|
|
||||||||||
|
J Am Coll Cardiol, 2005; 45:1932-1938, doi:10.1016/j.jacc.2005.02.074 © 2005 by the American College of Cardiology Foundation |



* Cleveland Clinic Foundation, Cleveland, Ohio
University of Virginia Health Systems, Charlottesville, Virginia
The Lindner Center and The Ohio Heart Health Center, Cincinnati, Ohio
Duke Clinical Research Institute, Durham, North Carolina
|| Methodist-DeBakey Heart Center and Baylor College of Medicine, Houston, Texas
¶ New York University School of Medicine, New York, New York
# Oscala Heart Institute, Munroe Regional Medical Center, Oscala, Florida
** University Hospital Nottingham, Nottingham, United Kingdom
Manuscript received January 6, 2005; revised manuscript received February 17, 2005, accepted February 21, 2005.
* Reprint requests and correspondence: Dr. A. Michael Lincoff, Department of Cardiovascular Medicine, Desk F25, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195. (Email: lincofa{at}ccf.org).
| Abstract |
|---|
|
|
|---|
BACKGROUND: Previous studies have shown a survival benefit with use of platelet glycoprotein (GP) IIb/IIIa inhibitors in diabetic patients undergoing PCI. The Randomized Evaluation in Percutaneous Coronary Intervention Linking Angiomax to Reduced Clinical Events (REPLACE)-2 trial showed the non-inferiority of a strategy of bivalirudin with provisional GP IIb/IIIa inhibition compared with routine GP IIb/IIIa inhibition. The relative efficacy of these two strategies in diabetic patients has not been studied.
METHODS: We evaluated the diabetic patients enrolled in the REPLACE-2 trial to assess the impact of these antithrombotic strategies on the short- and long-term outcome after PCI.
RESULTS: The REPLACE-2 trial enrolled 1,624 diabetic patients and 4,368 non-diabetic patients. Compared with non-diabetic patients, diabetic patients had similar short-term outcome but higher mortality at 1 year (3.06% vs. 1.85%, p = 0.004). There was no difference in short-term or long-term ischemic events among the diabetic patients randomized to the two arms. Specifically, the 1-year mortality rate was non-significantly lower in the bivalirudin arm, suggesting no differential survival impact of the two strategies (2.3% vs. 3.9%). There was less minor bleeding in the bivalirudin arm in diabetic patients (12.6% vs. 24.4%, p < 0.001), whereas no difference was seen in the incidence of major bleeding (3.0% vs. 3.3%, p = 0.69).
CONCLUSIONS: Compared with routine GP IIb/IIIa inhibition, the use of bivalirudin with provisional GP IIb/IIIa inhibitors in diabetic patients is associated with no differences in clinical outcomes at 30 days, a trend toward lesser mortality at 1 year, and a reduction in minor bleeding.
| |||||
Based on this line of evidence, the use of GP IIb/IIIa inhibitors has been strongly advocated for diabetic patients undergoing PCI. The GP IIb/IIIa inhibitors, although useful for reducing peri-procedural events, are associated with an increased hazard of bleeding. Recently the Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events (REPLACE)-2 study showed that bivalirudin with provisional GP IIb/IIIa inhibition is not inferior to heparin plus GP IIb/IIIa inhibition in patients undergoing elective or urgent PCI (4).Given the benefits of GP IIb/IIIa inhibitors in diabetic patients, however, it is important to confirm that the efficacy and safety of PCI in this population are not compromised by a bivalirudin-based antithrombotic strategy in which GP IIb/IIIa inhibitors are used only selectively.
Therefore, we evaluated the subgroup of diabetic patients enrolled in the REPLACE-2 trial to assess whether there was a difference in the short- or long-term outcome with the use of the two anticoagulation strategies. Further, we compared the outcomes of diabetic and non-diabetic patients to assess whether advances in medical care have ameliorated the high-risk profile associated with diabetes after percutaneous coronary revascularization.
| Methods |
|---|
|
|
|---|
The primary end point was the composite of death, myocardial infarction, ischemia requiring urgent revascularization, or in-hospital major bleeding within 30 days. Myocardial infarction was centrally adjudicated by an independent panel and was defined as either: 1) creatine kinase or creatine kinase-myocardial band elevation
3 times the upper limit of normal if within two days of revascularization, or
2 times the upper limit of normal if not associated with revascularization; or 2) by new Q waves in two or more adjacent electrocardiographic leads. Major bleeding was considered to have occurred if any of the following definitions was satisfied: a hemoglobin decrease >4 g/dl; overt bleeding with hemoglobin decrease >3 g/dl; two or more units of blood transfused; or retroperitoneal, intraocular, or intracranial hemorrhage. Minor bleeding was defined as overt bleeding not meeting the criteria for major bleeding.
