In this meta-analysis, enoxaparin was superior to unfractionated heparin in reducing mortality and bleeding outcomes during percutaneous coronary intervention, particularly in patients undergoing primary percutaneous coronary intervention for ST elevation myocardial infarction. Since early 2000 data have accumulated on enoxaparin in varied percutaneous coronary intervention settings. This current meta-analysis, with information on more than 30
000 patients, showed a 34% statistically significant reduction in mortality (1.66% absolute risk reduction) in patients receiving enoxaparin during percutaneous coronary intervention compared with unfractionated heparin. This survival benefit is supported by concomitant reductions in both ischaemic and major bleeding complications. All sensitivity analyses of mortality confirmed a genuine difference between the two drugs. Subgroup analyses suggested that the benefits on mortality and ischaemic complications were largely driven by the superiority measured in patients undergoing primary percutaneous coronary intervention for ST elevation myocardial infarction, whereas the better safety outcomes might be driven by the intravenous (versus subcutaneous) use of enoxaparin.
This meta-analysis confirms the results recently reported in the ATOLL (Acute ST-elevation myocardial infarction Treated with primary angioplasty and intravenous enoxaparin Or unfractionated heparin to Lower ischemic and bleeding events at short- and Long-term follow-up) randomised trial.14
Compared with unfractionated heparin, intravenous enoxaparin at a dose of 0.5 mg/kg reduced death or resuscitated cardiac death in patients undergoing primary percutaneous coronary intervention by 42% (P=0.049) and death or myocardial infarction by 37% (P=0.02).14
Although the 40% relative risk reduction in all cause mortality associated with enoxaparin in ATOLL was not significant (P=0.08) owing to lack of power, it is consistent with the 38% reduction in mortality found in the group with ST elevation myocardial infarction in the current meta-analysis (P<0.001), and more specifically with the 48% reduction of mortality in patients undergoing primary percutaneous coronary intervention (P<0.001). The survival benefit associated with enoxaparin was present in both risk of bias subgroups in our meta-analysis; in low risk of bias studies (randomised trials and retrospective analysis of randomised trials) and in those showing higher risk of bias (non-randomised studies).
This reduction in mortality is likely to be related to the favourable effects of enoxaparin in the prevention of ischaemic complications, which were also shown in this meta-analysis. Consistent reductions in ischaemic end points were observed in the overall analysis as well as in the five largest randomised studies,13
which together represented two thirds of the weight of ischaemic events in our meta-analysis. In comparison with unfractionated heparin, enoxaparin has been shown to be more stable and have more predictable pharmacokinetics,1
providing an optimal level of anticoagulation at the time of the procedure in more than 90% of patients, by whatever route the drug is administered.9
These optimal levels of anticoagulation have also been related to better ischaemic and survival outcomes.10
Moreover, additional endothelial and anti-inflammatory properties of enoxaparin6
may play an additional part in the prevention of ischaemic complications of acute coronary syndrome.
Improved safety might also contribute to the reduction in mortality rates. Previous studies have shown that bleeding and red blood cell transfusion have deleterious effects52
and have an effect on ischaemic outcomes as well as on mortality.53
Therefore the 20% reduction in major bleeding associated with enoxaparin might also have affected ischaemic and mortality outcomes. This result is consistent among all subgroups, with the exception of studies in which subcutaneous enoxaparin was used in comparison with unfractionated heparin, when no difference was seen. It seems that the reduction in major bleeding was mostly observed with intravenous enoxaparin, but the P value for interaction was not significant probably because of the heterogeneity in risk levels of populations in the two subgroups (subcutaneous versus intravenous). Indeed, the intravenous route provides immediate anticoagulation, with rapid clearance,49
avoiding exposure to prolonged anticoagulation after percutaneous coronary intervention, and in this study was associated with a 34% reduction in major bleeding (absolute risk reduction 1.52%) compared with unfractionated heparin. Therefore this meta-analysis confirms the benefit of enoxaparin measured in the individual randomised STEEPLE1
(elective angioplasty) and ATOLL14
(primary angioplasty) studies, with enough power to show a reduction in mortality. Patients with ST elevation myocardial infarction obviously gain a large benefit from enoxaparin on ischaemic end points and mortality. Although favourable, the magnitude of these effects seems less important in other clinical situations.
Comparisons with other reviews
Other new anticoagulants have been compared with unfractionated heparin in the setting of percutaneous coronary intervention. Bivalirudin alone compared with unfractionated heparin plus glycoprotein IIb/IIIa inhibitors has consistently shown improved safety in percutaneous coronary intervention, associated with a reduction in mortality in one trial.54
Head to head comparisons of unfractionated heparin alone in percutaneous coronary intervention have also suggested a better safety profile of bivalirudin, but without significant advantage on the net clinical benefit or mortality.58
Finally, a recent meta-analysis of nine trials, totalling almost 30
000 patients, confirmed the reduction in major bleeding complications from use of bivalirudin compared with unfractionated heparin, but failed to show any benefit on mortality, whereas a trend for higher risk of myocardial infarction was noted.61
In contrast with this, a meta-analysis of nine studies, totalling 16
286 patients, comparing low molecular weight heparin with unfractionated heparin in the setting of percutaneous coronary intervention for ST elevation myocardial infarction reported a reduction in both mortality and major bleeding consistent with our findings.62
An alternative anticoagulant, the synthetic factor Xa inhibitor fondaparinux, has been tested in acute coronary syndromes.63
Results were not favourable in patients with ST elevation myocardial infarction undergoing primary percutaneous coronary intervention64
and the drug has been tested only sparingly in elective percutaneous coronary intervention.65
A significant increase in catheter related thrombosis with fondaparinux prompted guidelines committees on both sides of the Atlantic to recommend unfractionated heparin as adjunctive treatment at the time of percutaneous coronary intervention. The recent results of the randomised Fondaparinux Trial With Unfractionated Heparin During Revascularization in Acute Coronary Syndromes (FUTURA) suggest that a standard unfractionated heparin dose of 85 IU/kg bolus, with an additional bolus if needed to achieve activated clotting time of 300 to 350 seconds, is preferable to a lower dose in patients previously treated with subcutaneous fondaparinux.3
In a pooled analysis of 19
085 patients with acute coronary syndrome invasively managed, fondaparinux reduced major bleeding compared with a heparin based strategy, with similar rates of ischaemic events and no reduction in mortality.66
Strengths and limitations of this meta-analysis
Our meta-analysis has limitations and was not carried out on individual patients’ data, which if possible would have strengthened the results, especially for subgroup analyses. Although differences in trial designs and definitions should be considered when interpreting the overall results, mortality is a hard end point not affected by study definitions. Duration and dose of study drugs also differed between trials, as did the use of concomitant treatments and types of revascularisation. Of note, many of the non-randomised studies were not pure head to head comparisons of the two anticoagulants. However, the absence of heterogeneity in the overall analysis and the sensitivity and subgroup analyses all showing consistent reductions in mortality, suggest that the results are robust. Regarding safety, it seems that the intravenous route for administering enoxaparin drives the reduction in major bleeding, confirming previous information from randomised trials.11
Duration of anticoagulation is a possible confounder for this problem, although this information is usually not available in published data.
Meaning of the study and implications for policymakers
The profound reduction in mortality found in this meta-analysis could be explained by the additional reductions of serious ischaemic events and major bleeding. The global reduction in ischaemic events and mortality was driven by the major effect observed in the setting of primary percutaneous coronary intervention for ST elevation myocardial infarction and is in line with the results of the ATOLL randomised trial.14
This effect was obtained from anticoagulation using intravenous enoxaparin and the favourable pharmacodynamic profile of the 0.5 mg/kg dose. A similar benefit on mortality has been seen recently in other studies of ST elevation myocardial infarction using bivalirudin56
or radial access.68
In lower risk populations the same interventions did not reduce mortality.1
The results of this meta-analysis should influence the next guidelines dealing with anticoagulation in percutaneous coronary intervention or in ST elevation myocardial infarction. The superiority of enoxaparin over unfractionated heparin is now well documented in the setting of percutaneous coronary intervention, by randomised controlled trials, registry based studies, and this meta-analysis, building the case for an update of current guidelines on anticoagulation. This is particularly true for primary percutaneous coronary intervention, where the benefit is most striking.
Unanswered questions and future research
Two new anticoagulants (bivalirudin and enoxaparin) have proved to be superior to unfractionated heparin during percutaneous coronary intervention, particularly in patients with ST elevation myocardial infarction undergoing primary percutaneous coronary intervention. A head to head comparison between intravenous enoxaparin and intravenous bivalirudin is needed in the setting of primary percutaneous coronary intervention using contemporary techniques (radial access, last generation of stent, and thromboaspiration) and new antiplatelet agents such as prasugrel or ticagrelor.
During percutaneous coronary intervention, enoxaparin seems to be superior to unfractionated heparin in reducing all cause mortality and ischaemic and bleeding end points. This superiority was particularly evident in patients with ST elevation myocardial infarction undergoing primary percutaneous coronary intervention. Given the relatively low cost of enoxaparin (and its wide availability), this seems to be an attractive strategy to improve outcomes in the large number of patients undergoing percutaneous coronary intervention worldwide.
What is already known on this topic
- Enoxaparin has a more stable and predictable anticoagulant effect than unfractionated heparin, which requires tight monitoring and dose adjustment
- In randomised studies, intravenous enoxaparin 0.5 mg/kg was superior to unfractionated heparin in elective percutaneous coronary intervention (reduction of bleeding) and primary percutaneous coronary intervention (reduction of ischaemic events)
What this study adds
- Enoxaparin use during percutaneous coronary intervention reduced mortality by 34% (absolute risk reduction 1.66%) compared with unfractionated heparin
- The mortality benefit was particularly noticeable in patients with ST elevation myocardial infarction undergoing primary percutaneous coronary intervention
- This survival benefit is supported by concomitant reductions in both ischaemic and major bleeding complications