|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Medicine (Drs. Macie, Forbes, and Douketis) and Clinical Epidemiology and Biostatistics (Dr. Foster), McMaster University, Hamilton, ON, Canada.
Correspondence to: James D. Douketis, MD, St. Josephs Hospital, Room F-538, 50 Charlton Ave East, Hamilton, ON, Canada, L8N 4A6; e-mail: jdouket{at}mcmaster.ca
| Abstract |
|---|
|
|
|---|
Design: Retrospective chart review.
Setting: Coronary care unit of a tertiary-care teaching hospital.
Patients: Patients with a discharge diagnosis of an ACS who received at least one dose of enoxaparin, 1 mg/kg, were eligible for this study. Enoxaparin dosing practices, factors that might influence the safety of enoxaparin administration, and bleeding events were documented. Multivariable regression analysis was used to identify independent predictors of bleeding in this clinical setting.
Results: Of 208 patients with an ACS who received enoxaparin, 48 patients (23%) received a dose that was > 10% or < 10% of the recommended 1 mg/kg dose, 18 patients (9%) did not have body weight documentation to guide enoxaparin dosing, and 17 patients (8%) had significant renal impairment (serum creatinine > 150 µmol/L), with the potential for bioaccumulation of enoxaparin. There were 35 bleeding events (17%), of which 8 events (4%) were major. Risk factors for any bleeding (major or minor) were increasing patient age (odds ratio [OR], 1.57; 95% confidence interval [CI], 1.13 to 2.20), coadministered nonsteroidal anti-inflammatory or antiplatelet drug therapy (OR, 2.38; 95% CI, 1.06 to 5.38), and number of enoxaparin doses (OR, 2.15; 95% CI, 1.25 to 3.68). Risk factors for major bleeding were increasing patient age (OR, 2.56; 95% CI, 1.05 to 6.28) and coadministered clopidogrel (OR, 7.70; 95% CI, 1.16 to 51.9).
Conclusion: In this clinical practice assessment of patients with an ACS, the use of enoxaparin was suboptimal, with the potential to increase bleeding complications. Coadministered clopidogrel, other drugs that affect hemostasis, and increasing age conferred an increased bleeding risk.
Key Words: acute coronary syndrome bleeding low-molecular-weight heparin
| Introduction |
|---|
|
|
|---|
However, the randomized controlled trials that investigated enoxaparin and other LMWHs in patients with an ACS have limitations in addressing practical issues that may impact on the bleeding risk of patients assessed in "real-world" clinical practice. First, patients with renal insufficiency, in whom there is the potential for bioaccumulation of LMWH,8 were excluded from these clinical trials, as were patients weighing > 100 kg in whom there are scant data as to the safety of weight-based LMWH dosing.1213 Second, the bleeding risks associated with enoxaparin therapy were not addressed in the context of coadministered aspirin and clopidogrel therapy, which are frequently used for the initial treatment of an ACS.1011 Furthermore, co-administration of other drugs that affect hemostasis, such as nonsteroidal anti-inflammatory drugs (NSAIDs), was also not addressed in previous studies. Third, these randomized trials enrolled selected patients who were at low risk for bleeding complications,14 and LWMH was administered in a closely monitored clinical setting. Consequently, the incidence of bleeding complications in these studies may not be representative of the bleeding risks in a real-world clinical setting. We, therefore, performed a 1-year audit of clinical practice relating to the administration of enoxaparin for the initial treatment of patients with an ACS. The objectives of this study were as follows: (1) to describe real-world enoxaparin dosing practices, and (2) to identify risk factors for bleeding complications in this clinical setting.
| Materials and Methods |
|---|
|
|
|---|
To be to be eligible for this chart review, patients had to satisfy prespecified criteria so that the study population was homogeneous in terms of a diagnosis of an ACS, and received a standardized antithrombotic regimen that involved enoxaparin, 1 mg/kg bid. In this way, we would minimize potential co-interventions that might influence the determinants of bleeding, and the study findings would be generalizable to a prespecified patient group. Thus, patients included in the analysis had to satisfy the following inclusion criteria: (1) age > 18 years; (2) discharge diagnosis of unstable angina or nonST-segment elevation myocardial infarction, defined by chest pain of > 30 min duration or increased frequency of angina at rest or with minimal exertion in the previous 12 h, and new ST-segment depression or T-wave inversion on electrocardiography or cardiac troponin-I > 0.5 µmol/L; (3) received at least one dose of enoxaparin, 1 mg/kg, by subcutaneous injection. Patients were therefore excluded if they received an antithrombotic treatment that might bias an assessment of bleeding risks associated with the prespecified treatment of interest (ie, enoxaparin, 1 mg/kg bid): (1) IV unfractionated heparin, (2) thrombolytic agent; and (3) a LMWH dose or preparation other than the prespecified enoxparin regimen.