The trial excluded patients undergoing PCI as reperfusion therapy for acute myocardial infarction. Patients were also excluded if they required ongoing warfarin therapy or had been treated with unfractionated heparin within 6 h (unless the activated partial thromboplastin time was <50 s), low-molecular-weight heparin within 8 h, bivalirudin within 24 h, abciximab within 7 days, or eptifibatide or tirofiban within 12 h before randomization.
Diabetes was defined by patient report, the treating physician, or both. One-year follow-up was based on telephone contact and was obtained beyond 270 days in 98% of the cohort.
Statistical analysis. Baseline characteristics were summarized by the use of frequencies and percentages for categorical variables and means and standard deviation for continuous factors. Differences in baseline characteristics were tested with Pearson chi-square tests for categorical variables, Wilcoxon rank sum test for creatinine clearance, and activated clotting time and unpaired t test for age and weight. One-year event-free survival was illustrated with Kaplan-Meier curves, and outcomes were compared using the log-rank test. All analysis was intention to treat. A p value of < 0.05 was used as the level for statistical significance. The statistical software used was SAS version 8.0 (SAS Inc, Cary, North Carolina).
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Multiple studies have shown a worse outcome among diabetic patients after myocardial infarction (5), unstable angina (6,7), and coronary bypass grafting (810). Diabetic patients undergoing PCI seem to have a similar short-term outcome compared with non-diabetic patients, but have an increased risk of subsequent myocardial infarction, restenosis, and mortality (11). Indeed the Bypass Angioplasty Revascularization Investigation (BARI) trial showed better survival in patients randomized to coronary artery bypass grafting compared with those undergoing PCI (12). These findings were, however, not replicated in the BARI trial registry (13) and other studies (10) suggesting that PCI may be a viable revascularization strategy in carefully selected diabetic patients.
Multiple studies have focused on the short- and long-term outcomes of PCI among diabetic patients treated using contemporary techniques. In a study of 689 patients undergoing multivessel stenting at a single center, although there was no difference in the incidence of in-hospital death or myocardial infarction, the risk of target lesion revascularization or long-term mortality was significantly higher in diabetic patients (11). Similarly, data from the National Heart, Lung, and Blood Institute Dynamic registry described an almost two-fold increase in the risk of long-term mortality among diabetic patients compared with non-diabetic patients (14). Retrospective analysis from other large clinical trials replicates these findings. In the Prevention of Restenosis With Tranilast and Its Outcomes (PRESTO) trial, compared with the 8,798 non-diabetic patients, diabetic patients (n = 2,694) were more likely to have worse clinical and angiographic characteristics, although this risk profile did not translate into any apparent increase in short-term complications (15). Diabetic patients were, however, more likely to have new lesions, to have progression of previous lesions, or to die by nine months (2.1% vs. 0.9%).
Against this backdrop, the role of platelet GP IIb/IIIa inhibitors among diabetic patients undergoing PCI has undergone much scrutiny after publication of the combined analysis of the three trials of abciximab (3). In that study of 1,262 diabetic patients, use of abciximab was associated with a reduction in mortality from 4.5% to 2.5% (p = 0.031). Furthermore, in the diabetic substudy of the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting Trial (EPISTENT) at six months, abciximab therapy was associated with a reduction in target vessel revascularization or the composite of death and myocardial infarction (16). The effects of abciximab on restenosis, however, could not be confirmed among the overall cohort of patients in the Evaluation of ReoPro and Stenting to Eliminate Restenosis (ERASER) trial (17) nor among diabetic patients in the Diabetes Abciximab Stent Evaluation (DANTE) study (18). Further support for the survival advantage of GP IIb/IIIa inhibitors came from a meta-analysis of six large trials that showed a significant mortality reduction with use of intravenous GP IIb/IIIa among diabetic patients with acute coronary syndromes (2). Among the 1,279 diabetic patients undergoing PCI during index hospitalization, the mortality rate was reduced from 4.0% to 1.2% (p = 0.002) with the use of GP IIb/IIIa inhibitors. Based on these data, the use of GP IIb/IIIa inhibitors had been advocated in all diabetic patients undergoing PCI (19,20). Recently, this hypothesis was directly tested in the Is Abciximab a Superior Way to Eliminate Elevated Thrombotic Risk in Diabetics (ISAR-SWEET) study (21). Among the 701 diabetic patients undergoing elective PCI after pre-loading with clopidogrel, no difference in either peri-procedural death, myocardial infarction, or urgent revascularization or in the primary end point of death and myocardial infarction at one year (8.3% vs. 8.6%, p = 0.91) was noted among patients randomized to abciximab or placebo. Although the study has been criticized for being underpowered, it remains the only direct test of GP IIb/IIIa inhibitors in this population (22).
