Abstract
Background
The anticoagulant of choice during percutaneous coronary intervention (PCI) in women is not well established.
Methods
An electronic search was conducted for trials that randomized patients undergoing PCI to bivalirudin versus heparin, and reported outcomes of interest in women. Random effects DerSimonian–Laird risk ratios (RR) were calculated. Main outcome was net adverse clinical events (NACE) at 30-days. Other outcomes included major adverse cardiac events (MACE), all-cause mortality, myocardial infarction (MI), target vessel revascularization (TVR), and major bleeding at 30-days. 1-year all-cause mortality and MACE were also examined.
Results
Nine trials that randomized women undergoing PCI to bivalirudin ( n = 3960) versus heparin ( n = 4050) were included. At 30-days, bivalirudin was associated with reduced risk of NACE (RR = 0.85; 95% CI 0.73–0.98; p = 0.03), mainly driven by reduction in major bleeding (RR = 0.59; 95% CI 0.49–0.71; p < 0.001) compared with heparin. No difference in MACE ( p = 0.92), all-cause mortality ( p = 0.23), MI ( p = 0.86); or TVR ( p = 0.53) was demonstrated between both groups. At 1-year, the risk of MACE and all-cause mortality was similar in both groups. On a subgroup analysis, the benefit associated with bivalirudin appeared to be less evident when Glycoprotein IIb/IIIa inhibitors (GPI) was used as bailout therapy with heparin, however without significant interaction. Furthermore, in STEMI population, no difference in NACE, MACE, or major bleeding was observed between both groups.
Conclusion
In women undergoing PCI, bivalirudin is associated with reduced risk of major bleeding and NACE compared with heparin especially when GPI is routinely used.
Highlights
- •
Bivalirudin is associated with reduced risk of net adverse clinical events and major bleeding at 30 days in women undergoing PCI compared with heparin especially in the setting of routine Glycoprotein IIb/IIIa inhibitors use.
- •
Bivalirudin offers no benefit on long term mortality compared with heparin.
- •
In STEMI patients, the benefit associated with bivalirudin compared to heparin was no longer demonstrated.
1
Introduction
Anticoagulation during percutaneous coronary intervention (PCI) prevents ischemic events by reducing thrombus formation on intravascular equipment, as well as at the site of coronary endothelial disruption resulting from balloon dilation and stent implantation . Heparin with or without glycoprotein IIb/IIIa inhibitors (GPI) has long been the standard anticoagulant during PCI. The direct thrombin inhibitor bivalirudin (Angiomax®, The Medicines Company) became an attractive alternative after studies showed a possible reduction in the risk of major bleeding and net adverse clinical events (NACE) with bivalirudin compared to heparin . This was however challenged by meta-analyses suggesting that the benefit of bivalirudin over heparin is largely related to routine administration of GPI with heparin, as well as the high doses of heparin used .
Women undergoing PCI are at higher risk of bleeding complications and mortality compared with men, thus the weighted risk versus benefit of the type of anticoagulation used is of utmost importance . Unfortunately, only a few studies were designed to assess the anticoagulant of choice during PCI in women . While favorable bleeding outcomes with bivalirudin compared to heparin have been observed in women, the impact of routine versus bailout use of GPI, and clinical presentation on such outcomes has not been investigated. Furthermore, a recent pooled analysis showed a reduced risk of 1-year mortality with bivalirudin compared to heparin in women undergoing PCI, however that analysis was limited to 3 studies . We aim to conduct a more comprehensive meta-analysis including all randomized clinical trials to date, with the objective to assess short and long-term clinical outcomes of bivalirudin versus heparin in women undergoing PCI, and to explore whether routine use of GPI and clinical presentation affect these outcomes.
2
Methods
We conducted this meta-analysis according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines . A systematic electronic search of PubMed, Cochrane Library, Web of Science and EMBASE databases was conducted from inception until January 2017, without language restriction, using the key words “bivalirudin”, “angiomax”, “heparin”, “percutaneous coronary intervention”, “women”, “female”, “sex” and “gender” both separately and in combination. Supplemental Fig. 1 illustrates the search strategy. We also searched the reference lists of all the retrieved articles and previous meta-analyses for additional trials not retrieved from the original search strategy. Furthermore, we performed a comprehensive search using the titles of the retrieved trials for any subgroup or pooled analyses performed. The current meta-analysis was registered with the International Prospective Register for Systematic Reviews or PROSPERO (CRD42016042288).
2.1
Inclusion criteria and data extraction
We included all clinical trials that randomized patients undergoing PCI to bivalirudin versus heparin (with or without GPI), and reported outcomes of interest in women. Data were extracted either from the main publication or from a further subgroup/pooled analysis. Four reviewers extracted data including the design of included trials, population size, patients’ baseline demographics, various treatment strategies, as well as outcomes of interest reported in female gender group. ClinicalTrials.gov registry was checked by the reviewers to ascertain the inclusion of all outcomes that might not be stated in the main publication. Independent reviewers, (M.S and R.N), revised the extracted data, and any discrepancy was resolved by consensus of all the authors. The number of events for clinical outcomes in both arms was tabulated. Data were tabulated separately for outcomes at 30-days and 1-year if reported by the study.
2.2
Risk of bias and quality assessment
The risk of bias among the included studies was assessed using the Cochrane Collaboration’s tool that utilizes 7 points to evaluate each study for selection bias (random sequence generation, and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), reporting bias (selective reporting), and other biases . Trials with <2 high risk criteria were considered to have a low risk of bias, while those with >4 high risk criteria were considered to have a high risk of bias. The GRADE (Grades of Recommendation, Assessment, Development and Evaluation) tool was used to further assess the quality of evidence for each outcome as recommended by the Cochrane Handbook for Systematic Reviews of Interventions.
2.3
Outcomes
The main outcome of this meta-analysis was the net adverse clinical events (NACE) at 30-days. NACE was defined as the composite of ischemia and major bleeding. Other outcomes of interest evaluated in our study included major adverse cardiac events (MACE), all-cause mortality, myocardial infarction (MI), target vessel revascularization (TVR), and major bleeding at 30-days. 1-year all-cause mortality and MACE were also examined. The definitions of outcomes in the included trials are detailed in Supplemental Table 1.
2.4
Statistical analysis
We analyzed the outcomes reported by an intention-to-treat analysis. The weighted mean follow-up duration of each outcome was calculated, and the sample size was used as the weight. The mean difference for continuous variables was calculated using simple t-test. As the event rates of all the outcomes were more than 1%, we performed random-effects summary risk ratios (RR) using a DerSimonian and Laird model as recommended by the Cochrane Handbook for Systemic Reviews of Intervention. p -Values were 2-tailed, and p < 0.05 was considered statistically significant. Confidence intervals (CIs) were calculated at the 95% level for the overall estimates effect. Statistical heterogeneity was evaluated using I 2 statistic value with values <25% indicating low heterogeneity, values between 25% and 50% indicating moderate heterogeneity, and values >50% indicating high heterogeneity . Publication bias was calculated using Egger method . All analyses were performed using STATA software version 14 (StataCorp, College Station, Texas).
2.5
Sensitivity and subgroup analyses
A sensitivity analysis was performed for NACE after excluding the BRIGHT trial since it was found to be an outlier, to evaluate its impact on this outcome. To evaluate the effect of GPI use as an adjunctive therapy to heparin on the outcomes of NACE and major bleeding, subgroup analyses were performed based on whether GPI was routinely used with heparin versus left to the physicians’ discretion (i.e., bailout therapy). Random-effects analysis was performed to calculate the p value for interaction in subgroup analyses, and was considered to be significant if p < 0.05. Furthermore, subgroup analyses were performed including the trials enrolling only acute coronary syndrome (ACS) patients, and then another analysis for trials enrolling only ST-segment elevation myocardial infarction (STEMI) patients, to evaluate whether the clinical presentation in women will affect the outcomes of NACE and major bleeding with bivalirudin versus heparin.
2
Methods
We conducted this meta-analysis according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines . A systematic electronic search of PubMed, Cochrane Library, Web of Science and EMBASE databases was conducted from inception until January 2017, without language restriction, using the key words “bivalirudin”, “angiomax”, “heparin”, “percutaneous coronary intervention”, “women”, “female”, “sex” and “gender” both separately and in combination. Supplemental Fig. 1 illustrates the search strategy. We also searched the reference lists of all the retrieved articles and previous meta-analyses for additional trials not retrieved from the original search strategy. Furthermore, we performed a comprehensive search using the titles of the retrieved trials for any subgroup or pooled analyses performed. The current meta-analysis was registered with the International Prospective Register for Systematic Reviews or PROSPERO (CRD42016042288).
2.1
Inclusion criteria and data extraction
We included all clinical trials that randomized patients undergoing PCI to bivalirudin versus heparin (with or without GPI), and reported outcomes of interest in women. Data were extracted either from the main publication or from a further subgroup/pooled analysis. Four reviewers extracted data including the design of included trials, population size, patients’ baseline demographics, various treatment strategies, as well as outcomes of interest reported in female gender group. ClinicalTrials.gov registry was checked by the reviewers to ascertain the inclusion of all outcomes that might not be stated in the main publication. Independent reviewers, (M.S and R.N), revised the extracted data, and any discrepancy was resolved by consensus of all the authors. The number of events for clinical outcomes in both arms was tabulated. Data were tabulated separately for outcomes at 30-days and 1-year if reported by the study.
2.2
Risk of bias and quality assessment
The risk of bias among the included studies was assessed using the Cochrane Collaboration’s tool that utilizes 7 points to evaluate each study for selection bias (random sequence generation, and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), reporting bias (selective reporting), and other biases . Trials with <2 high risk criteria were considered to have a low risk of bias, while those with >4 high risk criteria were considered to have a high risk of bias. The GRADE (Grades of Recommendation, Assessment, Development and Evaluation) tool was used to further assess the quality of evidence for each outcome as recommended by the Cochrane Handbook for Systematic Reviews of Interventions.
