Prognostic Modeling of Individual Patient Risk and Mortality Impact of Ischemic and Hemorrhagic Complications
Assessment From the Acute Catheterization and Urgent Intervention Triage Strategy Trial
Background— Both ischemic and hemorrhagic complications increase mortality rate in acute coronary syndromes. Their frequency and relative importance vary according to individual patient risk profiles. We sought to develop prognostic models for the risk of myocardial infarction (MI) and major bleeding to assess their impact on risk of death and to examine the manner in which alternative antithrombotic regimens affect these risks in individual patients.
Methods and Results— The Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial randomized 13 819 patients with acute coronary syndrome to heparin plus a glycoprotein IIb/IIIa inhibitor, bivalirudin plus a glycoprotein IIb/IIIa inhibitor, or bivalirudin alone. By logistic regression, there were 5 independent predictors of MI within 30 days (n=705; 5.1%) and 8 independent predictors of major bleeding (n=645; 4.7%), only 2 of which were common to both event types. In a covariate-adjusted, time-updated Cox regression model, both MI and major bleeding significantly affected subsequent mortality rate (hazard ratios, 2.7 and 2.9, respectively; both P<0.001). Treatment with bivalirudin versus heparin plus a glycoprotein IIb/IIIa inhibitor was associated with a nonsignificant 8% increase in MI and a highly significant 50% decrease in major bleeding. Given the individual patient risk profiles and the fact that bivalirudin prevented ≈6 major bleeds for each MI that might occur from its use, the estimated reduction in bleeding was greater than the estimated increase in MI by bivalirudin alone rather than heparin plus a glycoprotein IIb/IIIa inhibitor for nearly all patients.
Conclusions— Consideration of the individual patient risk profile for MI and major bleeding and the relative treatment effects of alternative pharmacotherapies permits personalized decision making to optimize therapy of patients with acute coronary syndrome.
Clinical Trial Registration— clinicaltrials.gov Identifier: NCT00093158.
Received May 7, 2009; accepted October 2, 2009.
A growing body of evidence has demonstrated that risk of death in patients with acute coronary syndrome (ACS) and in those undergoing percutaneous coronary intervention (PCI) is affected not only by recurrent ischemic events but also by major bleeding.1–3 In an analysis of 34 146 patients with non–ST-segment–elevation myocardial infarction (MI) enrolled in the Organization to Assess Ischemic Syndromes (OASIS) and The Clopidogrel in Unstable Angina to Prevent Recurrent Events (CURE) studies, major bleeding was independently associated with increased mortality rate at 30 days and 6 months.1 In a pooled analysis from 4 placebo-controlled randomized trials, patients developing an MI or major bleeding within 30 days of PCI had comparable rates of death at 1 year (11.3% versus 12.2%, respectively), both of which significantly increased compared with patients without such complications.4 Similarly, in the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial, the relative impact of MI and major bleeding on subsequent mortality rate was comparable regardless of antithrombotic strategy.5
Editorial see p 5
Clinical Perspective on p 51
Although both MI and major bleeding are strongly related to mortality rate in patients with ACS and in those undergoing PCI, the risk factors for the development of these 2 complications may vary, as may the relative effectiveness of therapeutic alternatives to suppress these events. Identifying patients who are at relatively higher risk of ischemic versus hemorrhagic complications (or vice versa), coupled with the characterization of the relative treatment effects of alternative adjunctive pharmacological agents, may thus allow personalized decision making to optimize therapies and subsequent clinical outcomes for individual patients. We therefore sought to develop prognostic models for the risk of MI and major bleeding in patients with ACS, to assess their impact on mortality rate, and to examine the manner in which alternative antithrombotic treatment strategies affect these risks in individual patients.