The REPLACE-2 trial showed the non-inferiority of a strategy of bivalirudin with provisional GP IIb/IIIa inhibitors when compared with heparin plus routine GP IIb/IIIa inhibitors in a broad spectrum of patients undergoing PCI. Given the apparent extraordinary benefit of GP IIb/IIIa blockade among diabetic patients, however, this subgroup deserves particular scrutiny. Our study provides reassuring data regarding the efficacy of the bivalirudin anticoagulant strategy among diabetic patients undergoing PCI. Not only was there no difference in short-term outcome among diabetic patients randomized to the two arms, but also the patients treated with bivalirudin had numerically better (although statistically non-significant) survival by 1-year follow-up. The number of patients in our analysis (n = 1,624) was similar to that in the prior analysis by Bhatt et al. (3) (n = 1,462) of abciximab, and the one-year mortality with the bivalirudin-based strategy (2.28%) or in the GP IIb/IIIa inhibitor arm (3.9%) is also comparable with that seen in their analysis (2.5% in the abciximab arm), or in the more contemporary Do Tirofiban and ReoPro Give Similar Efficacy Outcomes (TARGET) trial (2.1% in the tirofiban group and 2.9% in the abciximab group) (23). Although not definitive, the relatively large size of our cohort suggests that our findings are likely to be extant.
The biological basis of a long-term protective effect of a strategy of GP IIb/IIIa inhibition in diabetic patients has never been fully elucidated. Hypothesized mechanisms have included plaque stabilization, suppression of inflammatory mediators, reduced distal embolization, and an assuaged inflammatory response to vascular injury (24). Given that there are no data regarding whether bivalirudin as a direct thrombin inhibitor has these effects, the similar outcome with these two strategies cannot be fully explained. It is conceivable that parallel changes in interventional techniques, primary and secondary preventive measures, and improvements in clinical care that have occurred over the past decade may have diminished the additional long-term survival benefit of routine GP IIb/IIIa inhibition. However, it is equally plausible that a strategy of bivalirudin with provisional GP IIb/IIIa inhibition provides an analogous protective effect on the PCI site, thus explaining the similar survival in the two arms. However, these hypotheses have never been tested and need further investigation to support or refute them.
We did not find any reduction in major bleeding or the need for transfusion among diabetic patients treated with bivalirudin. This may relate to the small number of events and the lack of statistical power or may be a spurious finding given the relatively weak interaction in the setting of a secondary analysis. The incidence of major bleeding in the bivalirudin arm was similar among diabetic and non-diabetic patients, and was markedly higher among non-diabetic patients treated with GP IIb/IIIa inhibitors. Any conclusion from such data, however, needs to be made cautiously because such a protective interaction between diabetes and risk of bleeding with GP IIb/IIIa inhibitors has never been previously described.
Our results are based on a retrospective analysis of prospectively collected data and thus are prone to the limitations inherent to such studies. Further patients with ST-segment elevation myocardial infarction and non-ST-segment elevation myocardial infarction with ongoing ischemia were excluded from the REPLACE-2 trial and, therefore, our findings cannot be extended to this cohort. Data on glycemic control were not available. The small number of events combined with multiple anti-diabetes regimens prohibits a study of association of anti-diabetes therapy with short- or long-term end points. Our study was not powered to show either superiority or non-inferiority of one versus the other strategy in diabetic patients. However, given that the number of diabetic patients or events in our analysis is larger than in the previous analysis from the GP IIb/IIIa inhibitor trials, our data provide reassurance that the strategy of bivalirudin with provisional GP IIb/IIIa inhibitors does not carry any relevant mortality hazard. Given the reduction in bleeding events (albeit minor), shorter duration of infusion, and cost factors, it is likely that a large number of diabetic patients would be treated with bivalirudin and these data should serve to provide some evidence base for a clinician contemplating such a strategy.
| Conclusions |
|---|
|
|
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Feit, S. V. Manoukian, R. Ebrahimi, C. V. Pollack, E. M. Ohman, M. J. Attubato, R. Mehran, and G. W. Stone Safety and Efficacy of Bivalirudin Monotherapy in Patients With Diabetes Mellitus and Acute Coronary Syndromes: A Report From the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) Trial J. Am. Coll. Cardiol., April 29, 2008; 51(17): 1645 - 1652. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Mulukutla, H. A. Vlachos, O. C. Marroquin, F. Selzer, E. M. Holper, J. D. Abbott, W. K. Laskey, D. O. Williams, C. Smith, W. D. Anderson, et al. Impact of Drug-Eluting Stents Among Insulin-Treated Diabetic Patients: A Report From the National Heart, Lung, and Blood Institute Dynamic Registry J. Am. Coll. Cardiol. Intv., April 1, 2008; 1(2): 139 - 147. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |