ORIGINAL CONTRIBUTIONS

Bivalirudin versus Heparin Plus Glycoprotein IIb/IIIa Inhibitors in Drug-Eluting Stent Implantations in the Absence of Acute Myo

*†‡Gregory J. Mishkel, MD, †Anna L. Moore, MPH, ‡Stephen J. Markwell, MA, *Robert W. Ligon, MS
*†‡Gregory J. Mishkel, MD, †Anna L. Moore, MPH, ‡Stephen J. Markwell, MA, *Robert W. Ligon, MS
Bare-metal stents (BMS) lower the risk of restenosis and the need for repeat coronary revascularization compared to balloon angioplasty alone.1 The introduction of drug-eluting stents (DES) has further markedly decreased angiographic restenosis rates to approximately 3–4% at 6 months.2–4 DES were developed to reduce the risk of major adverse cardiac events (MACE) after percutaneous coronary interventions (PCI) by inhibiting neointimal proliferation and reducing the risk of restenosis. Bivalirudin is a direct thrombin inhibitor that has several advantages over heparin including inhibition of circulating and clot-bound thrombin, no activation of platelets, a predictable dose response, a short half-life and no risk of drug-induced thrombocytopenia.5 In the REPLACE-2 randomized, controlled trial, bivalirudin was as effective as heparin plus glycoprotein (GP) IIb/IIIa inhibition, though was associated with significantly fewer bleeding events.6 Gurm and colleagues confirmed the results of REPLACE-2 in a “real-world” tertiary care setting.7 However, the REPLACE-2 study was conducted before DES were available, and the study by Gurm et al included only the first 8 months of DES availability. The use of bivalirudin in DES implantations has been studied and recent published reports suggest that in this setting it is safe, effective and associated with fewer vascular complications than heparin.8-10 Furthermore, compared with heparin plus GP IIb/IIIa inhibition, treatment with bivalirudin results in a shorter length of stay and lower hospital costs.9 Our objective was to contribute information from a high-volume tertiary care cardiac center on the administration, clinical outcomes and costs associated with bivalirudin versus heparin + GP IIb/IIIa inhibitors in a population of patients who underwent DES implantion. Patient Population and Methods We retrospectively reviewed the records of all patients undergoing DES interventions in our tertiary care hospital between May 1, 2003 and December 31, 2004. We excluded patients who: (a) received a BMS in the same procedure; (b) underwent PCI for a vein graft lesion; or (c) presented with myocardial infarction (MI) with or without ST-segment elevation. Patients who received a GP IIb/IIIa inhibitor and bivalirudin concomitantly were identified, though were not included in this analysis because: (a) the size of this subgroup (n = 71) was disproportionately smaller than the other subgroups, precluding its inclusion in the logistic regression used to generate propensity scores; and (b) “bailout” procedures could not be reliably distinguished from planned treatment with GP IIb/IIIa + bivalirudin. The administration of bivalirudin and/or heparin + GP IIb/IIIa inhibitor and pretreatment with clopidogrel was at the discretion of each of the 18 operators who performed the index procedures. Among these 18 operators, 4 performed less than 10 procedures, 8 performed between 10 and 100, and 6 performed more than 100 procedures. Since this was an observational study, and since no attempt was made to control the use of certain types of stents or specific antithrombotic regimens, the use of sirolimus- versus paclitaxel-eluting stents, and of heparin + GP IIb/IIIa inhibitors versus bivalirudin varied widely among operators. Bivalirudin was administered as a 0.75 mg/kg bolus followed by an infusion of 1.75 mg/kg/hour for the duration of the procedure. Heparin was administered as an initial bolus of 30–70 units/kg to reach a target activated clotting time (ACT) of 200–250 seconds. Our Ethics Review Board approved the study protocol, which complied with our institutional guidelines and those of the American Physiological Society. Data Collection and Analysis The treatment, procedural and outcome data were obtained from our facility’s American College of Cardiology National Cardiovascular Data Registry (ACC-NCDR) dataset. Angiographic and