REPLACE II Substudies: Effects of Clopidogrel Pretreatment and Cost-Effectiveness Analysis

Luis Gruberg, MD, FACC


December 30, 2003

Editorial Collaboration

Medscape &

The second Randomized Evaluation in PCI Linking Angiomax to Reduced Clinical Events (REPLACE-2) was a double-blinded, multicenter, noninferiority trial designed to access whether the use of bivalirudin (Angiomax; The Medicines Company, Parsippany, New York) plus provisional glycoprotein (GP) IIb/IIIa inhibition could be equivalent or not inferior to the gold standard of heparin and a GP IIb/IIIa inhibitor for the prevention of ischemic complications in patients undergoing urgent or elective percutaneous coronary intervention (PCI).[1]

The noninferiority of bivalirudin to heparin plus GP IIb/IIIa inhibitor therapy was confirmed by the 30-day and 6-month results showing that there was no significant difference noted between the 2 groups with respect to the incidence of the primary quadruple composite endpoint of death, myocardial infarction (MI), urgent revascularization at 30 days, and major bleeding or in the secondary triple composite endpoint of death/MI/urgent revascularization at 30 days.

Two respective REPLACE-2 substudies presented at the American Heart Association Scientific Sessions 2003 sought to determine the effect of clopidogrel pretreatment and cost.


Between October 2001 and August 2002, a total of 4651 patients in the United States who underwent nonemergent PCI were randomized to receive bivalirudin (0.75 mg/kg bolus, 1.75 mg/kg/hour infusion) with provisional GP IIb/IIIa (n = 2319) vs heparin (65 U/kg, maximum 7000 U) plus GP IIb/IIIa (n = 2332) (Figure 1). Patients were included if they needed planned or elective PCI. Patients with MI, chronic renal failure, prior GP IIb/IIIa or low-molecular-weight heparin administration, and recent major trauma or surgery were excluded from the study. All patients received aspirin. Clopidogrel pretreatment was strongly encouraged 2-12 hours prior to the procedure (300-mg loading dose followed by 75 mg/day for 30 days).

Figure 1. REPLACE-2: trial design.

The majority of patients received eptifibatide (60%) and 40% received abciximab. Among patients in the bivalirudin group, only 7.7% were concomitantly treated with a GP IIb/IIIa inhibitor (Table 1).

Table 1. REPLACE-2: Baseline Clinical Characteristics
Characteristic Bivalirudin
(n = 2319)
(n = 2332)
Age (yrs) 63 ± 11 63 ± 11
Female (%) 27 27
Diabetes (%) 30 28
ST elevation MI < 7 days (%) 8 8
Acute coronary syndrome < 2 days (%) 16 16
Acute coronary syndrome > 2 days (%) 20 21
Eptifibatide (%) 60 60
Abciximab (%) 40 40
Thienopyridine > 2 hours pretreatment (%) 55 54

There was no difference in the rate of the primary or secondary endpoints between the bivalirudin and heparin groups. Thirty-day ischemic events including death, MI, and urgent revascularization were similar in the 2 groups, but major and minor bleeding complications and thrombocytopenia were significantly higher in the heparin plus GP IIb/IIIa inhibitor group (Figure 2).

Figure 2. REPLACE-2: death/MI/revascularization/bleeding at 30 days.

The same noninferiority results noted at 30-day follow-up remained consistent at 6-month follow-up; there was no significant difference in the rate of death, MI, or the need for revascularization at 6 months (Figure 3).

Figure 3. REPLACE-2: death/MI/urgent revascularization at 6-month follow-up.
Clopidogrel Pretreatment Substudy

Presenter: Jacqueline Saw, MD, Cleveland Clinic Foundation (Cleveland, Ohio)

As per the REPLACE-2 trial protocol, clopidogrel pretreatment was recorded for a prespecified subgroup analysis. Testing the premise that clopidogrel pretreatment may improve patient outcomes following PCI, investigators sought to determine whether the outcomes observed with bivalirudin were influenced by clopidogrel pretreatment.[2]

Pretreatment Substudy: Aim

The current study was a posthoc analysis of REPLACE-2 patients that evaluated the effect of clopidogrel pretreatment on the relative efficacy of bivalirudin vs heparin on both the primary quadruple endpoint and the secondary triple endpoint.

Pretreatment Substudy: Results

The present study stratified patients according to whether they received clopidogrel pretreatment. Of the 6010 patients enrolled in REPLACE-2, 5893 of them received clopidogrel (85% in both groups); 841 patients were not pretreated vs 5052 patients who did receive pretreatment (Figure 4).

Figure 4. Pretreatment substudy design.

Baseline clinical characteristics and the procedures performed were well balanced between patients who did not receive clopidogrel pretreatment vs those who did (Table 2).

Table 2. Pretreatment Substudy: Baseline Characteristics
Characteristic No Clopidogrel Pretreatment
(n = 841)
Clopidogrel Pretreatment
(n = 5052)
Age (yrs) 63 ± 11 63 ± 11
Female (%) 28 25
Diabetes (%) 24 28
Hypertension (%) 67 67
S/P MI (%) 35 38
S/P CVA (%) 3 2
Interventions Performed
PCI (%) 94 97
Stent (%) 80 87
Plain balloon (%) 11 8
GP IIb/IIIa blockade (%) 5 7
CVA, cerebrovascular accident; GP, glycoprotein; MI, myocardial infarction; PCI, percutaneous coronary intervention; S/P, status post

Investigators found that pretreatment with clopidogrel, irrespective of the timing of pre-administration, did not affect the relative efficacy of bivalirudin compared with heparin for primary and secondary endpoints. For the bivalirudin group, clopidogrel pretreatment was associated with lower event rates compared with no pretreatment. In addition, a nonsignificant trend was observed in those randomized to heparin.

By multivariate analysis, clopidogrel pretreatment was associated with a trend toward a reduction of primary and secondary endpoints (OR 0.85, P = .18; and OR 0.82, P = .14, respectively). The interaction test between clopidogrel pretreatment and study drug randomization was not significant, indicating no difference in outcomes between treatment groups, regardless of clopidogrel pretreatment.

Pretreatment Substudy: Conclusions
  1. Clopidogrel pretreatment at all doses and times administered in the REPLACE-2 study did not influence the relative efficacy of bivalirudin vs heparin plus GP IIb/IIIa inhibitors in patients undergoing PCI

  2. Clopidogrel pretreatment was associated with a trend toward improved clinical outcomes in both groups following PCI

Pretreatment Substudy: Comments

Although the results of this posthoc analysis of patients enrolled in the REPLACE-2 study failed to show a statistically significant benefit of pretreatment with clopidogrel, there was a clear trend in that direction. We have to realize that pretreatment is not always possible in this era of same-day hospitalization to PCI to discharge, but based on the results of these trials, it seems that every effort should made to pretreat patients with clopidogrel. Nevertheless, we have to take into account the problem that some of these patients will not have previous diagnostic angiography and, therefore, there is the possibility that they may require urgent or emergent coronary artery bypass surgery, which could be complicated by higher bleeding rates.

Cost-Effectiveness Analysis

Presenter: David J. Cohen, MD, Beth Israel Deaconess Medical Center (Boston, Massachusetts)

Despite the fact that GP IIb/IIIa inhibition has been shown to reduce ischemic complications in a broad range of patients undergoing PCI, concerns over bleeding complications and costs dissuade some physicians from prescribing this therapy. On the basis of the results of REPLACE-2, the new direct thrombin inhibitor bivalirudin with the provisional use of GP IIb/IIIa inhibitors may be a suitable alternative to heparin plus GP IIb/III therapy. The ability to only use GP IIb/IIIa inhibitors on a provisional basis when patients are concomitantly treated with bivalirudin may result in significant cost savings.

Cost-Effectiveness Analysis: Aim

The current substudy was conducted to determine the impact of both ischemic and bleeding complications on the cost of PCI in contemporary practice, investigators compared the in-hospital and 30-day costs for patients undergoing PCI using the 2 anticoagulation regimens.[3]

Cost-Effectiveness Analysis: Methods

Cost data were prospectively collected through 30-day follow-up for all US patients and analyzed on an intention-to-treat basis. Hospital costs were determined by standard methods, such as measuring or evaluating:

  • Anticoagulants according to the number of vials and average wholesale costs for each drug assuming that any unused drug would be discarded

  • Cath lab procedures (to determine resource utilization and current unit costs)

  • Hospital charges obtained for 2709 patients and converted to costs based on department-level cost-to-charge ratios

  • Physician services based on the 2002 Medicare Fee Schedule

Cost-Effectiveness Analysis: Results

There was no difference in procedural resource utilization between patients randomized to bivalirudin or heparin (Table 3).

Table 3. Cost-Effectiveness Analysis: Procedural Resource Utilization
(n = 2319)
(n = 2332)
Procedure duration (min) 69 ± 55 68 ± 53
Number of stents 1.36 ± 1.0 1.31 ± 1.0
Number of balloons 1.27 ± 1.1 1.24 ± 1.1
Number of guides 1.6 ± 1.1 1.6 ± 1.1
Contrast volume (mL) 209 ± 119 208 ± 120

As was to be expected, when analyzing the total procedural costs, treatment with heparin was significantly more expensive than bivalirudin, accounting for an increase of $335, all of it driven by the higher cost of GP IIb/IIIa inhibitors (Table 4).

Table 4. Cost-Effectiveness Analysis: Procedural Costs
(n = 2319)
(n = 2332)
Bivalirudin ($) 453 ± 299 0 < .001
GP IIb/IIIa inhibitors ($) 77 ± 301 932 ± 545 < .001
Other supplies ($) 715 ± 238 709 ± 236 NS
Devices ($) 2075 ± 1339 2024 ± 1257 NS
Personnel ($) 123 ± 56 122 ± 54 NS
Overhead ($) 1162 ± 496 1153 ± 482 NS
Total ($) 4606 ± 1916 4931 ± 1793 < .001

Among patients in the bivalirudin group, only 7.7% were concomitantly treated with a GP IIb/IIIa inhibitor. Independent predictors of hospital cost included unplanned coronary bypass surgery, repeat PCI, major bleeding, thrombocytopenia, and MI. In-hospital and 30-day costs were reduced by $405/patient and by $374/patient in the bivalirudin group (P < .001 for both values).

Cost-Effectiveness Analysis: Conclusions

On the basis of their results, investigators concluded that:

  • In the REPLACE-2 population, use of bivalirudin with provisional GP IIb/IIIa inhibition compared with heparin plus routine GP IIb/IIIa inhibition led to similar outcomes, fewer bleeding complications, and reduced medical care costs by $374/patient

  • Cost savings were consistent across a range of clinical subgroups, but tended to be less among patients preselected for eptifibatide (-$185) and greater for patients selected for abciximab (-$560/patient)

  • In addition to the anticoagulants themselves, approximately 20% of the cost savings seen with bivalirudin were directly related to reductions in bleeding complications and thrombocytopenia

  • In the future, greater savings might be possible in selected patients by using the short infusion duration of bivalirudin to support outpatient PCI in low-risk procedures

Cost-Effectiveness Analysis: Comments

Bivalirudin has simplified the use of antithrombotic adjuvant therapy in patients undergoing PCI. The results from the REPLACE-2 trial proved that the use of this drug was not only as good as heparin plus a GP IIb/IIIa inhibitor, but it was also associated with lower rates of bleeding complications and thrombocytopenia. With this excellent cost analysis by Cohen and colleagues, we can now be assured that it is also less expensive -- certainly a major concern in these days of cost containment. As the authors state, bivalirudin may pave the way for same-day PCI in low-risk patients.

  1. Lincoff AM, Bittl JA, Harrington RA, for the REPLACE-2 Investigators. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention REPLACE-2 randomized trial. JAMA. 2003;289:853-863.

  2. Saw J, Desmet W, Lincoff AM, et al. Clopidogrel pretreatment does not influence the relative efficacy of bivalirudin with provisional GP IIb/IIIa blockade vs heparin with routine GP IIb/IIIa blockade. A REPLACE-2 substudy. Circulation. 2003;108(Suppl IV):IV-570. Abstract 2597.

  3. Cohen DJ, Lavelle T, Chen H-L, et al. Cost-effectiveness of bivalirudin with provisional glycoprotein 2b/3a inhibition vs heparin + routine glycoprotein 2b/3a inhibition for contemporary PCI: results from the REPLACE II trial. Circulation. 2003;108(Suppl IV):IV-570. Abstract 2601.