) follow-up information is available for all 649 patients. Follow-up angiography was not performed

401

routinely; however, 27 patients had subsequent cardiac catheterization for clinical indications 1 to 35 months (mean, 9.5 ⫾ 8.3 months) postoperatively. Data are presented as the mean ⫾ standard deviation or as percentages.

Results Complete arterial revascularization was achieved with a single ITA and single RA in 606 patients (93%). Complete arterial grafting is likely possible in a higher percentage of cases; however, this series includes our early experience as we were becoming comfortable with this grafting strategy and were more likely to use additional grafts. A total of 937 distal anastamoses were done with the left ITA (1.4 per patient) and 1,452 with the RA (2.2 per patient) as shown in Table 3. Cardiopulmonary bypass times and aortic occlusion times were prolonged, reflecting the technical complexity of the procedure (Table 2). Mortality and morbidity rates are shown in Table 4. There was one early (30-day) death, for an operative mortality rate of 0.2%. The incidence of q-wave myocardial infarction and low output syndrome were also gratifyingly low. Fourteen patients (2%) had low output syndromes requiring significant inotropic support or an intraaortic balloon pump. Hypoperfusion was recognized in only 1 patient and resolved with mobilization of the ITA. The incidence of deep sternal wound infection was low, and there were no ischemic hand complications. Twenty-seven patients had angiographic follow-up studies for clinical indications or because of noninvasive studies suggestive of ischemia 9.5 ⫾ 8.3 months postoperatively. All studies were reviewed by the operating surgeon. As shown in Table 5, the patency of ITA anastomoses did not appear to be compromised by construction of the T graft. One T anastomosis was occluded, but all distal anastomoses of the radial artery were open. Those distal anatomoses were classified as patent. Three additional patients had completely occluded RA grafts, whereas 7 had a patent RA to the first anastomosis, most often to the circumflex artery, and occlusion of a more distal anastomosis, most often a branch of the right

Table 5. Results of Postoperative Catheterizations Internal Thoracic Artery Target LAD Diagonal RI Circumflex RPLV PDA RCA Total grafts

Radial Artery

Grafts Studied

Grafts Patent

Grafts Studied

Grafts patent

Total

26 17 1 0 0 0 0 44

26 17 1 0 0 0 0 44 (100%)

0 0 2 31 4 22 3 62

0 0 2 27a 3 14b 1 51

100% 100% 100% 87% 75% 64% 33% 82%

a

One radial artery graft to the circumflex coronary system was patent but showed a string sign, for a perfect patency rate of 83%. grafts to the posterior descending coronary artery were patent but showed string signs, for a perfect patency rate of 56%. LAD ⫽ left anterior descending artery; PDA ⫽ posterior descending artery; RPLV ⫽ posterior left ventricular branch of the right coronary artery.

RCA ⫽ right coronary artery;

b

Two radial artery

RI ⫽ ramus intermedius;

402

SUNDT ET AL RADIAL ARTERY T GRAFTS

coronary artery. Graft patency was best to the obtuse marginal branches of the circumflex artery (87%) and decreased to the progressively more distal target vessels. Angiographic results and noninvasive studies were frequently discordant.

Comment This study demonstrated that complete arterial revascularization can be achieved in most patients using a single ITA and single RA. The low perioperative risk observed in our series, albeit in a low-risk population, suggests that this strategy is safe despite dependence of the entire revascularization on the proximal ITA. Although others have cautioned against the potentially catastrophic consequences of acute hypoperfusion resulting from inadequate ITA flow [9, 10], hypoperfusion was not a clinically evident problem in our experience. We believe that hypoperfusion is more likely related to technical errors, such as conduit injury or angulation, than to inadequate flow reserve of the ITA. We [11] and others [12] previously reported the use of bilateral ITAs for revascularization of the left coronary system in the presence of left main disease without incremental risk. On the basis of these data we applied the T graft strategy regardless of the presence of hemodynamically significant left main stenosis, an anatomic feature present in 14% of patients in this series. We have been gratified by the results. More subtle hypoperfusion caused by early limitation of flow reserve could have been responsible for the results of noninvasive studies suggesting ischemia despite patent grafts at angiography. Recent evidence indicates that flow reserve of the ITA might increase significantly with time [13], and we would expect such subtle ischemia to improve. We favor the use of the RA based proximally on the ITA both to obtain adequate length to permit revascularization with two conduits alone and to avoid anastomosis of the arterial conduit directly on the aorta. Patency of anastomoses of the ITA to the aorta have been reported to be only 80% to 90% [14, 15] whereas that for the inferior epigastric artery or gastroepiploic artery is 75% to 85% [16 –18]. Patency is likely lower if the aorta is thickened unless a vein hood or pericardial patch is used as an intermediary. Furthermore, we share the concern of Calafiore and colleagues [19] that placement of what is normally a third or fourth order artery on the aorta could impose abnormal and potentially harmful sheer stress on the conduit. Using the left ITA for inflow, Tector and associates [20] achieved a 91% patency rate with a T graft configuration, and Calafiore and colleagues [19] a 93% patency for anastomosis of the RA to the left ITA. We chose to explore the use of the RA for construction of a T graft because it has favorable handling characteristics and can be harvested simultaneously with the ITA. Since the revival of the RA as a bypass conduit by Acar and associates [21], it has been embraced with remarkable rapidity by many surgeons, likely for these same reasons. Furthermore, several large series of CAB with

Ann Thorac Surg 1999;68:399 – 405

this conduit in more conventional configurations have already been reported with good early results [19, 21–23]. The patency rate we observed for RA grafts to the branches of the right coronary artery was disappointing. As the postoperative angiograms in this study were prompted by clinical indications rather than routine control studies, this sampling might be biased negatively with regard to patency. Although other authors have observed lower patency rates of arterial grafts to targets other than the left anterior descending artery [24, 25], this observation suggests that we need to reassess our strategy with respect to right-sided vessels. Given the T graft configuration, the graft segments to progressively more right-sided targets are the most distal on the RA and are therefore most likely to be under tension. The observation of diminishing patency with more distal anastomoses of the RA supports this hypothesis. During the conduct of this series, we preferred to use longitudinal (parallel) technique for all side-to-side anastomoses. We recognize, however, that conduit length might be better conserved by the use of crossing (perpendicular) anastomoses, and we have changed our philosophy accordingly. Alternatively, a greater willingness to use a third conduit might be indicated. Three RA segments demonstrated string signs at postoperative angiography. Two of these grafts were to target vessels with less-than-critical stenoses. String signs therefore might represent an autoregulatory phenomenon in the presence of competitive flow. The persistent patency despite this competitive flow is, we believe, encouraging.

Comparison With Other Studies Sauvage and associates [26] first reported the use of composite arterial grafts in a T graft configuration in 1986; however, it has been only recently that large series have been reported. Tector and colleagues [20] reported complete arterial revascularization in 287 patients using both ITAs in a T graft configuration with an operative mortality rate of 1.7%. Early graft patency among the 26 patients who had postoperative angiography was somewhat higher in that series than in our study, with 91% of free right ITA grafts. In contrast with our experience, 4 patients of Tector and colleagues were returned to the operating room for additional saphenous vein grafts in the setting of apparent hypoperfusion, and 4 other patients, excluded from analysis in that series, had additional saphenous grafts placed before weaning from cardiopulmonary bypass because of regional wall motion abnormalities and suspected hypoperfusion. Subsequently, Barra and associates [27] reported on 80 highly selected patients, receiving a left ITA and right ITA T graft with a somewhat higher operative mortality rate of 3.75%. Sixty patients (84.2%) had control angiography at 12 to 24 months postoperatively, demonstrating 93.4% patency of the free right ITA. No episodes of hypoperfusion were observed. Calafiore and colleagues [19] introduced the use of the RA as a free Y or T graft from the side of the pedicled left ITA. There were no operative deaths in their series of

Ann Thorac Surg 1999;68:399 – 405

composite arterial grafts, including 103 radial artery T grafts and 124 T grafts constructed with the inferior epigastric artery. The cumulative patency rate was 93.1% for RA grafts at a mean follow-up interval less than 1 year. In the series as a whole, hypoperfusion was seen in 2.5%. More recently, Weinschelbaum and associates [28] studied 164 patients, all of whom had ITA-RA T grafts. Among the 46 patients who had early control angiography, all arterial conduits were patent. The operative mortality rate was also low, at 1.8%. The results of our larger study are consonant with these reports. Our operative mortality rate compares favorably with the rates of those studies. In contrast to the experience of Tector and associates with two ITAs, we did not observe hypoperfusion even though we applied this technique more broadly and aggressively to older patients, those with left main stenosis, and to those with poor left ventricular function. Our observed patency rate was somewhat lower than the rates in those studies, perhaps reflecting selection bias in patients who had undergoing angiography in our series or the fact that our series included the learning curves of all surgeons involved. It could also reflect, in part, our aggressive approach to complete revascularization including a willingness to place grafts to small target vessels. Additionally, in contrast to both Weinschelbaum and associates and Calafiore and colleagues, calcium channel blockers were not administered postoperatively in approximately 90% of our patients at the discretion of the surgeon (HBB). There are no clear differences in patency rates among surgeons involved.

Rationale for Complete Arterial Revascularization It is well established that use of a single ITA to the left anterior descending artery reduces the risk of recurrent angina [2], cardiac reoperation [29], late myocardial infarction, and death [1]. This is presumably attributable to the superior late patency of the ITA. Based on this observation, Fiore and associates [3] and Pick and colleagues [4] attempted to show improved event-free survival with bilateral ITA grafting. A statistically significant reduction in subsequent myocardial infarction and recurrence of angina pectoris was demonstrated by Fiore and associates [3] in their retrospective analysis of 100 patients who had bilateral ITA grafting as compared with 100 case-matched controls. A survival benefit was also suggested, as was improved freedom from coronary reintervention. These findings have been confirmed by Pick and colleagues [4] in a similar case-controlled study of patients who had single (n ⫽ 161) or bilateral (n ⫽ 160) ITA grafting. Bilateral ITA grafting was an independent predictor of lower rates of angina recurrence and late myocardial infarction. More recently, Lytle and colleagues [7] reported reduced risks of reintervention (reoperation or angioplasty) and enhanced survival for all age groups except for those less than 50 years of age when two ITAs were used instead of one. Those findings have been challenged. Using the Katholieke Universiteit Leuven Coronary Surgery Database, Sergeant and coworkers [5] have been unable to demon-

SUNDT ET AL RADIAL ARTERY T GRAFTS

403

strate an influence of multiple arterial grafting on patient survival or on the return of angina pectoris and the occurrence of infarction [6]. The explanation for this disparity might be the frequent inappropriate choice of targets to which these additional arterial grafts were placed until 1987 and the limited number of patients with sufficiently long follow-up to permit demonstration of late benefit. A strategy that provides complete arterial revascularization neutralizes the issue of target selection and should provide the best possible long-term result.

Study Limitations The principle limitations of this study are the small number of angiographic controls available and the short duration of follow-up in most patients. Unfortunately, routine follow-up angiography is not practical in the current era of cost containment. As the rationale for using T grafts is predicated on long-term benefit, these data are insufficient to argue the superiority of this strategy over the more standard single ITA and additional saphenous vein grafts. These data do, however, indicate that despite the increased complexity of this approach and the theoretic potential for hypoperfusion, this strategy can be used without incurring an increased operative risk. Whether this safety is predominantly reflective of flow through the ITA or the RA grafts is unknown. We conclude from these data that complete arterial revascularization can be achieved in most patients with a single ITA and a single RA. The operative risk is low. Early graft patency, particularly to the circumflex system, was encouraging. Disappointing patency to more distal targets was likely related to technical factors that are being addressed. These findings support the continued exploration of this grafting strategy in pursuit of a more durable procedure for the treatment of coronary artery disease. Larger, ideally randomized, prospective trials of complete arterial revascularization versus CAB with a single ITA and additional saphenous vein grafts are warranted.