follow-up data were obtained from a dedicated interventional database that is maintained at our facility. Clinical follow ups were routinely collected on patients who underwent stent implantation at 6 months, 1 year and annually thereafter, via telephone contact, mailed surveys, medical records and/or death certificate review. Economic data were obtained from our hospital’s accounting system (McKesson Trendstar, Alpharetta, Georgia) and represent actual costs. Vascular complication was defined as: (a) bleeding from the vascular access site requiring a blood transfusion and/or prolonging the hospital stay, and/or a >3.0 g/dl decrease in hemoglobin concentration; (b) total arterial occlusion by thrombus requiring surgical repair; (c) loss of distal pulse; or (d) dissection or development of pseudoaneurysm or arteriovenous fistula at the site of percutaneous vascular access. Postprocedural renal failure was defined as an increase in serum creatinine to >2.0 mg/dl, a ?50% increase over an abnormal baseline value, or new requirement of dialysis. Stent thrombosis was defined as an intraluminal filling defect with contrast staining on 3 sides, representing total or partial stent occlusion present at the time of a clinically-driven repeat angiography. All stent thromboses were confirmed by a review of the angiogram and the procedural report. Target vessel revascularization (TVR) was defined as a reintervention on the stented segment for chest pain or >70% stenosis on follow-up angiography. MI was defined as the occurrence of chest pain accompanied by new electrocardiographic changes consistent with myocardial ischemia and a rise in CK-MB enzyme to >3 times the upper limit of normal. MACE was a composite endpoint including death, MI or TVR. Statistical analysis. Chi-square tests of independence, or Fisher’s exact test as appropriate, and independent groups t-tests were used to compare the groups on baseline characteristics. Logistic regression was used to generate propensity scores for patients included in the planned heparin + GP IIb/IIIa inhibitor alone and the planned bivalirudin alone groups. The propensity scores indicated the predicted probability of being treated with heparin + GP IIb/IIIa inhibitor or bivalirudin alone based on all of the covariates listed in Table 1, plus body mass index, payor, physician operator and number of stents per lesion. Chi-square tests of independence or Fisher's exact test were used, as appropriate, to perform the unadjusted comparisons between the groups. Logistic regression was used to adjust these comparisons for the propensity score. Independent groups t-tests were used to assess group differences on the economic outcomes of length of stay and cost. Analysis of covariance was performed to adjust the comparisons for propensity score. One-year event rates were estimated using the Kaplan-Meier method. The Cox proportional hazards model was used to assess group differences on 1-year clinical outcomes of death, TVR and MACE. Propensity scores were included in the Cox regressions for the adjusted comparisons. Differences were considered statistically significant for p-values Results Our analysis included 1,842 patients, of whom 1,305 (68.2%) received heparin + GP IIb/IIIa inhibitor, and 537 (28.1%) received bivalirudin alone. Femoral vascular access was used in >99% of the procedures. Clinical follow up was available for 1,813 patients (98.4%), and the average follow up was 782 ± 204 days. The demographic, clinical and procedural characteristics of the study groups are shown in Table 1. The patients in the bivalirudin-treated group had a higher prevalence of current smoking and histories of congestive heart failure, peripheral vascular disease, cerebrovascular disease, MI, PCI and coronary artery bypass graft surgery, and had a greater mean number of diseased vessels. The heparin + GP IIb/IIIa-treated group, however, included a greater percentage of patients who presented with unstable angina. Nearly all patients received aspirin (99.6%) and clopidogrel (98.8%). All patients received at least 1 sirolimus- or paclitaxel-eluting stent, and 11 patients received both types of stents in the same procedure. Although we do not have an explanation for this marked difference, bivalirudin was used in association with approximately one-fourth of sirolimus-eluting stents, versus nearly two-thirds of paclitaxel-eluting stents. The rates of in-hospital adverse events were similarly low in both treatment groups (Table 2), except for a significantly higher rate of vascular complications in the heparin + GP IIb/IIIa inhibitor-treated group after adjustment by the propensity score (p = 0.04). A single patient had a hemorrhage in the bivalirudin-treated group, in contrast with 11 patients with vascular hemorrhages, 4 patients with dissections and 4 patients with pseudoaneurysms in the heparin + GP IIb/IIIa inhibitor-treated group. The 1-year MACE-free survival rates were similar in both groups. The mean 1-year mortality estimates were 3.3%, with a 95% confidence interval (CI) of 2.4–4.4% in the heparin + GP IIb/IIIa inhibitor-treated group versus 5.5% (95% CI: 3.8–7.8%) in the bivalirudin-treated group. Corresponding estimates for TVR and MACE were 3.3% (95% CI: 2.5–4.4%) versus 3.2% (95% CI: 2.0–5.0%) and 8.0% (95% CI: 6.6–9.6%) versus 9.2% (95% CI: 7.0–12.1%), respectively. These differences were all statistically nonsignificant. A total of 8 patients (0.4%) presented to our catheterization laboratory with clinical manifestations consistent with stent thromboses during follow up. They occurred at 3, 28, 202, 470 and 557 days, respectively, in 5 patients from the heparin + GP IIb/IIIa inhibitor group, and at 5, 2 and 161 days, respectively, in 3 patients who received bivalirudin. The incidence of angiographically-confirmed thrombosis was similar in both groups. Using a Cox regression analysis, hazard ratios were calculated for patients who received bivalirudin alone with respect to those who received heparin + GP IIb/IIIa inhibitor for the long-term endpoints of death, nonfatal MI, TVR and MACE. Treatment with bivalirudin was not associated with an increased risk of adverse events. Similar results were obtained after propensity adjustment in both treatment groups (Table 3). The unadjusted and propensity-adjusted overall cost analyses showed no significant difference between the two treatment groups with respect to hospital cost and length of stay (Table 4). Furthermore, after subdivision of the heparin + GP IIb/IIIa inhibitor group according to drug type, there were no significant differences between the heparin + eptifibatide and the bivalirudin-alone groups (Table 4). The hospital costs per patient incurred in the heparin + abciximab group ($12,105 unadjusted and $12,379 propensity-adjusted) were significantly higher than those incurred in the bivalirudin group ($11,018 unadjusted and $11,086 propensity-adjusted; p = 0.05 for both comparisons). The mean costs per patient of abciximab, eptifibatide and bivalirudin were $1,241 ± 435, $553 ± 195 and $374 ± 142, respectively. Discussion Main findings of our study. This study suggests that in patients undergoing routine PCI for implantation of a DES, bivalirudin is as effective as heparin + GP IIb/IIIa inhibitor, and potentially associated with a lower in-hospital vascular complication rate. In addition, the use of bivalirudin was associated with lower hospital costs than the use of abciximab as the GP IIb/IIIa inhibitor. These observations are concordant with the results of previous studies of bivalirudin in similar clinical settings.14 Safety issues with DES are important, given the concerns regarding a possible increased thrombogenicity raised by reports of stent thrombosis soon after the approval of sirolimus-eluting stents for commercial use.11 While firm evidence for increased rates of stent thrombosis with DES compared to BMS has not been provided,11 the issue of the safety of DES implantation has been brought to the forefront. Therefore, while the REPLACE-2 trial demonstrated the safety of bivalirudin with BMS, whether it can safely replace heparin as the procedural antithrombotic during DES implantation remains unknown. The observation of a significantly lower vascular complication rate after risk adjustment among bivalirudin-treated patients suggests that, had the treatment groups been randomized with a balanced distribution of risk factors, a significant difference between the groups may have been present. All other rates of adverse events were similar in unadjusted and adjusted analyses, indicating that there was no disadvantage in the use of bivalirudin as compared to heparin + GP IIb/IIIa inhibitor. Bivalirudin also appeared cost-effective in PCI with DES implants. Rha et al reported a significantly shorter mean length of hospital stay associated with bivalirudin compared to heparin,9 as observed in our overall comparison of heparin + GP IIb/IIIa inhibitor with bivalirudin alone, although our findings were not statistically significant. When the heparin + GP IIb/IIIa inhibitor-treated group was subdivided by type of inhibitor, it became apparent that patients treated with abciximab incurred higher hospital costs, both before and after propensity adjustment. It should be noted that this cost comparison was limited to patients who did not present with a MI. Previous studies. Rha and colleagues were first to report on the safety of bivalirudin in PCI with sirolimus-eluting stents.9 In a retrospective comparison of clinical outcomes in 323 bivalirudin-treated versus 352 heparin-treated patients, they found no significant between-groups differences with respect to major in-hospital clinical outcomes, including death, Q-wave MI, coronary artery bypass or repeat angioplasty. The incidence of subacute stent thrombosis was not significantly different between the study groups. It is, however, noteworthy that the 4 cases observed in that study occurred in the heparin-treated group. While the bleeding complications were similar in both groups, overall vascular complications were significantly lower in the bivalirudin-treated group. In the heparin-treated group, 36.7% of patients received concomitant eptifibatide compared with 5.1% in the bivalirudin-treated group. No numbers were provided regarding the use of abciximab or tirofiban, which might have contributed to the vascular complication rate.9 ADEST was the first prospective trial to examine the safety of bivalirudin in 1,182 consecutive patients undergoing PCI with implantation of sirolimus-eluting stents.8 In-hospital MACE occurred in 4.9% of patients, including a 0.3% mortality, 4.0% MI, 0.5% TVR, and 0.3% stent thrombosis. Major bleeding was reported in only 0.8% of patients.8 Chieffo et al reported the results of the ENSEMBLE trial, which examined the safety and efficacy of bivalirudin in 70 patients who underwent implantation of sirolimus-eluting stents and 41 patients who underwent implantation of paclitaxel-eluting stents.10 The rates of in-hospital ischemic events were low, including 1.8% MI, 0.9% TVR and 0.9% stent thrombosis. No patient died, suffered major or minor hemorrhages, required transfusion, or sustained vascular complications.10 The rates of acute and subacute stent thrombosis and overall in-hospital MACE with bivalirudin during DES implantation were similar to the rates reported for sirolimus-eluting stents in the pivotal SIRIUS trial3 (0.4% and 2.4%, respectively) and for paclitaxel-eluting stents in the TAXUS-IV trial12 (0.0% and 2.9%, respectively), in which heparin and GP IIb/IIIa inhibitors were administered. Since DES are more expensive than BMS, the economical impact of this new technology has been evaluated.13 In this perspective, bivalirudin might be an attractive alternative to heparin, as it has been shown to reduce length of stay and hospital costs when compared to heparin + GP IIb/IIIa inhibition.14 This economic advantage was first demonstrated in the REPLACE-2 trial, where the in-hospital and 30-day costs were lowered by $405 and $374 per patient, respectively. Regression modeling revealed that, besides the cost of drugs, the hospital savings were primarily due to lower rates of bleeding complications and of drug-induced thrombocytopenia.14 Study limitations. We acknowledge several limitations in this study. First, it was not randomized, and the antithrombotic treatment administered was left to the discretion of each operator. Since the treatment groups were different with respect to several baseline characteristics, propensity scores were utilized to adjust for the nonrandom treatment assignments. Second, our analyses were limited by the retrospective design of the study. Third, our results cannot be