Ann Thorac Surg 2004;78:471-475
© 2004 The Society of Thoracic Surgeons
Original article: cardiovascular
Comparison of bilateral thoracic artery grafting with percutaneous coronary interventions in diabetic patients
Chaim Locker, MDa,
Rephael Mohr, MDa*,
Oren Lev-Ran, MDa,
Gideon Uretzky, MDa,
Aharon Frimerman, MDa,
Yael Shaham, MDa,
Itzhak Shapira, MDa
a Department of Cardiothoracic Surgery, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel
Accepted for publication February 6, 2004.
* Address reprint requests to Dr Mohr, Dept of Cardiothoracic Surgery, Tel Aviv Sourasky Medical Center, 6 Weizman St, Tel Aviv 64239, Israel
e-mail: shapiraiz{at}tasmc.health.gov.il
 |
Abstract
|
|---|
BACKGROUND: This study compares the outcome of percutaneous coronary interventions (PCI) with bilateral internal thoracic grafting (BITA) in diabetic patients.
METHODS: From May 1996 to December 1999, 802 consecutive diabetic patients underwent myocardial revascularization: 363 by PCI and 439 by BITA. The two groups were similar; however, left main disease (28% versus 3.3%), ejection fraction less than 0.35 (14.5% versus 5.5%), and chronic obstructive lung disease (8.4% versus 3%) were more prevalent in the BITA group, and prior percutaneous transluminal coronary angioplasty, in the PCI group (16.8% versus 10.5%).
RESULTS: The number of coronary vessels treated per patient was higher in the BITA group (3.4 versus 1.2; p < 0.001). Thirty-day mortality was similar: 3.4% in the BITA group and 2.8% in the PCI group. Late follow-up (3 to 6.5 years) showed decreased return of angina (11% versus 64%; p < 0.001), fewer reinterventions (2.7% versus 55%; p < 0.001), and increased cardiovascular event-free survival (80% versus 30%; p < 0.001) in the BITA group. Six-year survival of BITA and PCI patients was 85.5% and 81.2%, respectively (not significant). However, survival of the subgroups of patients with left main or three-vessel coronary artery disease was significantly better with BITA (86% versus 76%; p = 0.003).
CONCLUSIONS: Despite higher risk profile of diabetic patients treated surgically by BITA, their late outcome is better than that of patients treated by PCI. The results of this study support referring diabetics with single-vessel or double-vessel disease to PCI and those with three-vessel and left main coronary artery disease to surgery.
|
Introduction
|
|---|
The left internal thoracic artery (ITA) is the conduit of choice for myocardial revascularization because of its superior graft patency when compared with saphenous vein grafts [1]. When ITA is connected to the left anterior descending coronary artery, this superior patency rate is associated with improved long-term survival and reduced occurrence of adverse cardiovascular events [2]. Long-term survival and freedom from myocardial infarction (MI), recurrence of angina, percutaneous reinterventions, and repeat operations may further be reduced by the use of bilateral ITA (BITA) [3, 4]. However, most surgeons are reluctant to use BITA grafting routinely because of increased reported risk of sternal dehiscence and deep sternal infection associated with harvesting of BITA, especially in diabetic or elderly patients [5]. This increased risk of sternal complications is probably related to sternal devascularization, and may be reduced by harvesting both ITAs using the skeletonizing technique [6]. In this technique, the ITA is separated from the chest wall without its accompanying veins and endothoracic fascia, and its collaterals are divided gently between silver clips without the use of electrocautery, thus preserving collateral sternal blood supply from the chest wall through the intercostal arteries.
In May 1996 we decided to adopt the technique of bilateral skeletonized ITA grafting as our preferred myocardial revascularization technique, and since then have used BITA routinely in 71% of all coronary artery bypass grafting (CABG) procedures, including diabetic and elderly patients. The relative number of diabetic patients referred to CABG during this period increased after the publications from the Bypass Angioplasty Revascularization Investigation (BARI) study [7, 8]. This study showed that angioplasty in diabetic patients with multivessel coronary disease is associated with increased rates of restenosis and reinterventions, and significantly lower 4- and 7-year survival, compared with CABG.