2.3
Outcomes
The main outcome of this meta-analysis was the net adverse clinical events (NACE) at 30-days. NACE was defined as the composite of ischemia and major bleeding. Other outcomes of interest evaluated in our study included major adverse cardiac events (MACE), all-cause mortality, myocardial infarction (MI), target vessel revascularization (TVR), and major bleeding at 30-days. 1-year all-cause mortality and MACE were also examined. The definitions of outcomes in the included trials are detailed in Supplemental Table 1.
2.4
Statistical analysis
We analyzed the outcomes reported by an intention-to-treat analysis. The weighted mean follow-up duration of each outcome was calculated, and the sample size was used as the weight. The mean difference for continuous variables was calculated using simple t-test. As the event rates of all the outcomes were more than 1%, we performed random-effects summary risk ratios (RR) using a DerSimonian and Laird model as recommended by the Cochrane Handbook for Systemic Reviews of Intervention. p -Values were 2-tailed, and p < 0.05 was considered statistically significant. Confidence intervals (CIs) were calculated at the 95% level for the overall estimates effect. Statistical heterogeneity was evaluated using I 2 statistic value with values <25% indicating low heterogeneity, values between 25% and 50% indicating moderate heterogeneity, and values >50% indicating high heterogeneity . Publication bias was calculated using Egger method . All analyses were performed using STATA software version 14 (StataCorp, College Station, Texas).
2.5
Sensitivity and subgroup analyses
A sensitivity analysis was performed for NACE after excluding the BRIGHT trial since it was found to be an outlier, to evaluate its impact on this outcome. To evaluate the effect of GPI use as an adjunctive therapy to heparin on the outcomes of NACE and major bleeding, subgroup analyses were performed based on whether GPI was routinely used with heparin versus left to the physicians’ discretion (i.e., bailout therapy). Random-effects analysis was performed to calculate the p value for interaction in subgroup analyses, and was considered to be significant if p < 0.05. Furthermore, subgroup analyses were performed including the trials enrolling only acute coronary syndrome (ACS) patients, and then another analysis for trials enrolling only ST-segment elevation myocardial infarction (STEMI) patients, to evaluate whether the clinical presentation in women will affect the outcomes of NACE and major bleeding with bivalirudin versus heparin.
3
Results
Nine clinical trials with a total of 8010 women met our inclusion criteria. The included trials randomized PCI patients to bivalirudin versus heparin (with or without GPI) and reported outcomes of interest in a subgroup analysis of female gender, with a total of 3960 women in the bivalirudin arm and 4050 women in the heparin arm. Seven trials enrolled only ACS patients, while 2 trials included both ACS and elective PCI patients. Three out of the 7 ACS trials exclusively included patients with ST-segment elevation myocardial infarction (STEMI) . Routine use of GPI with heparin was performed in 4 trials (99% of the patients received heparin plus GPI), while the other 5 trials used GPI as a bailout therapy (only 30% of the patients received heparin plus GPI). Baseline population characteristics of included trials are summarized in Table 1 .
| Study | Year | Patient population | Patients, n a | Treatment strategies | GPI use in heparin arm, % | Primary outcome |
|---|---|---|---|---|---|---|
| BRIGHT | 2015 | STEMI | 127/265 | Bivalirudin vs UFH alone or plus tirofiban | 50 | Death/MI/TVR, stroke or bleeding |
| MATRIX | 2015 | STEMI, NSTEMI and UA | 879/839 | Bivalirudin vs UFH | 0.2 | Death/MI or stroke and net adverse events (MACE + major bleeding) |
| BRAVE-4 | 2014 | STEMI | 65/58 | Bivalirudin plus prasugrel vs. UFH plus clopidogrel | 6.1 | Death/MI/unplanned TVR/definite in stent thrombosis, stroke or major bleeding |
| EUROMAX | 2013 | STEMI | 275/248 | Bivalirudin vs UFH or LMWH plus optional GPI | 69.1 | Death and non-CABG related major bleeding |
| ISAR-REACT 4 | 2011 | NSTEMI | 199/200 | Bivalirudin vs. UFH (70 μ/kg bolus) plus abciximab | 99.6 | Death/MI/urgent TVR, and major bleeding |
| HORIZONS-AMI | 2008 | STEMI | 412/430 | Bivalirudin vs UFH plus GPI | 97.7 | Major bleeding and combined adverse events (death/MI/TVR/stroke and major bleeding) |
| ISAR-REACT 3 | 2008 | Elective PCI or UA | 545/530 | Bivalirudin vs UFH 140 μ/kg bolus | 0.2 | Death/MI/urgent TVR, and in-hospital bleeding |
| ACUITY | 2006 | NSTEMI or UA | 700/701 | Bivalirudin vs UFH or LMWH plus GPI | 96.6 | Death/MI/unplanned TVR, and major bleeding |
| REPLACE-2 | 2003 | Elective PCI, UA or MI >7 days old | 758/779 | Bivalirudin vs UFH (bolus 65 IU/kg) and GPI | 96.5 | Death/MI/urgent TVR and in-hospital bleeding |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