For patients with multiple hospital admissions during the surveillance period, only the first hospital admission was counted to minimize bias that may result from including patients on multiple occasions who might or might not have had a predisposition to bleeding. Patients who required a percutaneous coronary intervention or coronary bypass surgery were transferred to the regional cardiovascular center, and did not undergo further clinical follow-up after transfer out of the coronary care unit.
Outcomes
The following data were extracted from the patient medical records: (1) clinical characteristics and laboratory test results; (2) initial enoxaparin dose, changes in enoxaparin dose and reason for dose changes, and number of enoxaparin doses administered; and (3) bleeding events, defined a priori as major or minor. Major bleeds were episodes that were clinically overt and associated with one of the following outcomes: transfusion of
2 U of packed RBCs; decrease in hemoglobin > 20 g/L within 24 h; retroperitoneal, intracranial or body cavity bleeding; or death. Minor bleeding was defined as any other bleeding episode, including enoxaparin injection site bruising and transient epistaxis or hematuria.
Potential Risk Factors for Bleeding
The following patient clinical and treatment characteristics were identified, a priori, as potential determinants of bleeding risk: (1) age; (2) sex; (3) body weight; (4) serum creatinine; (5) creatinine clearance, calculated based on the Cockroft-Gault formula;15 (6) number of enoxaparin doses received; (7) co-administered aspirin, clopidogrel, or NSAIDs; and (8) previous peptic ulcer disease.
Statistical Analysis
The proportion and clinical characteristics of patients who had body weight documentation and received a weight-based dose of enoxaparin was described. Univariable and multivariable logistic regression analysis was used to identify independent predictors of bleeding. The odds ratio (OR) and corresponding 95% confidence interval (CI) was reported for each variable in the model. Variables were retained in the final multivariable model if their level of significance was
0.10. Statistical significance was declared for a predictor variable at p < 0.05.
| Results |
|---|
|
|
|---|
|
|
2 U of packed RBCs. None of the bleeding events were fatal. Risk factors for any (major or minor) and major bleeding, based on the univariable analysis, are presented in Table 3
. In the multivariable analysis, risk factors for any bleeding, and major bleeding are presented in Table 4and Table 5 , respectively. Risk factors that were independently associated with an increased risk for any bleeding were increasing age, coadministered nonsteroidal anti-inflammatory or antiplatelet drug therapy, and number of enoxaparin doses administered. Risk factors that were independently associated with an increased risk for major bleeding were increasing age, and coadministered clopidogrel therapy. Two factors were associated with an increased risk for major bleeding, although these associations did not quite attain statistical significance: the number of enoxaparin doses administered (p = 0.084), and increasing serum creatinine concentration (p = 0.079). The use of angiotensin-converting enzyme inhibitors were associated with a significantly lower risk of any bleeding (p = 0.25), although this association did not reach statistical significance with major bleeding (p = 0.062).
|
|
|
| Discussion |
|---|
|
|
|---|
There were missed opportunities for optimal enoxaparin administration that would incorporate weight-based dosing and dose adjustments, or alternative drug treatment with unfractionated heparin because of renal impairment. Observational studies1617 have provided some insight as to the determinants of drug-related adverse events that could apply to LMWH for the treatment of patients with an ACS. Suggestions that may improve the safety of LMWH administration include standing orders for patient weight documentation, dose limits in patients weighing > 100 kg, and serum creatinine thresholds. In addition, documentation of risk factors for bleeding, including a previous history of bleeding, active peptic ulcer disease, thrombocytopenia, and coadministration of drugs affecting hemostasis, might heighten awareness of the bleeding risks associated with LMWH.
Clinically important, or major, bleeding occurred in eight patients (4%), a rate that may be higher than that observed in structured clinical trials. There are two clinical trials involving patients with an ACS who received an enoxaparin dose regimen (ie, 1 mg/kg bid) and treatment duration (ie, 2 to 14 days) that was comparable to that of our study population. In these studies,57 the incidence of major bleeding was lower than that in our study population, with an incidence of 1.5% (29 of 1,953 patients) and 1.9% (6 of 309 patients), respectively. Although a higher rate of major bleeding of 6.3% (102 of 1,607 patients) was reported in one clinical trial6 of patients with an ACS who received the same enoxaparin dose regimen, most of these bleeds occurred during or after coronary artery bypass surgery and were not attributable to enoxaparin therapy administered for the initial treatment of an ACS, as in our study.