Patient Population and Study Protocol
To develop prognostic risk models in patients with ACS undergoing an early invasive strategy, we utilized the detailed ACUITY trial database, in which 13 819 moderate- and high-risk patients were enrolled. The ACUITY study protocol, inclusion and exclusion criteria, and principal results have been reported previously.5–7 In brief, eligible patients were randomized to the open-label use of 1 of 3 antithrombin regimens begun before angiography: a heparin (either unfractionated heparin or enoxaparin at site discretion) plus a glycoprotein (GP) IIb/IIIa inhibitor, bivalirudin plus a GPIIb/IIIa inhibitor, or bivalirudin alone. Patients assigned to a GPIIb/IIIa inhibitor were randomized again in a 2×2 factorial design to routine upstream GPIIb/IIIa inhibitor initiation with eptifibatide or tirofiban in all patients immediately after randomization versus deferred selective GPIIb/IIIa inhibitor initiation with eptifibatide or abciximab in the catheterization laboratory only for patients undergoing immediate PCI.
Coronary angiography was performed within 72 hours after randomization, with subsequent treatment by PCI, coronary artery bypass grafting (CABG) surgery, or conservative medical care without revascularization. All patients with CAD received dual antiplatelet therapy as described previously.5–7
Clinical End Points
The ACUITY trial was powered for 3 primary 30-day end points: composite ischemia, defined as death from any cause, MI, or unplanned revascularization for ischemia; major bleeding not related to CABG; and net adverse clinical events (composite ischemia or major bleeding). The specific definition of MI depended on the presence or absence of baseline cardiac biomarker elevation, its time of development, and its association with PCI, CABG, or medical treatment.6 Non-CABG major bleeding was defined as intracranial or intraocular bleeding, access site hemorrhage requiring intervention, ≥5-cm diameter hematoma, reduction in hemoglobin ≥4 g/dL without or ≥3 g/dL with an overt bleeding source, reoperation for bleeding, or blood product transfusion.6 Anemia was defined with the use of World Health Organization criteria as baseline hematocrit <39% for men and <36% for women.8 Chronic renal insufficiency was defined as an estimated creatinine clearance <60 mL/min with the use of the Cockcroft-Gault equation.9
As described previously, from 37 candidate variables (including baseline demographics, medical history, cardiovascular risk factors, cardiac biomarkers, laboratory measures, ECG findings, and randomization medication assignment), 2 forward stepwise logistic regression models were used to identify those variables that were independent predictors of MI and non-CABG major bleeding within 30 days; P<0.01 was the criterion for inclusion in each final model.10
To investigate the relationships between MI and major bleeding with the subsequent occurrence of death, each adverse event was entered into a time-updated Cox model11 adjusted for the 13 previously identified baseline predictors of death (listed in Table 1).10 To further estimate the time-dependent risk of major bleeding and MI on mortality rate, additional Cox models were developed with different time-updated binary covariates for discrete time intervals (ie, days 0 to 1, days 2 to 7, days 8 to 30, and days ≥31 after the event). To estimate the numbers of deaths attributable to MI and major bleeding, respectively,12 the adjusted hazard ratios (HRs) for each interval after the event were then applied to the actual number of deaths in that interval (ie, the number attributed=number of deaths among patients with the attribute [MI or bleeding]× [HR−1]/HR). All analyses were performed with the use of STATA version 9.2. All significance levels are 2 sided.
Predictors of MI and Major Bleeding
Of the 13 819 ACUITY patients, 645 (4.7%) and 705 (5.1%) developed an MI and a non-CABG major bleed, respectively, within 30 days of randomization. Table 1 details the univariate prognostic associations of the 13 baseline variables (including treatment assignment) that were shown previously10 to be independent predictors for the development of MI or non-CABG major bleeding within 30 days (Table 1).
By logistic regression, the occurrence of a new MI within 30 days was independently predicted by 5 variables, including (in order of statistical significance) elevated baseline cardiac biomarkers, family history of coronary artery disease, older age, ST-segment deviation ≥1 mm on the baseline ECG, and previous MI (Table 2). Treatment assignment was not significantly related to the risk of MI, although there was a nonsignificant trend for an 8% increase in MI risk in patients treated with bivalirudin alone rather than heparin plus a GPIIb/IIIa inhibitor. Figure 1 (top) shows the distribution of the individual 30-day risk of MI for all 13 819 patients. Categorizing patients into 3 equally sized groups of risk for 30-day MI revealed a modest gradient in MI incidence from 3.3% in the lowest risk tertile to 7.3% in the highest risk tertile. At any level of risk, there was no discernible effect of randomized treatment on the risk of MI (Table 3).
The occurrence of non-CABG major bleeding within 30 days was independently predicted by 8 variables, including (in order of statistical significance) female sex, baseline anemia, older age, randomization to heparin plus a GPIIb/IIIa inhibitor rather than bivalirudin alone, elevated baseline serum creatinine, elevated baseline white blood cell count, no previous PCI, previous cerebrovascular accident, baseline ST-segment deviation ≥1 mm, and randomization to heparin plus upstream routine rather than deferred selective GPIIb/IIIa inhibitors (Table 2). Figure 1 (bottom) shows the distribution of the individual risk of non-CABG major bleeding for all 13 819 patients, calculated from the described model assuming treatment with heparin plus upstream GPIIb/IIIa inhibitors. Categorization of patients into 3 equally sized groups for risk of 30-day non-CABG major bleeding (Table 3) shows that risk of major bleeding increased markedly from the lowest to highest third, independently of randomized treatment. For each tertile of risk, treatment with bivalirudin alone rather than either heparin or bivalirudin plus a GPIIb/IIIa inhibitor significantly reduced bleeding risk, with the greatest absolute benefit for patients in the highest risk tertile.
Impact of MI and Major Bleeding on Subsequent Mortality Rate
When introduced as time-updated covariates into the Cox model along with other baseline variables, both the occurrence of MI and non-CABG major bleeding within 30 days after presentation with an ACS were independent predictors of subsequent death within 1 year, with HRs of 2.7 and 2.9, respectively (both P<0.0001) (Table 4). The increased risk of death after MI and non-CABG major bleeding was consistent across all planned treatments (PCI, CABG, or medical treatment). From these models, one can estimate that of the 524 deaths within 1 year in ACUITY, 47.8 (9.1%) were attributable to an MI, and 61.2 (11.7%) were attributable to a non-CABG major bleed.
As shown in Table 4, the temporal impact of the effect of an MI versus non-CABG major bleeding on subsequent death differed. An MI increased the likelihood of death 15.6 times within the first day after its occurrence, but then its prognostic impact declined rapidly such that the risk of dying >30 days after an MI was not elevated significantly. In contrast, the likelihood of dying after a non-CABG major bleed was 4-fold increased within the first 30 days of the event, and the risk remained significant (2.2-fold increased mortality rate) beyond 30 days after the bleed.
Among 1647 patients with ACS triaged to CABG, 607 patients (37%) experienced a CABG-related major bleed. Death subsequently occurred in 57 patients with and in 67 patients without a CABG-related major bleed. In contrast to the increased risk from non-CABG major bleeding, CABG-related major bleeding was not an independent predictor of subsequent death within 1 year (HR, 1.01; 95% confidence interval, 0.70 to 1.46; P=0.96).
Risk of MI Versus Major Bleeding in the Individual Patient
Based on the multivariable models in Table 2, Figure 2 (top) depicts the individual predicted risks of MI and non-CABG major bleeding for all 13 819 patients in the ACUITY trial, assuming that each patient was administered heparin plus a GPIIb/IIIa inhibitor. Approximately half of the study population are above the 45° line of equal risk, and hence their predicted risk of major bleeding is higher than their predicted risk of MI. The bottom panel of Figure 2 similarly depicts the predicted risks of both an MI and a non-CABG major bleed, assuming that all patients were administered bivalirudin alone. As a result of this change in treatment, the population distribution shifts slightly to the right (more MIs) and markedly lower (fewer major bleeds).
Assuming conservatively that treatment with bivalirudin alone rather than heparin plus a GPIIb/IIIa inhibitor is associated with a true 8% increase in the odds of MI (even though in the multivariable model this risk was nonsignificant), given the highly significant 50% reduction in the odds of developing a non-CABG major bleed with bivalirudin (prevention of ≈6 major bleeds for each additional MI that might occur), one can then estimate for each individual patient the absolute reduction in bleeding risk and the absolute increase in MI risk by treating with bivalirudin alone instead of heparin plus a GPIIb/IIIa inhibitor. For the overwhelming majority of patients, the estimated reduction in bleeding risk is greater than the estimated increase in MI risk (Figure 3).
The main findings of the present analysis from the ACUITY trial are that in moderate- and high-risk patients with ACS treated with contemporary antithrombotic strategies before an early invasive management strategy, the following occurred: (1) The risk factors for the development of MI and non-CABG major bleeding in individual patients differed substantially, with only baseline ST-segment deviation and older age predicting both; (2) in a covariate-adjusted, time-updated multivariable analysis, treatment with bivalirudin alone rather than heparin plus a GPIIb/IIIa inhibitor was associated with an 8% nonsignificant increase in MI but a highly significant 50% reduction in major bleeding; (3) both the occurrence of MI and non-CABG major bleeding within 30 days (but not major bleeding related to CABG) had a significant and roughly comparable impact on subsequent death through 1 year of follow-up, although with different timing of maximal effect; and (4) when one considers the individual risk profile for MI and non-CABG major bleeding for each patient and the relative risk of these events with alternative antithrombotic strategies, a personalized selection of the optimal pharmacological regimen for the individual patient can be made that would be expected to minimize the occurrence of MI and major bleeding and their impact on subsequent risk of death. In the present trial, given the risk profile of the patients enrolled and the fact that bivalirudin prevents ≈6 non-CABG bleeds for each MI that might occur as a result of its use, for nearly all patients, the estimated reduction in bleeding was greater than the estimated increase in MI by treatment with bivalirudin alone rather than heparin plus a GPIIb/IIIa inhibitor.
In the present study, the occurrence of MI was independently predicted by cardiac biomarker elevation, family history of coronary disease, age, ST-segment deviation ≥1 mm, and prior MI, consistent with prior reports.7,13 Indeed, all of these factors are components of the Thrombolysis in Myocardial Infarction (TIMI) unstable angina risk score.14 Also consistent with previous studies, independent predictors of major bleeding included female sex, age, baseline anemia, and serum creatinine.15–18 The independent association between higher admission white blood cell count, first PCI, history of cerebrovascular accident, ST-segment deviation ≥1 mm, and major bleeding has not been described previously and deserves further study. The observation that the 5 independent baseline predictors of MI and the 8 independent baseline predictors of major bleeding shared only 2 variables in common demonstrates that individual patients with ACS may be identified who are at relatively higher risk of ischemic than hemorrhagic complications or, conversely, greater risk of hemorrhagic than ischemic complications.
The occurrence of both MI and non–CABG-related major bleeding within 30 days significantly increased the subsequent risk of dying within the next year.10 The point estimate for hazard was comparable for non-CABG major bleeding and MI (2.9 versus 2.7, respectively); by multivariable modeling, when the frequency and HR are taken into account, ≈12% and ≈9% of the deaths occurring within 1 year in ACUITY could be attributed to non-CABG major bleeding and MI, respectively. Moreover, temporal variability of the maximal impact of these events on subsequent mortality rate was noted. Specifically, although the occurrence of an MI strongly increased the risk of death within the first days and weeks of the event, patient survival beyond 1 month after the MI was not affected. In contrast, major bleeding had a highly significant impact on mortality rate not only during the first 30 days but between 30 days and 1 year as well (HR, ≈4 and 2.2, respectively; both P<0.001). Similarly, the OASIS and CURE investigators reported that the HR for death from major bleeding (defined somewhat differently than in ACUITY) was 5.4 during the first 30 days (P<0.0001) and 1.5 between 30 days and 6 months (P=0.047).1 The mechanisms through which bleeding affects mortality rate (both early and late after the event) are speculative. Major bleeding is occasionally fatal (eg, intracranial bleeding), or the procedures required to treat severe bleeding may themselves be complicated; bleeding may result in hypotension, ischemia, and arrhythmias19,20; transfusion of banked blood products may impair vasodilatation because of nitric oxide depletion, resulting in red blood cell capillary sludging, and may result in prolonged systemic inflammation and apoptosis.21,22 Functional capillary density, blood flow, and oxygen distribution in microvascular networks are also reduced after stored red blood cell transfusions.23 Potentially lifesaving drugs (eg, aspirin, thienopyridines, β-blockers, and angiotensin-converting enzyme inhibitors) are often discontinued to manage bleeding and are frequently not restarted even after the hemorrhagic event has resolved,24,25 perhaps because of rebleeding propensity. Unmeasured confounders associated with late deaths may be present in patients who bleed, although the outcomes of randomized trials in which agents that decrease bleeding (without other obvious benefit) result in decreased mortality rate suggest that the relationship is at least in part causal.26,27
Given the strong influence of both MI and major bleeding on subsequent risk of death in ACS and PCI, the optimal antithrombotic regimen would effectively suppress ischemic complications while minimizing iatrogenic hemorrhagic risk. However, most agents that reduce ischemia increase bleeding. Given the varying risk factors for MI and non-CABG major bleeding, the optimal antithrombotic regimen for a given individual may vary according to each patient’s relative risk of MI versus bleeding, as well as the relative effectiveness of the alternative pharmacotherapies in suppressing ischemia versus bleeding. In the present analysis, bivalirudin alone compared with heparin plus a GPIIb/IIIa inhibitor was shown to result in 50% reduced odds of a major bleed, an especially desirable attribute in subsets of patients at high propensity to bleed.17 In the highest tertile of bleeding risk in ACUITY, bivalirudin alone compared with heparin plus a GPIIb/IIIa inhibitor reduced non-CABG major bleeding rates from 10.2% to 5.5%. Bivalirudin had a lesser absolute but similar relative effect in patients with a lower risk of bleeding. Conversely, bivalirudin monotherapy was associated with a (nonsignificant) 8% increase in the risk of MI within 30 days. Given this favorable balance between ischemia and bleeding (≈6 major bleeds prevented for each additional MI when bivalirudin is used), the individual risk profiles of the 13 819 patients with ACS in the trial, and the approximately comparable impact of MI and non-CABG bleeding on subsequent morality, the selection of bivalirudin alone would be expected to optimize outcomes for the vast majority of patients, as shown in Figure 3.
Importantly, although the present study suggests that most patients with ACS undergoing an early invasive management strategy would benefit by preferential treatment with bivalirudin alone rather than heparin plus a GPIIb/IIIa inhibitor, modeling of individual patient risk for MI and major bleeding events and the relative treatment effects of alternative pharmacological agents may be useful for personalized decision making in different scenarios. For example, in patients with ACS undergoing PCI in the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel (TRITON)–TIMI 38 trial, treatment with the more potent thienopyridine prasugrel rather than clopidogrel resulted in marked suppression of MI and stent thrombosis but increased rates of total and fatal bleeding.28 As a result, all-cause mortality rate was not significantly different between the 2 agents. The approach described in the present report might identify those ACS patients at relatively high risk for MI and stent thrombosis and/or relatively low risk for hemorrhagic complications in whom use of prasugrel compared with clopidogrel would be expected to be advantageous (and, conversely, those patients likely to experience more harm than benefit with this potent new agent). In accord with recent recommendations on the treatment of patients with ACS,29 our study confirms that anticoagulation decisions should be based on the individual risk of both ischemic and bleeding events.
Given the retrospective nature of this analysis, the exact causal relationships between MI, bleeding, and death cannot be established. Although multivariable modeling was used to adjust for baseline differences, unmeasured confounders may not have been controlled for, and data on the exact triggers for MI and major bleeding are unavailable. The 1-year survival rates between the bivalirudin and heparin plus GPIIb/IIIa inhibitor arms in the main randomized trial were not significantly different,30 most likely as a result of the relatively low rate of death of patients with ACS after invasive management and the fact that MI and major bleeding were collectively responsible for only 109 of 524 deaths (21%) occurring by 1 year. However, the observation that bivalirudin alone compared with heparin plus a GPIIb/IIIa inhibitor reduced deaths among higher-risk patients with ST-segment–elevation MI due to reduced hemorrhagic complications with comparable rates of MI27 demonstrates that the principles outlined in the present analysis are likely valid and would emerge if applied to a larger population with non–ST-segment–elevation MI.