On the basis of the BARI report, a combined approach was tailored to diabetic patients in our institution: most multivessel diabetic patients were referred to surgery, and angioplasty was offered mainly to patients with single-vessel and double-vessel coronary artery disease.
Results of this combined approach including early results and midterm (up to 6.5 years) follow-up are described in this retrospective report.
|
Patients and methods
|
|---|
The study evaluates all diabetic patients who underwent percutaneous coronary intervention (PCI) or CABG with BITA between May 1996 and December 1999 in the Tel Aviv Sourasky Medical Center. Our definition for diabetes consisted of patients diagnosed as having diabetes mellitus type 2 and treated by oral hypoglycemics at least 6 months before their procedure. Patients with diabetes mellitus type 2 who were treated with insulin were also included and comprised 30 patients (8%) in the PCI group and 39 patients (9%) in the CABG group (Table 1). Patients treated with diet only were excluded.
Preoperative characteristics of both groups and details of operative procedures performed are listed in Table 1.
The total number of PCIs performed in the 363 PCI patients during the first procedure was 488, including 376 (77.5%) procedures with stents. Stent implantation was performed after balloon angioplasty dilatation. All patients received 100 mg of aspirin starting the first postprocedural day, platelet glycoprotein (Gp) IIb IIIa receptor blockers (abciximab or epifibratide) were used in 202 patients [9], and ticlopidine (250 mg) was used for 1 month after the procedure in 161 patients, specifically, PCI patients had been treated with ticlopidine until the beginning of 1998 when abciximab was introduced in our institution. We separated the patients according to medication and compared the results between the patients who were treated with abciximab to those who were not.
In the CABG group the ITAs were dissected as skeletonized arteries [10] before heparin administration to decrease the risk of damage and hematoma formation in the region of side branches during dissection. The operations were performed with cardiopulmonary bypass. Myocardial preservation involved intermittent infusion of warm cardioplegic solution (30° to 32°C) [6].
We used three different techniques for right coronary artery grafting. In 87 patients (20%) the right gastroepiploic artery was used as a third arterial conduit to bypass the posterior descending branch of the right coronary artery. Revascularization of the right system with saphenous vein graft was used in 80 (18%) of the patients. Composite T-graft with end-to-side anastomosis of free ITA (mostly the right ITA) connected to an in situ ITA (mostly the left ITA) was used in 335 patients (76%). The distal end of the free right ITA was used for posterior descending coronary artery grafting in 93 patients (21%). The remaining 179 patients had no bypass to the right system. Sequential grafting was used in 259 patients (59%). To decrease the risk of spasm of the arterial grafts, we treated all patients with high-dose intravenous infusion of isosorbide dinitrate (Isoket; 4 to 20 mg/h) during the first postoperative 24 to 48 hours [11]. Systolic blood pressure was maintained greater than 100 to 120 mm Hg. From the second postoperative day, patients who received a gastroepiploic artery were treated with calcium-channel blockers (diltiazem 90 to 180 mg/d orally) for at least 3 months.
Data are expressed as mean ± standard deviation or proportions as appropriate. The 
2 test and Fisher's exact test were used for discrete variables. The Cox proportional hazard model was used to evaluate risk factors for overall mortality. Hazard ratio and 95% confidence intervals are given. Postoperative survival is expressed by the Kaplan-Meier method, and survival was compared by the log rank test. All analyses were performed by SPSS 9.0 software (SPSS Inc, Chicago, IL).
|
Results
|
|---|
The two groups were similar regarding age, sex, and the prevalence of acute MI, peripheral vascular disease, and chronic renal failure (Table 1). However, the CABG group included more patients with chronic obstructive pulmonary disease (COPD; 37 versus 11, 8.4% versus 3%) and severe left ventricular dysfunction (ejection fraction < 0.35; 64 versus 20, 15% versus 5.5%). On the other hand, prior percutaneous transluminal coronary angioplasty was more prevalent in the PCI group (61 versus 46, 16.8% versus 10.5%).
Left main coronary artery disease and triple-vessel coronary artery disease were more common among CABG patients, and single-vessel or double-vessel disease was more prevalent in the PCI group. The number of vessels treated was higher in the CABG group (3.4 versus 1.2), and more patients in the CABG group underwent revascularization of the left anterior descending coronary artery territory (Table 1).
Thirty-day mortality was 3.4% in the CABG group and 2.8% in the PCI group (not significant). Sternal wound infection in the CABG group occurred in 3.2% of the patients.
The rate of periprocedural MI was significantly higher in the PCI group (4.1% versus 0.7%; p = 0.000). The rate of total major periprocedural complications in the two groups was similar (Table 2).
Postprocedural follow-up from 3 to 6.5 years was available in 762 (96%) of the patients.
The two groups had similar late and overall (early + late) mortality. Six- year survival (Kaplan-Meier) was slightly better in the CABG group (85.5% versus 81.2%). However, the difference in survival did not reach statistical significance (log-rank test; Table 2, Fig 1).
View larger version (11K):
[in this window]
[in a new window]
|
Fig 1. Survival curve (Kaplan-Meier) for diabetic patients undergoing percutaneous coronary interventions (PCI) compared with those undergoing coronary artery bypass grafting (CABG).
|
|
Symptomatic status of diabetic patients who underwent CABG was better. Only 11% of them experienced return of angina during the follow-up period, compared with 64% of the PCI patients. Only 12 of the CABG group (2.7%) had to undergo reintervention, compared with 200 (55%) of the PCI patients. One hundred five (30%) of the PCI group were free of angina or reintervention at the time of latest follow-up, compared with 353 (80%) of the CABG group. Reintervention in the PCI group included 16 (4.4%) CABG and 184 (50.4%) re-PCI.
In the CABG group 2 (0.5%) patients underwent repeat operations and 10 (2.2%) had to have postoperative PCI. In the PCI group, the 202 patients treated with abciximab had no significant difference in early, late, or overall mortality compared with the 161 patients who received ticlopidine. Moreover, there was no difference in the rate of return of angina or reinterventions.
To determine whether the similar overall mortality in the two groups is affected by the fact that the CABG group contains more patients with ejection fraction less than 0.35, three-vessel coronary artery disease, left main coronary artery disease, and COPD, and the PCI group had more patients with prior percutaneous transluminal coronary angioplasty, we performed a Cox regression analysis with patient's group (PCI or CABG) as a dependent variable. We first included the preoperative risk factors to be controlled (left main coronary artery disease, three-vessel coronary artery disease, COPD, ejection fraction less than 0.35, prior percutaneous transluminal coronary angioplasty, and emergency procedure) and then overall mortality. The regression model showed that even after controlling for the above risk factors, overall mortality was not statistically different between the two groups.
Overall mortality in the PCI group was related to the extent of coronary artery involvement: overall mortality of patients with single-vessel disease was 9.6% (14 patients) and that of patients with double-vessel disease was 15.9% (17 patients). Triple-vessel disease patients had 23.5% mortality (23 patients), and patients who underwent PCI to their left main had 25% (3 patients) overall mortality (Fig 2). It is important to note that only 2 patients in the three-vessel disease subgroup underwent PCI of all three vessels, 30 underwent PCI of two vessels, and 74 had PCI of only one vessel. Overall mortality of three-vessel disease patients who underwent CABG was significantly lower than that of PCI patients (13% versus 23.5%, p = 0.009). Six-year survival (Kaplan-Meier) of PCI patients with three-vessel disease was 76% compared with 86% in CABG patients (p = 0.008, log-rank test; Fig 3).