References 1. Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal-mammary artery graft on 10-years survival and other cardiac events. N Engl J Med 1986;314:1– 6. 2. Acinapura AJ, Rose DM, Jacobwitz IJ, et al. Internal mammary artery bypass grafting: influence on recurrent angina and survival in 2100 patients. Ann Thorac Surg 1989;48: 186–91. 3. Fiore AC, Naunheim KS, Dean P, et al. Results of internal thoracic artery grafting over 15 years: single versus double grafts. Ann Thorac Surg 1990;49:202–9. 4. Pick AW, Orszulak TA, Anderson BJ, Schaff HV. Single versus bilateral internal mammary artery grafts: 10 year outcome analysis. Ann Thorac Surg 1997;64:599 – 605. 5. Sergeant P, Blackstone E, Meyns B. Does extensive arterial revascularization decrease the early and long term risk of myocardial infarction after coronary artery bypass grafting? Ann Thorac Surg 1998;66:1–11. 6. Sergeant P, Blackstone E, Meyns B. Is return of angina after coronary artery bypass grafting immutable, can it be delayed, and is it important? J Thorac Cardiovasc Surg 1998; 116:440–53.

404

SUNDT ET AL RADIAL ARTERY T GRAFTS

7. Lytle BW, Arnold JH, Loop FD, et al. Two internal thoracic artery grafts are better than one. J Thorac Cardiovasc Surg 1999;117:855–72. 8. Barner HB, Johnson SH. The radial artery as a T-graft for coronary revascularization. Operative Tech Cardiac Thorac Surg 1996;1:117–36. 9. Jones EL, Lattouf OM, Weintraub WS. Catastrophic consequences of internal mammary artery hypoperfusion. J Thorac Cardiovasc Surg 1989;98:902–7. 10. Vajatai P, Ravichandran PS, Fessler CL, et al. Inadequate internal mammary artery graft as a cause of postoperative ischemia: incidence, diagnosis and management. Eur J Cardiothorac Surg 1992;6:603– 8. 11. Barner HB, Naunheim KS, Willman VL, Fiore AC. Revascularization with bilateral internal thoracic artery grafts in patients with left main coronary stenosis. Eur J Cardiothorac Surg 1992;6:66–71. 12. Paolini G, Zuccari M, Di Credico G, et al. Myocardial revascularization with bilateral internal thoracic artery in patients with left main disease: an incremental risk? Eur J Cardiothorac Surg 1994;8:576–9. 13. Akasaka T, Yoshikawa J, Yoshida K, et al. Flow capacity of internal mammary artery grafts: early restriction and later improvement assessed by doppler guide wire. Comparison with saphenous vein grafts. J Am Coll Cardiol 1995;25:640–7. 14. Verhelst R, Etienne PY, El Khoury G, Noirhomme P, Rubay J, Dion R. Free internal mammary artery graft in myocardial revascularization. Cardiovasc Surg 1996;4:212– 6. 15. Tatoulis J, Buxton B, Fuller JA. Results of 1,454 free right internal thoracic artery-to-coronary artery grafts. Ann Thorac Surg 1997;64:1263–9. 16. Buche M, Schroeder E, Gurne O, et al. Coronary artery bypass grafting with the inferior epigastric artery: midterm clinical and angiographic results. J Thorac Cardiovasc Surg 1995;109:553– 60. 17. Manapat AE, McCarthy PM, Lytle BW, et al. Gastroepiploic and inferior epigastric arteries for coronary artery bypass: early results and evolving applications. Circulation 1994;90: II-144 –7. 18. Suma H, Amano A, Horii T, Kigawa I, Fukuda S, Wanibuchi

Ann Thorac Surg 1999;68:399 – 405

19.

20. 21. 22.

23. 24. 25. 26.

27.

28.

29.

Y. Gastroepiploic artery graft in 400 patients. Eur J Cardiothorac Surg 1996;10:6–11. Calafiore AM, Di Giammarco G, Teodori G, et al. Radial artery and inferior epigastric artery in composite grafts: improved midterm angiographic results. Ann Thorac Surg 1995;60:517–24. Tector AJ, Amundsen S, Schmal TM, et al. Total revascularization with T-grafts. Ann Thorac Surg 1994;57:33–9. Acar C, Jebara BA, Portoghese N, et al. Revival of the radial artery for coronary artery bypass grafting. Ann Thorac Surg 1992;54:652– 60. Brodman RF, Frame R, Camacho M, Hu E, Chen A, Hollinger I. Routine use of unilateral and bilateral radial arteries for coronary artery bypass graft surgery. J Am Coll Cardiol 1996; 28:959– 63. Tatoulis J, Buxton BF, Fuller JA. Bilateral radial artery grafts in coronary reconstruction: technique and early results in 261 patients. Ann Thorac Surg 1998;66:714–20. Huddleston CB, Stoney WS, Alford WC, et al. Internal mammary artery grafts: technical factors influencing patency. Ann Thorac Surg 1986;42:543–9. Chow MST, Sim E, Orszulak TA, Schaff HV. Patency of internal thoracic artery grafts: comparison of right versus left and importance of vessel grafted. Circulation 1994;90:II-129 –32. Sauvage MR, Wu H-D, Kowalsky PE, et al. Healing basis and surgical techniques for complete revascularization of the left ventricle using only the internal mammary arteries. Ann Thorac Surg 1986;42:449– 65. Barra JA, Bezon E, Mansourati J, Rukbi I, Mondine P, Youssef Y. Reimplantation of the right internal thoracic artery as a free graft into the left in situ internal thoracic artery (Y Procedure). One year angiographic results. J Thorac Cardiovasc Surg 1995;109:1042– 8. Weinschelbaum EE, Gabe ED, Macchia A, Smimmo R, Suarez LD. Total myocardial revascularization with arterial conduits: radial artery combined with internal thoracic arteries. J Thorac Cardiovasc Surg 1997;114:911– 6. Cosgrove DM, Loop FD, Lytle BW, et al. Predictors of reoperation after myocardial infarction. J Thorac Cardiovasc Surg 1986;92:811–21.

DISCUSSION DR WILLIAM A. BAUMGARTNER (Baltimore, MD): I would like to congratulate Dr Sundt and his associates for both an excellent study and a well-presented paper. The authors have shown in this study that the radial artery as a T graft off the internal thoracic artery in a large number of patients is associated with good functional results. Dr Sundt, you may have answered one of my first questions. I thought that although angiography was performed for clinical reasons when you might expect a higher occlusion rate, this early patency rate of 82% seems somewhat low and comparable to early graft patency. My question is do you think this is a function of the T graft or the radial artery itself? Also, have you done any noninvasive follow-up studies to assess adequacy of revascularization in your other patients? I noticed that the mean age of your patients is about 4 or 5 years younger than those series of isolated bypass operations. Do you have any current selection criteria for patients being considered for total arterial revascularization? Because you stated in your paper that there is a difference of opinion within your group, would you please comment as to the use of calcium antagonists in the postoperative period. The long-term efficacy of radial artery conduits still needs to be determined. Your group was a pioneer in the use of internal thoracic conduits when most surgeons thought that saphenous veins and internal thoracic arteries were comparable conduits. I

commend your group for proceeding with another arterial conduit in a somewhat similar atmosphere of questioning. I enjoyed your paper very much and I thank the Association for the privilege of commenting. Thank you. DR SUNDT: Thank you Dr Baumgartner. I think we have touched on the first issue, patency of grafts to the right coronary system. We are not entirely happy with those numbers and I think that only in time will we know what the true patency is going to be for the T graft. I think it can be a little complex determining your intergraft distances, and it might be that there is some tension on the graft segments to the right. We are therefore considering the use of perpendicular side-to-side anastomoses to conserve conduit and the use of additional conduits to the right coronary system. I would note also that some authors argue that there might be differences in graft patency related to the target vessel itself independent of the conduit. We have not embarked on routine postoperative noninvasive studies but some of our skeptical cardiologic colleagues have. We look forward to having an opportunity to review the noninvasive studies which they are obtaining for us. Thus far, however, the studies have been gratifying. A number of the postoperative angiograms were done because of positive results of stress tests, and the angiograms subsequently showed all grafts to be patent. In terms of the selection criteria, specifically age, that is a