Sources of Funding
The ACUITY trial was sponsored and funded by The Medicines Company, Parsippany, NJ, and Nycomed, Roskilde, Denmark.
Dr Pocock has received research funding and honoraria from The Medicines Company and has served on the Steering Commiittee of Euromax trial. Dr Mehran has received research funding and honoraria from The Medicines Company, Boston Scientific, Cordis, Regado, Therox, sanofi-aventis, Lilly, Diachi Sankyo, Medtronic, Abbott, Guebert, and Abiomed; and has served as a consultant to or on the advisory board of Cordis, Boston Scientific, The Medicines Company, Abbott, Bracco, Medtronic, and Guebert. Dr Nikolsky has received honoraria and lecture fees from Abbott. Dr Lansky has received research funding from The Medicines Company. Dr Bertrand has received consulting fees and lecture fees from Servier and sanofi-aventis. Dr Lincoff has received research funding from The Medicines Company and Bristol Myers Squibb. Dr Moses has received consulting fees from Johnson & Johnson and has served on the speakers’ bureau for Astra Zeneca. Dr Ohman has received research funding from Bristol Myers Squibb, CV Therapeutics, Daiichi Sankyo, Datascope, Eli Lilly & Company, sanofi-aventis, Schering-Plough, Berlex, and The Medicines Company; has served as an expert witness for CV Therapeutics, Inovise, Liposcience, and Response Biomedical; has ownership interest in Inovise; and has served as a consultant to or on the advisory board of Northpoint Domain, Pozen, Inc., The Medicines Company, Savacor, and WebMD. Dr White has received research funding from Alexion, sanofi-aventis, Eli Lilly, The Medicines Company, the NIH, Pfizer, Roche, Johnson & Johnson, Shering Plough, Merck Sharpe & Dohme, AstraZeneca, GlaxoSmithKline, Daichi Sankyo, Neuren Pharmaceuticals, Fournier Laboratories, and Procter & Gamble; and has served as a consultant to or on the advisory board of Regado Biosciences; and has received consulting and lecture fees from sanofi-aventis and The Medicines Company. Dr Stone has received research funding from Boston Scientific, Abbott Vascular, and The Medicines Company. The remaining authors report no disclosures.
Eikelboom JW, Mehta SR, Anand SS, Eikelboom JW, Mehta SR, Anand SS, Xie C, Fox KA, Yusuf S. Adverse impact of bleeding on prognosis in patients with acute coronary syndromes. Circulation. 2006; 114: 774–882.
Rao SV, O'Grady K, Pieper KS, Granger CB, Newby LK, Mahaffey KW, Moliterno DJ, Lincoff AM, Armstrong PW, Van de Werf F, Califf RM, Harrington RA. A comparison of the clinical impact of bleeding measured by two different classifications among patients with acute coronary syndromes. J Am Coll Cardiol. 2006; 47: 809–816.
Ndrepepa G, Berger PB, Mehilli J, Seyfarth M, Neumann FJ, Schömig A, Kastrati A. Periprocedural bleeding and 1-year outcome after percutaneous coronary interventions: appropriateness of including bleeding as a component of a quadruple end point. J Am Coll Cardiol. 2008; 51: 690–697.
Stone GW, Bertrand ME, Moses JW, Ohman EM, Lincoff AM, Ware JH, Pocock SJ, McLaurin BT, Cox DA, Jafar MZ, Chandna H, Hartmann F, Leisch F, Strasser RH, Desaga M, Stuckey TD, Zelman RB, Lieber IH, Cohen DJ, Mehran R, White HD; ACUITY Investigators. Routine upstream initiation vs deferred selective use of glycoprotein IIb/IIIa inhibitors in acute coronary syndromes: the ACUITY timing trial. JAMA. 2007; 297: 591–602.
Stone GW, McLaurin BT, Cox DA, Bertrand ME, Lincoff AM, Moses JW, White HD, Pocock SJ, Ware JH, Feit F, Colombo A, Aylward PE, Cequier AR, Darius H, Desmet W, Ebrahimi R, Hamon M, Rasmussen LH, Rupprecht HJ, Hoekstra J, Mehran R, Ohman EM; ACUITY Investigators. Bivalirudin for patients with acute coronary syndromes. N Engl J Med. 2006; 355: 2203–2216.
Nutritional Anemias: Report of a WHO Scientific Group. Geneva, Switzerland: World Health Organization; 1968.
Mehran R, Pocock SJ, Stone GW, Clayton TC, Dangas GD, Feit F, Manoukian SV, Nikolsky E, Lansky AJ, Kirtane AJ, White HD, Colombo A, Ware JH, Moses JW, Ohman EM. Associations of major bleeding and myocardial infarction with the incidence and timing of mortality in patients presenting with non-ST-elevation acute coronary syndromes: a risk model from the ACUITY trial. Eur Heart J. 2009; 30: 1457–1466.
Eide GE, Heuch I. Average attributable fractions: a coherent theory for apportioning excess risk to individual risk factors and subpopulations. Biomed J. 2006; 48: 820–37.
Boersma E, Pieper KS, Steyerberg EW, Wilcox RG, Chang WC, Lee KL, Akkerhuis KM, Harrington RA, Deckers JW, Armstrong PW, Lincoff AM, Califf RM, Topol EJ, Simoons ML; the PURSUIT Investigators. Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation: results from an international trial of 9461 patients. Circulation. 2000; 101: 2557–2267.
Avezum A, Makdisse M, Spencer F, Gore JM, Fox KA, Montalescot G, Eagle KA, White K, Mehta RH, Knobel E, Collet JP; GRACE Investigators. Impact of age on management and outcome of acute coronary syndrome: observations from the Global Registry of Acute Coronary Events (GRACE). Am Heart J. 2005; 149: 67–73.
Kirtane AJ, Piazza G, Murphy SA, Budiu D, Morrow DA, Cohen DJ, Peterson E, Lakkis N, Herrmann HC, Palabrica TM, Gibson CM; TIMI Study Group. Correlates of bleeding events among moderate- to high-risk patients undergoing percutaneous coronary intervention and treated with eptifibatide: observations from the PROTECT-TIMI-30 trial. J Am Coll Cardiol. 2006; 47: 2374–2379.
Nikolsky E, Mehran R, Dangas G, Fahy M, Na Y, Pocock SJ, Lincoff AM, Stone GW. Development and validation of a prognostic risk score for major bleeding in patients undergoing percutaneous coronary intervention via the femoral approach. Eur Heart J. 2007; 28: 1936–1945.
Manoukian SV, Feit F, Mehran R, Voeltz MD, Ebrahimi R, Hamon M, Dangas GD, Lincoff AM, White HD, Moses JW, King SB III, Ohman EM, Stone GW. Impact of major bleeding on 30-day mortality and clinical outcomes in patients with acute coronary syndromes: an analysis from the ACUITY trial. J Am Coll Cardiol. 2007; 49: 1362–1368.
Yan AT, Yan RT, Huynh T, DeYoung P, Weeks A, Fitchett DH, Langer A, Goodman SG; INTERACT Investigators. Bleeding and outcome in acute coronary syndrome: insights from continuous electrocardiogram monitoring in the Integrilin and Enoxaparin Randomized Assessment of Acute Coronary Syndrome Treatment (INTERACT) trial. Am Heart J. 2008; 156: 769–775.
Pawloski JR, Stamler JS. Nitric oxide in RBCs. Transfusion. 2002; 12: 1603–1609.
Reynolds JD, Ahearn GS, Angelo M, Zhang J, Cobb F, Stamler JS. S-Nitrosohemoglobin deficiency: a mechanism for loss of physiological activity in banked blood. Proc Natl Acad Sci U S A. 2007; 104: 17058–17062.
Nikolsky E, Aymong EA, Halkin A, Grines CL, Cox DA, Garcia E, Mehran R, Tcheng JE, Griffin JJ, Guagliumi G, Stuckey T, Turco M, Cohen DA, Negoita M, Lansky AJ, Stone GW. Impact of anemia in patients with acute myocardial infarction undergoing primary percutaneous coronary intervention: analysis from the CADILLAC trial. J Am Coll Cardiol. 2004; 44: 547–553.
Roy P, Bonello L, Torguson R, de Labriolle A, Lemesle G, Slottow TL, Steinberg DH, Kaneshige K, Xue Z, Satler LF, Kent KM, Suddath WO, Pichard AD, Lindsay J, Waksman R. Impact of “nuisance” bleeding on clopidogrel compliance in patients undergoing intracoronary drug-eluting stent implantation. Am J Cardiol. 2008; 102: 1614–1617.
Yusuf S, Mehta SR, Chrolavicius S, Afzal R, Pogue J, Granger CB, Budaj A, Peters RJ, Bassand JP, Wallentin L, Joyner C, Fox KA; Fifth Organization to Assess Strategies in Acute Ischemic Syndromes Investigators. Comparison of fondaparinux and enoxaparin in acute coronary syndromes. N Engl J Med. 2006; 354: 1464–1476.
Stone GW, Witzenbichler B, Guagliumi G, Peruga JZ, Brodie BR, Dudek D, Kornowski R, Hartmann F, Gersh BJ, Pocock SJ, Dangas G, Wong SC, Kirtane AJ, Parise H, Mehran R; HORIZONS-AMI Trial Investigators. Bivalirudin during primary PCI in acute myocardial infarction. N Engl J Med. 2008; 358: 2218–2230.
Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, Neumann FJ, Ardissino D, De Servi S, Murphy SA, Riesmeyer J, Weerakkody G, Gibson CM, Antman EM; TRITON-TIMI 38 Investigators. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2007; 357: 2001–2015.
Bassand JP, Hamm CW, Ardissino D, Boersma E, Budaj A, Fernández-Avilés F, Fox KA, Hasdai D, Ohman EM, Wallentin L, Wijns W; Task Force for Diagnosis and Treatment of Non-ST-Segment Elevation Acute Coronary Syndromes of European Society of Cardiology. Guidelines for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes. Eur Heart J. 2007; 28: 1598–1660.
Stone GW, Ware JH, Bertrand ME, Lincoff AM, Moses JW, Ohman EM, White HD, Feit F, Colombo A, McLaurin BT, Cox DA, Manoukian SV, Fahy M, Clayton TC, Mehran R, Pocock SJ; ACUITY Investigators. Antithrombotic strategies in patients with acute coronary syndromes undergoing early invasive management: one-year results from the ACUITY trial. JAMA. 2007; 298: 2497–2506.
On the basis of analysis of 13 819 patients with an acute coronary syndrome treated with contemporary antithrombotic strategies in the randomized Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial, we developed prognostic models for the risk of myocardial infarction (MI) and risk of non–coronary artery bypass grafting major bleeding and examined the manner in which bivalirudin compared with heparin plus a glycoprotein IIb/IIIa inhibitor affected these risks in individual patients. The risk factors for MI and major bleeding differed substantially, with only baseline ST-segment deviation and older age predicting both. In a covariate-adjusted analysis, treatment with bivalirudin alone rather than heparin plus a glycoprotein IIb/IIIa inhibitor was associated with a nonsignificant 8% increase in MI but a significant 50% reduction in major bleeding. Both MI and major bleeding affected subsequent mortality rate, with hazard ratios of 2.7 and 2.9, respectively (both P<0.001). Given the individual patient risk profiles and the fact that bivalirudin prevented ≈6 major bleeds for each MI that might occur, bivalirudin alone rather than heparin plus a glycoprotein IIb/IIIa inhibitor had an estimated reduction in bleeding risk that was greater than the estimated increase in MI risk for nearly all patients. When one considers each patient’s risk profile for MI and non–coronary artery bypass grafting major bleeding and the relative risk of these events with alternative antithrombotic strategies, a selection of the optimal pharmacological regimen for the individual patient can be made that would minimize the occurrence of these adverse events and their impact on subsequent mortality rate.
Guest Editor for this article was David J. Moliterno, MD.