extrapolated to patients presenting with MI, with or without ST-segment elevation, or to patients treated with combined bivalirudin and GP IIb/IIIa inhibitor therapy. GP IIb/IIIa inhibitors are often added to bivalirudin under conditions of “bailout” associated with potentially poor outcomes. While we were able to identify patients who received concomitant GP IIb/IIIa inhibitor + bivalirudin, we could not distinguish those for whom this combined treatment had been planned from those who received it provisionally as a “bailout” strategy. Consequently, these patients could not be included in the comparison of clinical outcomes, since the intention to treat was unknown. Furthermore, we were unable to identify the small number of patients who received unplanned (“bailout”) GP IIb/IIIa use in addition to primary heparin therapy. Fourth, data were not available regarding the duration and dosage of pretreatment with clopidogrel. Fifth, since ACT measurements were not available, it is possible that ACT was systematically higher in the heparin + GP IIb/IIIa inhibitor group, explaining the higher rate of bleeding complications in that group. Despite these limitations, this study contributes information collected in a large patient population and reflects the actual clinical presentation, outcomes and costs of treatment in routine, daily practice. Conclusions This study indicates that bivalirudin is as effective as heparin + GP IIb/IIIa inhibition in patients undergoing routine DES implantation, and that it might be associated with lower rates of vascular complications. In addition, its use was associated with lower in-hospital costs in selected cases. These observations are concordant with the results of previously published studies of bivalirudin and suggest that it is a sound alternative to heparin + GP IIb/IIIa inhibition for DES implantations outside of the acute phase of MI.
References
1. Whitlow P. Drug-eluting stents. J Invasive Cardiol 2004;16(Suppl):2S–6S. 2. Morice MC, Serruys PW, Sousa JE, et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002;346:1773–1780. 3. Moses JW, Leon MB, Popma JJ, et al. SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315–1323. 4. Cohen MG, Ohman EM. Drug-eluting stents in acute myocardial infarction: Is science catching up with practice? JAMA 2005;293:2154–2156. 5. Bates SM, Weitz JI. Direct thrombin inhibitors for treatment of arterial thrombosis: Potential differences between bivalirudin and heparin. Am J Cardiol 1998;82:12P–18P. 6. Lincoff AM, Bittl JA, Harrington RA, et al. 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. 7. Gurm HS, Rajagopal V, Fathi R, et.al. Effectiveness and safety of bivalirudin during percutaneous coronary intervention in a single medical center. Am J Cardiol 2005;95:716–721. 8. Dangas G, Lasic Z, Mehran R, et.al. Effectiveness of the concomitant use of bivalirudin and drug-eluting stents (from the prospective, multicenter BivAlirudin and Drug Eluting STents [ADEST] study). Am J Cardiol 2005;96:659–663. 9. Rha S, Kuchulankanti PK, Pakala R, et al. Bivalirudin versus heparin as an antithrombotic agent in patients treated with a sirolimus-eluting stent. Am J Cardiol 2004;94:1047–1050. 10. Chieffo A, Melzi G, Rogacka R, et al. Safety and feasibility of bivalirudin with either Cypher or TAXUS drug-eluting stent during percutaneous coronary intervention. Eurointervention 2005;1:70–74. 11. FDA Public Health Web Notification: Final update of information for physicians on sub-acute thromboses (SAT) and hypersensitivity reactions with use of the Cordis CYPHER™ sirolimus-eluting coronary stent. Available at: http://www.fda.gov/cdrh/safety/cypher3.html. Accessed September 25, 2006. 12. Stone GW, Ellis SG, Cox DA, et al. for the TAXUS-IV Investigators. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221–231. 13. Kong DF, Eisenstein EL, Sketch MH, et al. Economic impact of drug-eluting stents on hospital systems: A disease-state model. Am Heart J 2005;147:449–456. 14. Cohen DJ, Lincoff AM, Lavelle TA, et al. Economic evaluation of bivalirudin with provisional glycoprotein IIb/IIIa inhibition versus heparin with routine glycoprotein IIb/IIIa inhibition for percutaneous coronary intervention. J Am Coll Cardiol 2004;44:1792–1800.