View larger version (11K):
[in this window]
[in a new window]
|
Fig 2. Survival of patients undergoing percutaneous coronary interventions and extent of three-vessel coronary artery disease (3VD) involvement (Kaplan-Meier). (1VD = single-vessel disease; 2VD = double-vessel disease.)
|
|
View larger version (9K):
[in this window]
[in a new window]
|
Fig 3. Comparison of survival in patients with three-vessel disease undergoing percutaneous coronary interventions (PCI) or coronary artery bypass grafting (CABG) (Kaplan-Meier).
|
|
Predictors of overall mortality in the CABG group were peripheral vascular disease, congestive heart failure, and age more than 70 years (Table 3). Predictors of overall mortality in the PCI group were age more than 70 years, COPD, emergency procedure, old MI, periprocedural use of stents, and the number of coronary vessels involved. Recurrence of angina had a protective effect, and early recurrence of angina was associated with better long-term survival (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 3. Overall (Early + Late) Mortality of Coronary Artery Bypass Grafting Patients: Results of a Cox Proportional Hazards Model
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Overall (Early + Late) Mortality of Percutaneous Coronary Intervention Patients: Results of a Cox Proportional Hazards Model
|
|
|
Comment
|
|---|
Randomized trials comparing multivessel PCI with CABG have demonstrated similar rates of death and MI, but higher rates of reinterventions and early return of angina in PCI patients [12, 13]. The comparable long-term survival was demonstrated in most patients with the exception of diabetics [1, 8, 14]. Restenosis rates after balloon angioplasty in diabetic patients can be as high as 63% [15]. Long-term survival benefit of CABG in diabetics in the prestent era was reported recently by the BARI investigators [8]. Similar findings were reported recently in the subgroup of 208 diabetic patients treated with stents from the Arterial Revascularization Therapy Study. Three-year survival of diabetic patients who underwent CABG was better than that of diabetics who underwent myocardial revascularization with stents [16].
The number of diabetic patients in our study is close to four times the number of diabetics in Arterial Revascularization Therapy Study, and 77% of the PCI patients were treated with stents.
Overall mortality in our study of CABG patients was similar to that of PCI. However, mortality of PCI patients with triple-vessel disease or left main coronary artery disease was higher than that of patients with triple-vessel disease and left main coronary artery disease who underwent CABG. This increased overall mortality may be related to incomplete revascularization. Moreover, use of stents and the extent of coronary artery disease were found to be independent predictors of late mortality in the PCI group.
Long-term survival and patency of vessels treated with stents may be improved by the use of platelet receptor Gp IIbIIIa blockers such as abciximab [9]. This drug was not used in the Arterial Revascularization Therapy Study. However, 55% of the diabetic PCI patients in our study were treated with platelet receptor blockers. Despite the aggressive treatment with antiplatelet medications used during our study period, the rates of reintervention and angina return were significantly higher in the PCI group.
Preliminary reports from studies of drug-eluting stents [17] suggest a significant decrease of neointimal growth and in-stent restenosis with drug-eluting stents, as compared with bare stents. Drug-eluting stents were also found to be effective in diabetic patients with small-vessel disease [17]. However, a longer follow-up and larger series of patients with diabetes mellitus are required to compare results with those of CABG.
Several recent studies recommend using stents more liberally in patients with increased risk to die or develop post-CABG complications such as cerebrovascular accident and sternal infection [18]. The results of our study do not support this notion. The rates of early major perioperative complications (MI, stroke, and sternal infection) in the surgical group were relatively low. The predictors of postoperative mortality in the CABG group were old age, peripheral vascular disease, and CHF. A predictor for sternal wound infection was COPD. Old age and COPD were also found to be predictors of mortality after PCI, and the mortality rate of peripheral vascular disease patients in both groups was similar. Therefore, PCI did not significantly decrease the risk of death or major complications after myocardial revascularization in those high-risk diabetic patients.
The mean follow-up time in our study was 4.5 years. Late results of the BARI study [8] showed that survival benefit after 7 years was better than that observed after 4 years [7]. The survival benefit in the BARI study was confined to patients who underwent revascularization of the left anterior descending artery with ITA. Use of BITA grafting may improve late outcome as compared with single ITA. However, a longer follow-up time is required.
In conclusion, despite higher risk profile of diabetic patients treated surgically by BITA, their late outcome is better than that of patients treated by PCI. The results of this study support our policy of referring diabetics with single-vessel or double-vessel disease to PCI and those with three-vessel and left main coronary artery disease to surgery.
|
References
|
|---|
- Barner H.B., Swartz M.I., Mudd J.G., Tyras D.H. Late patency of the internal mammary artery as a coronary artery bypass conduit. Ann Thorac Surg 1982;34:408-412.[Abstract]
- Loop D.F., Lytle B.W., Cosgrove D.M., et al. Influence of internal-mammary-artery-graft on 10 year survival and other cardiac events. N Engl J Med 1986;314:1-6.[Abstract]
- Lytle B.W., Blackstone E.H., Loop F.D., et al. Two internal thoracic artery grafts are better than one. J Thorac Cardiovasc Surg 1999;117:855-872.[Abstract/Free Full Text]
- Buxton B.F., Komeda M., Fuller J.A., Gordon I. Bilateral internal thoracic artery grafting may improve outcome of coronary artery surgery: risk-adjusted survival. Circulation 1998;98(Suppl 2):II-1-6.
- Kouchoukos N.T., Wareing T.H., Murphy S.F., Pelate C., Marshall W.G.J. Risks of bilateral internal mammary artery bypass grafting. Ann Thorac Surg 1990;49:210-217.[Abstract]
- Sofer D., Gurevitch J., Shapira I., et al. Sternal wound infections in patients after coronary artery bypass grafting using bilateral skeletonized internal mammary arteries. Ann Surg 1999;229:585-590.[Medline]
- BARI Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996;335:217-225.[Abstract/Free Full Text]
- BARI Investigators. Seven-year outcome in the Bypass Angioplasty Revascularization Investigation (BARI) by treatment and diabetic status. J Am Coll Cardiol 2000;35:1122-1129.[Abstract/Free Full Text]
- Bhatt D.L., Marso S.P., Lincoff M., Wolski K.E., Ellis S.G., Topol E.J. Abciximab reduces mortality in diabetics following percutaneous coronary intervention. J Am Coll Cardiol 2000;4:922-928.
- Gurevitch J., Kramer A., Locker C., et al. Technical aspects of double-skeletonized internal mammary artery grafting. Ann Thorac Surg 2000;69:841-846.[Abstract/Free Full Text]
- Gurevitch J., Miller H.I., Shapira I., et al. High-dose isosorbide dinitrate for myocardial revascularization with composite arterial grafts. Ann Thorac Surg 1997;63:382-387.[Abstract/Free Full Text]
- Sim I., Gupta M., McDonald K., Bourassa M., Hlatky M.A. A meta-analysis of randomized trials comparing coronary artery bypass grafting with percutaneous transluminal coronary angioplasty in multi-vessel coronary artery disease. Am J Cardiol 1995;76:1025-1029.[Medline]
- Ryan T.J. Editorial: a randomized comparison with a different focus and a new result. J Am Coll Cardiol 2001;37:59-62.[Free Full Text]
- Weintraub W.S., Stein R., Kosinski A., et al. Outcome of coronary bypass surgery versus coronary angioplasty in diabetic patients with multivessel coronary artery disease. J Am Coll Cardiol 1998;31:10-19.[Abstract/Free Full Text]
- Kip K., Faxon D., Detre K., et al. Coronary angioplasty in diabetic patients. The National Heart, Lung and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Circulation 1996;94:1818-1825.[Abstract/Free Full Text]
- Abizaid A., Costi M., Centereo M., et al. Clinical and economic impact of diabetes mellitus on percutaneous and surgical treatment of multi-vessel coronary disease patients: insights from the Arterial Revascularization Therapy Study (ARTS) trial. Circulation 2001;104:533-538.[Abstract/Free Full Text]
- Lepor N.E., Madyoon H., Kereiakes D. Effective and efficient strategies for coronary revascularization in the drug-eluting stent era. Rev Cardiovasc Med 2002;3(Suppl):S38-50.
- Sedlis S.P., Morrison D.A., Lorin J.D., et al. Percutaneous coronary intervention versus coronary bypass graft surgery for diabetic patients with unstable angina and risk factors for adverse outcomes with bypass. Outcome of diabetic patients in the AWESOME randomized trial and registry. J Am Coll Cardiol 2002;40:1555-1566.[Abstract/Free Full Text]
Top
Abstract
Introduction