There were several clinical factors that appeared to influence the risk of bleeding in our study population. Coadministration of clopidogrel, which was received by 26% of patients, was the strongest predictor of bleeding. Such patients had an almost eightfold increased risk of major bleeding compared to patients who did not receive clopidogrel (OR, 7.70; 95% CI, 1.16 to 51.9). The use of NSAIDs or other antiplatelet drugs affecting hemostasis was also a predictor of bleeding, as such patients had approximately a twofold increased risk of any bleeding (OR, 2.38; 95% CI, 1.06 to 5.38). These findings are not surprising because coadministration of multiple drugs affecting hemostasis will increase the risk of bleeding.18192021 Increasing age was also associated with an increased risk for bleeding (OR, 1.57; 95% CI, 1.13 to 2.20). Increasing age as a potential risk factor for bleeding has not been addressed in clinical trials of LMWH in patients with an ACS. In studies investigating LMWH therapy in patients with venous thromboembolism, the findings are variable. One study22 found that there was no increased risk of major bleeding in an elderly population, while a French survey23 noted increased bleeding episodes in such patients. Elderly patients may have clinical factors associated with an increased risk for bleeding, such as generalized frailty, a tendency for falls, and occult peptic ulcer disease,2425 but are not accounted for in assessments of bleeding risk factors. Another relevant issue is the potential for bioaccumulation of enoxaparin in patients with renal impairment because enoxaparin is cleared primarily by the kidney.26 In our study, a creatinine clearance < 25 mL/min, which corresponds to significant renal impairment, occurred in 3% of all patients but in 25% of patients with major bleeding. In a retrospective cohort study27 of 106 patients with an ACS who received enoxaparin, 1 mg/kg, patients with renal dysfunction had a higher risk of any bleeding (51% vs 22%, p < 0.001) and major bleeding (30% vs 2%, p < 0.001) than those with normal renal function. Furthermore, in a substudy of the Thrombolysis in Myocardial Infarction 11B trial28 involving patients with an ACS who received enoxaparin, 1 mg/kg, the incidence of major bleeding was significantly higher in patients with renal dysfunction than those with normal renal function (7.5% vs 1.2%, p < 0.01). Some authorities have suggested that laboratory monitoring of the anticoagulant effect of LMWH therapy, with anti-factor Xa heparin measurements, might be warranted if LMWHs are administered to patients with significant renal insufficiency.29 Our finding that use of angiotensin-converting enzyme inhibitor therapy appeared to decrease the risk of bleeding complications (OR, 0.25; 95% CI, 0.08 to 0.84) is difficult to explain, due to a lack of biological plausibility, and may be due to chance.
There are potential limitations of this study that should be addressed. First, we acknowledge that as a retrospective chart review, there may have been inadequate documentation of relevant data that could influence the validity of our results. However, our analysis was based on drug use, which requires specific documentation in the patient record, and objective clinical information, such as patient age and weight that is not subject to biased interpretation. Second, although the bleeding outcomes that we included were initially documented based on a subjective clinical assessment, we used prespecified standardized criteria to differentiate clinically important, or major, bleeding from other bleeding episodes that might have been preferentially documented in patients medical record because of prior knowledge of LMWH therapy. Third, we did not assess if patients who required invasive cardiac procedures, such as coronary angiography or coronary revascularization, had an increased risk for bleeding since these procedures, if required, were undertaken at another institution. In one study30 of patients with an ACS who, during treatment with enoxaparin, 1 mg/kg bid, or IV unfractionated heparin, required urgent coronary bypass surgery, there was a higher risk of postoperative bleeding in patients who received enoxaparin (7.9% vs 3.7%, p = 0.03). However, in patients with an ACS who required a percutaneous coronary intervention, concurrent therapy with enoxaparin, 1 mg/kg bid, did not appear to increase the risk for bleeding complications when the intervention was undertaken within 8 h of the last subcutaneous dose of enoxaparin.31
To summarize, our clinical practice assessment of enoxaparin therapy in patients with an ACS found that dosing practices were suboptimal, and might confer a higher-than-expected risk of clinically important bleeding complications. Increasing patient age, and co-administration of clopidogrel and other drugs that affect hemostasis were major determinants of bleeding.
| Footnotes |
|---|
Received for publication August 6, 2003. Accepted for publication November 3, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W. Lim, F. Dentali, J. W. Eikelboom, and M. A. Crowther Meta-Analysis: Low-Molecular-Weight Heparin and Bleeding in Patients with Severe Renal Insufficiency Ann Intern Med, May 2, 2006; 144(9): 673 - 684. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. A. Spinler, P. Dobesh, C. Macie, and J. Douketis Dose Capping Enoxaparin Is Unjustified and Denies Patients With Acute Coronary Syndromes a Potentially Effective Treatment Chest, June 1, 2005; 127(6): 2288 - 2290. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |