|Home | About | Journals | Submit | Contact Us | Français|
We sought to determine the safety and efficacy of enoxaparin versus unfractionated heparin during percutaneous coronary intervention (PCI). Four hundred ninety-three consecutive patients undergoing elective or emergency PCI received unfractionated heparin (70 U/kg, intravenously) or enoxaparin (1 mg/kg, intravenously). Patients who had received subcutaneous enoxaparin in the emergency department were given a supplementary 0.3-mg/kg intravenous dose. There was no crossover of therapies. All patients received oral antiplatelet therapy and eptifibatide. Primary safety outcomes were bleeding and a postprocedural hemoglobin decrease of ≥3 g/dL. Troponin I levels were considered a marker for myocardial injury.
Two hundred twenty-two patients received enoxaparin, and 271 received unfractionated heparin. There were no thrombotic events or in-hospital deaths. Multivariate logistic regression analysis showed that, compared with unfractionated heparin, enoxaparin yielded a lower risk of bleeding (odds ratio [OR]=0.47; 95% confidence interval [CI], 0.21–1.05) and significantly fewer >3-g/dL decreases in hemoglobin (OR=0.45; 95% CI, 0.22–0.94). Enoxaparin also produced less of a decrease in mean platelet count (41 ± 34 vs 55 ± 63 ×109/L; P = 0.02) and in platelets >30% from baseline (OR=0.56; 95% CI, 0.31–0.99). After elective PCI, fewer enoxaparin patients had troponin I levels ≥3 times the upper limit of normal (OR=0.40; 95% CI, 0.028–0.66).
Compared with unfractionated heparin, enoxaparin entailed less bleeding during both elective and emergent PCI and less cardiac enzyme elevation in patients undergoing elective PCI. Therefore, we believe that intravenous enoxaparin is a safe alternative to unfractionated heparin in both settings.
Unfractionated heparin (UFH), a mainstay of antithrombotic therapy during percutaneous coronary intervention (PCI), is usually adjusted on the basis of the patient's activated clotting time (ACT). However, UFH has numerous well-recognized limitations: its difficulty in achieving reliable levels of anticoagulation, because of its higher degree of protein binding; its inactivation by platelet factor 4; its tendency toward platelet activation1; and its risk of heparin-induced thrombocytopenia.2 The narrow therapeutic index of UFH has been revealed in numerous studies in which almost two thirds of UFH recipients did not reach therapeutic levels within the first 12 hours of treatment3,4 and only about half of the patients were within the therapeutic range at any given time. In contrast, low-molecular-weight heparin (LMWH) has been shown to provide more reliable anticoagulation, to release less von Willebrand factor,5 and to have a minimal effect on platelet activation.1
Enoxaparin, an LMWH, has emerged as an important tool in the antithrombotic management of patients who present with acute coronary syndromes (ACSs); and randomized, controlled trials have shown this agent to be superior to UFH.6–8 In the United States, the American College of Cardiology and American Heart Association (ACC/AHA) guidelines9 give both enoxaparin and UFH a Class IA recommendation for use in patients with non–ST-segment-elevation myocardial infarction (NSTEMI). In contemporary practice, early invasive management has been shown to be superior to a conservative approach in treating both high-risk10–12 and medium-risk ACS patients.13 Consequent to the more widespread use of LMWH in ACS patients, interventional cardiologists are more likely to encounter patients in the cardiac catheterization laboratory who already have received subcutaneous or intravenous LMWH.
The current study compares intravenous enoxaparin with UFH as an anticoagulant agent during elective and emergent PCI.
We reviewed the records of 549 consecutive patients who underwent elective or emergency PCI while receiving one of 2 different anticoagulation agents at a tertiary referral center. All patients had symptoms of coronary artery disease, a positive stress-test result, or ACS that required left-sided heart catheterization and subsequent PCI. Complete baseline characteristics and short-term follow-up data were available for 493 patients, who constituted the study group. The study was approved by the institutional review board of Tulane University Hospital and Clinic.
Interventional procedures were performed according to standard techniques. Patients received either intravenous UFH or intravenous enoxaparin, depending on the preference of the interventional cardiologist in the cardiac catheterization laboratory. Of the 4 cardiologists involved, 2 used UFH and 2 used enoxaparin. The UFH group received an intravenous bolus of 70 U/kg, which was then titrated to achieve an activated clotting time of 200 to 300 seconds. The enoxaparin group received an intravenous bolus of 1 mg/kg. Patients who already had received a subcutaneous dose of enoxaparin were given a 0.3-mg/kg intravenous dose if at least 6 hours had elapsed since the last subcutaneous dose (or if they had received fewer than 3 subcutaneous doses and were, therefore, not in a steady state). There was no crossover between the 2 therapies.
Intracoronary stents were deployed in all cases. All patients received oral aspirin (325 mg) and clopidogrel (300 mg) during or at the conclusion of the procedure, followed by a maintenance dose of clopidogrel (75 mg/day). All patients were given an intravenous double bolus and infusion of eptifibatide before the lesions were crossed (180 mg/kg, followed by an infusion of 2 mg/[kg·min] for 18–24 hours afterwards); this dosage was reduced by 50% in patients with renal insufficiency. Closure devices were used in most patients immediately after the procedure, regardless of whether they had received UFH or enoxaparin. If manual pressure was used, all sheaths were removed within 4 hours after the procedure in intravenous enoxaparin recipients, after 6 hours in subcutaneous enoxaparin recipients, or when the ACT reached <170 seconds in UFH recipients.
The primary endpoints of the study were a decrease in hemoglobin levels of >3 g/dL within 48 hours after the procedure and the presence of Thrombolysis-in-Myocardial-Infarction (TIMI) major or minor bleeding. Secondary endpoints included a decrease of 30% or more in the platelet count, and, in patients undergoing elective PCI, an increase in cardiac troponin I levels (cTnI; cutoff, >0.4 ng/mL; Dade Behring; Deerfield, Ill) to >3 times the upper limit of normal (ULN).
Continuous variables are presented as mean ± SD; intergroup comparisons were made by using the unpaired Student's t test. Proportions were compared by using the χ2 test. A P value of <0.05 was considered significant. Multivariate logistic regression was used to assess the risk of prespecified outcomes between the 2 groups. All analyses were performed with Stata® 9.0 software (StataCorp; College Station, Tex).
The study group comprised 493 patients who underwent PCI. Of these patients, 222 received enoxaparin and 271 received UFH. Table I shows the patients' baseline characteristics. Of the procedures, 309 (62%) were elective, and 184 (38%) were emergency interventions in ACS patients (enoxaparin, 100 cases; UFH, 84 cases). Overall, there were no significant intergroup differences in age; sex; or a history of diabetes, hypertension, increased lipid levels, smoking, or peripheral vascular disease. Of the patients who underwent elective procedures, fewer received enoxaparin (55%) than UFH (69%). There were no thrombotic events or deaths.
In UFH patients, the mean ACT was 262 ± 36 seconds. On the basis of our group's extensive experience with the use of intravenous enoxaparin, no monitoring of this drug was performed.3,6,13–19 All patients received concomitant aspirin, clopidogrel, and eptifibatide. Of the 100 patients with ACS who received enoxaparin, 40 patients received (at the time of PCI) a supplemental 0.3-mg/kg intravenous dose in addition to the subcutaneous doses received on admission, because these patients were not in a steady state (only 2 subcutaneous doses having been administered); 21 patients received a supplemental dose because more than 6 hours had elapsed since the last subcutaneous dose. Other emergently treated patients received a full intravenous dose (1 mg/kg).
The difference in hemoglobin levels before and after PCI was 0.72 ± 0.83 for enoxaparin patients versus 0.9 ± 0.75 for UFH patients (P = 0.054), and the difference in the decrease in platelet count (×109/L) was 41 ± 34 versus 55 ± 63, respectively (P = 0.002) (see Table II).
Table III shows the results of univariate and multivariate logistic regression analysis for the primary and secondary safety endpoints, comparing enoxaparin with UFH. Multivariate logistic regression analysis was performed to adjust for age; sex; a history of coronary artery disease, hypertension, diabetes, increased lipid levels, smoking, or peripheral vascular disease; and the type of intervention performed (elective or emergency).
Compared with the UFH patients, the enoxaparin group had a significantly lower risk of a 3-g/dL decrease in their hemoglobin levels (odds ratio [OR]=0.45; 95% confidence interval [CI], 0.22–0.94). Blood transfusion was required in 2 patients (0.9%) in the enoxaparin group versus 5 patients (1.84%) in the UFH group. Table IV shows the TIMI bleeding status. For this analysis, we used the Pearson χ2 coefficient to compare the difference in the proportions for every TIMI bleeding event. The enoxaparin group had a trend toward less TIMI major or minor bleeding (enoxaparin, 4.9%; UFH, 8.6%), but this difference did not achieve significance. A univariate unadjusted analysis of the difference in the bleeding rates in both groups revealed a lower incidence of any TIMI bleeding events in enoxaparin patients (9, 4.0%) versus UFH patients (23, 8.5%) (P = 0.04).
Enoxaparin entailed a lower decrease in the mean platelet count (41 ± 34 vs 55 ± 63 ×109/L; P = 0.02) and, in the univariate analysis, a lower decrease in platelets >30% from baseline (OR=0.56; 95% CI, 0.31–0.99).
Of the patients undergoing elective PCI (excluding those with unstable angina and NSTEMI), fewer enoxaparin recipients had elevated postprocedural cTnI levels ≥3 times the ULN (OR=0.40; 95% CI, 0.028–0.66) (Table V). There was a trend in the enoxaparin group toward a lower 5-fold cTnI elevation (OR=0.22; 95% CI, 0.021–2.31) (see Table V). Overall, the enoxaparin patients were consistently less likely to have elevated cardiac enzyme levels than were the UFH patients.
The present study summarizes our cumulative experience with the procedural use of enoxaparin in a cohort of patients who underwent either emergency or elective PCI. Patients who received enoxaparin had a significantly lower risk of decreased hemoglobin levels or a decreased platelet count than did patients who received UFH. In addition, patients who underwent elective PCI and received enoxaparin had lower rates of periprocedural cTnI elevation ≥3 times the UNL.
In comparison with UFH, LMWH has several theoretical advantages, including decreased platelet activation, a less pronounced effect on platelet function, and more efficient inhibition of thrombin generation.2,20 However, despite the fact that recently updated ACC/AHA guidelines recommend either LMWH or UFH in the treatment of ACS and despite the growing evidence in favor of early invasive management, LMWH is not widely used for procedural anticoagulation during PCI. One concern is the lack of a rapid bedside assay for reliably monitoring the anticoagulative effects of LMWH, although such an assay has been explored in preliminary studies.21,22 As clinical experience with LMWH (with or without concomitant GPIIb/IIIa inhibitors) during elective and emergency PCI15–17,20–27 continues to grow, emerging data appear to confirm the efficacy, safety, and feasibility of using LMWH in this setting.
The use of intravenous enoxaparin in patients undergoing elective PCI was initially supported by 2 nonrandomized, observational studies—the National Investigators Collaborating on Enoxaparin (NICE) 1 and NICE 4 trials.15 The NICE 1 trial involved a 1-mg/kg intravenous bolus of enoxaparin, and NICE 4 involved 0.75 mg/kg of enoxaparin in conjunction with a GPIIb/IIIa antagonist. The 30-day incidence of death, myocardial infarction, and urgent revascularization was 7% for NICE 1 and 6.8% for NICE 4. At 30 days, the TIMI major and minor bleeding rates were 1.1% and 6.1%, respectively, for NICE 1, and were 0.4% and 7%, respectively, for NICE 4. These results are comparable with outcomes achieved with the use of UFH in similar PCI studies.28,29
In the Coronary Revascularization Using Integrilin and Single bolus Enoxaparin (CRUISE) study, Bhatt and coworkers16 evaluated patients undergoing elective or emergency PCI with either enoxaparin or UFH in conjunction with a GPIIb/IIIa antagonist. All patients received intravenous eptifibatide (a double bolus of 180 μg/kg, followed by a 2-μg/[kg·min] infusion for up to 18–24 hr). The enoxaparin arm had a significantly lower composite endpoint of death, myocardial infarction, urgent revascularization, bleeding, and vascular access site complications.
Lower dosages of intravenous enoxaparin have also been used in the catheterization laboratory. In a study by Choussat and associates,24 0.5 mg/kg of intravenous enoxaparin was given during elective PCI in a low-risk population of 242 patients. Eptifibatide was also administered in 26% of those patients. Anti-Xa activity levels of >0.5 IU/mL were achieved in 97.5% of the patients. The 30-day death, myocardial infarction, and urgent revascularization rate was 2.5%, with major and minor bleeding rates of 0.4% and 1.2%, respectively.
Another study30 evaluated the safety and efficacy of low-dose (0.5 mg/kg) intravenous enoxaparin combined with GPIIb/IIIa inhibitor therapy during PCI. The TIMI minor bleeding rate was 1.3%, and no TIMI major bleeding or major adverse cardiac events were reported.
In the STEEPLE trial (Safety and Efficacy of Enoxaparin in Percutanous Coronary Intervention: An International Randomized Evaluation),31 Montalescot and coworkers compared UFH with enoxaparin at 2 different doses (0.75 mg/kg or 0.5 mg/kg) during PCI. Enoxaparin at 0.5 mg/kg was associated with lower bleeding rates. Target anticoagulation levels were also more readily reached with the use of enoxaparin.
In the setting of ACS, the Superior Yield of the New strategy of Enoxaparin, Revascularization and Glycoprotein IIb/IIIa inhibitors (SYNERGY) trial32 compared enoxaparin with UFH in high-risk ACS patients. Enoxaparin was non-inferior to UFH in the early invasive management of these patients, although bleeding events tended to be more frequent with enoxaparin. The ACS patients in the SYNERGY study usually received antithrombotic therapy before arriving in the catheterization laboratory, so the excess bleeding that occurred with enoxaparin may have resulted from a longer duration of therapy with multiple agents.
The results of the current study add to the earlier preliminary experience with enoxaparin for procedural anticoagulation and suggest that this agent may be superior to UFH, because enoxaparin is less likely to result in decreased hemoglobin levels and platelet counts after PCI. Moreover, the enoxaparin recipients had less cTnI elevation after elective PCI, which suggests that enoxaparin may reduce the risk of procedure-related thrombotic events in the microcirculation.
After PCI, elevated creatine kinase levels (especially ≥3 times the ULN) have been shown to correlate strongly with future adverse outcomes.30 Studies have also shown that increases in cardiac troponin T and I levels predict future adverse outcomes.31,33–36
Although our study was not randomized—the interventional cardiologists could not predict or select which patients would receive intervention—the allocation of patients to receive either enoxaparin or UFH (all comers in any given day of procedure coverage) provided an equal distribution of baseline characteristics. Even so, without formal randomization and blinding, there may still be unmeasured confounding variables that were not included in the logistic regression model. In addition, the present study was not powered to detect differences in long-term outcome events, such as major adverse cardiac events and death. Only events that occurred during hospitalization were documented, and those events were infrequent in this cohort. The present study does, however, include the use of thienopyridines and GPIIb/IIIa inhibitors during PCI.
Another possible limitation of our study is the lack of monitoring for the effects of enoxaparin (that is, anti-Xa activity) during PCI. However, despite the lack of monitoring and the fact that 40% of the enoxaparin-treated ACS patients received only 2 subcutaneous doses of enoxaparin before undergoing PCI, outcomes in the enoxaparin patients were at least as satisfactory as in the UFH recipients, and enoxaparin resulted in fewer bleeding complications. Presumably, administration of a 0.3-mg/kg intravenous bolus at the time of PCI helped overcome some of the rapid-treatment risks. Lower doses of enoxaparin may be effective during PCI when combined with more aggressive antiplatelet therapy for ACS.
Our experience with 493 consecutive patients undergoing PCI suggests that enoxaparin is a potentially safer and more effective antithrombotic agent than UFH for procedural anticoagulation during PCI. Enoxaparin was associated with less thrombocytopenia, fewer hematologic derangements, and less bleeding. Furthermore, during elective PCI, fewer enoxaparin patients had elevated cTnI levels; this finding suggests that enoxaparin more effectively prevents thrombotic events. We hypothesize that some of the beneficial effects of enoxaparin during PCI may be related to its role in reducing platelet activation. Our results highlight the need for prospective, randomized, controlled clinical trials to evaluate the safety and efficacy of enoxaparin during emergency PCI and to determine whether the decreased platelet activation associated with enoxaparin has significant clinical consequences.
Address for reprints: José G. Díez, MD, FACC, Department of Medicine–Cardiology, Baylor College of Medicine, Suite 9.85, BCM 620, 1709 Dryden Road, Houston, TX 77030. E-mail: ude.mcb@zeid
Dr. Díez has served as a consultant for Sanofi-Aventis. Dr. Ferguson has served as a consultant and has received research support from Sanofi-Aventis; he is currently an employee of The Medicines Company. This research study involved no industry support or participation.