Aspirin, Warfarin, or the Combination for Secondary Prevention of Coronary Events in Patients With Acute Coronary Syndromes and Prior Coronary ...

Aspirin, Warfarin, or the Combination for Secondary Prevention
 of Coronary Events in Patients With Acute Coronary Syndromes
           and Prior Coronary Artery Bypass Surgery
Thao Huynh, MD; Pierre Théroux, MD; Peter Bogaty, MD; James Nasmith, MD; Susan Solymoss, MD

Background—Patients with a non–ST-elevation acute coronary syndrome and prior CABG are at high risk of a recurrent
  ischemic event despite aspirin therapy. This trial investigated the potential benefit of secondary prevention with
Methods and Results—In a double-blind randomized trial, 135 patients with unstable angina or non–ST-segment elevation
  myocardial infarction, with prior CABG, and who were poor candidates for a revascularization procedure received
  therapy with aspirin and placebo⫹warfarin, warfarin and placebo⫹aspirin, or aspirin and warfarin for 12 months.
  Warfarin was titrated to an international normalized ratio of 2.0 to 2.5. The primary end point (death or myocardial
  infarction or unstable angina requiring hospitalization 1 year after randomization) occurred in 14.6% of the patients in
  the warfarin-alone group, in 11.5% of patients in the aspirin-alone group, and in 11.3% of patients randomized to the
  combination therapy (P⫽0.76). Subgroup analyses by risk features provided no indications that warfarin alone or in
  combination with aspirin could be of benefit over aspirin alone. Bleeding was more frequent in the 2 groups of patients
  administered warfarin.
Conclusions—Moderate-intensity oral anticoagulation alone or combined with low-dose aspirin does not appear to be
  superior to low-dose aspirin in the prevention of recurrent ischemic events in patients with non–ST-elevation acute
  coronary syndromes and previous CABG. (Circulation. 2001;103:3069-3074.)
                            Key Words: warfarin sodium 䡲 anticoagulants 䡲 aspirin 䡲 coronary disease

C    ardiac events are frequent in patients with previous
     CABG and are associated with an impaired prognosis.1– 4
Aspirin prevents graft closure during the first year after
                                                                                 unstable angina or non–ST-elevation myocardial infarction and prior
                                                                                 CABG were considered for the study. The diagnosis of unstable
                                                                                 angina was based on an accelerating pattern of chest pain occurring
                                                                                 at rest or with minimal exertion or of a prolonged chest pain and
surgery,5–12 but its long-term benefit in this specific setting                  normal serum levels of creatine kinase (CK). A non–ST-segment
has not been documented. Warfarin had been evaluated in                          elevation myocardial infarction was diagnosed when serum CK-MB
unstable angina,13 but no studies have addressed the specific                    increased above the upper normal limit of the site in the absence of
population of patients with unstable coronary syndromes                          new Q waves on the ECG. Patients who had coronary angioplasty or
without ST elevation developing late after CABG. Because of                      repeat CABG during the index hospitalization were excluded;
                                                                                 therefore, the study population was limited to patients who were poor
the high frequency of flow stasis, thrombi, and fibrin-rich                      candidates for a revascularization procedure. Other exclusion criteria
thrombi associated with venous graft disease,14 –16 long-term                    were as follows: contraindication to the use of aspirin or warfarin, a
anticoagulation with warfarin may be particularly beneficial                     treatable cause for angina pectoris, any major concomitant illness,
in this situation. We tested the hypothesis that moderate-                       congestive heart failure class 3 or 4 (New York Heart Association),
intensity warfarin (international normalized ratio [INR] 2 to                    uncontrolled systemic hypertension (blood pressure ⬎180/
                                                                                 95 mm Hg), recent major trauma, alcohol or drug abuse, females
2.5) either alone or in combination with low-dose aspirin will                   with child-bearing potential, coronary angioplasty within the last 6
be more effective than aspirin alone for the secondary                           months, conditions mandating treatment with aspirin (such as previ-
prevention of coronary events in patients with previous                          ous stroke) or with warfarin (such as metallic valve prosthesis), atrial
CABG.                                                                            fibrillation, or intracardiac thrombi. All patients gave informed
                                                                                 consent to participate, and the institutional review board at each
                                                                                 participating hospital approved the protocol.
Selection of Patients                                                            Study Protocol
The present study was conducted at 4 referral cardiology centers in              In a double-blind parallel-group study design, patients were random-
Quebec, Canada. All patients who presented with a diagnosis of                   ized to 1 of 3 parallel study groups: (1) aspirin plus placebo-warfarin,

  Received October 12, 2000; revision received April 9, 2001; accepted April 9, 2001.
  From the Montreal General Hospital (T.H., S.S.), the Montreal Heart Institute (P.T.), and Sacré-Coeur Hospital (J.N.), Montreal, Quebec, and the
Quebec Heart Institute (P.B.), Quebec, Quebec, Canada.
  Reprint requests to Pierre Theroux, MD, Montreal Heart Institute, 5000 Belanger East St, Montreal, Quebec, Canada H1T 1C8. E-mail
  © 2001 American Heart Association, Inc.
  Circulation is available at

3070        Circulation         June 26, 2001

                             TABLE 1.         Clinical Characteristics of Study Patients
                                                               Warfarin⫹Placebo    Aspirin⫹Placebo    Aspirin⫹Warfarin
                             Characteristic                         (n⫽45)              (n⫽46)             (n⫽44)             P
                             Age (mean⫾SD), y                       67⫾12               68⫾11              66⫾12          0.90
                             Females, %                              13.6                17.8                29.3         0.18
                             Mean LVEF, %                            30.9                29.4                29.4         0.99
                             LVEF ⬍40%, %                            53.3                41.3                56.8         0.29
                             Interval between randomization           8.5                 7.2                 6.9         0.20
                             and CABG, y
                             Prior MI, %                             62.2                56.5                72.7         0.38
                             Current smoking, %                      31.1                17.8                20.5         0.29
                             Hypertension, %                         37.8                34.8                38.6         0.73
                             Hyperlipidemia, %                       62.2                43.5                68.2         0.11
                             Diabetes mellitus, %                    15.6                17.4                25.0         0.50
                             CCS class ⬎3, %                         13.3                13.0                20.5         0.30
                             Cardiac catheterization,* %             57.7                65.2                52.2         0.46
                               MI indicates myocardial infarction. Values are mean percentage unless specified differently.
                               *During index hospitalization.

(2) warfarin plus placebo-aspirin, or (3) aspirin plus warfarin.                  and associated with elevated serum CK-MB levels (above the upper
Aspirin was administered as a daily dose of 80 mg in nonenteric                   normal limit of the site) or with new Q waves. Unstable angina was
coated form. Warfarin was given as crystalline warfarin sodium                    considered as a study end point only when it was severe enough to
tablets (Coumadin). Placebo tablets were supplied by Dupont                       require hospitalization and there was a confirmed diagnosis at
Pharma Canada. In-hospital follow-up visits were scheduled at 1, 3,               hospital discharge. Atypical chest pain requiring hospitalization was
6, and 12 months of treatment. The study drugs were terminated at                 not counted as an end point. All events were classified before
the 12-month follow-up visit, and aspirin (325 mg daily) was                      conditions of the blind study were revealed.
prescribed to all patients. Patients were reevaluated 1 month after the
discontinuation of the study drugs.                                               Statistical Analyses
                                                                                  The study was terminated prematurely after enrollment of half the
Study Monitoring                                                                  planned number of patients because of difficulty in recruiting
Measurements of INR were performed 3 days and 1 week after study                  because of the high rate of conventional or investigative procedures
drug initiation. Subsequent visits were scheduled depending on INR
                                                                                  performed in otherwise qualifying patients. The baseline character-
stability. Hemoglobin and hematocrit were monitored at the time of
                                                                                  istics of the 3 study groups were compared by ␹2 and Student tests.
INR determination. Unblinded pharmacists or physicians, not other-
                                                                                  End-point events were group-classified by the intention-to-treat
wise involved in the study and patient care, adjusted warfarin to a
target INR of 2.0 to 2.5 by using a predefined algorithm. To maintain             principle. Intergroup differences for event rates were compared by
the double-blind integrity, patients assigned to aspirin plus placebo-            the ␹2 statistics with the use of 2⫻3 tables. The event rates for the
warfarin therapy also had regular blood tests and mock placebo-                   primary end point of death, myocardial infarction, and documented
warfarin adjustments. Major bleeding was defined as a fall in                     recurrent unstable angina were displayed by the Kaplan-Meier
hemoglobin of ⱖ2 g/L or requiring blood product transfusion. Minor                survival method and tested for statistical significance by the log-rank
bleeding was defined as all other bleeding events reported by the                 statistic. Events rates were also compared in subgroups with higher
patients and/or health care professionals. In the event of clinically             risk features identified by univariate and multivariate analyses of the
significant bleeding, the study drugs were discontinued, and patients             baseline characteristics associated with an impaired prognosis. Sta-
were transfused as necessary. Compliance was defined as 100% of                   tistical significance was defined by a value of P⬍0.05. The SPSS
study medication taken (pills were counted by research coordina-                  Base 10.0 (SPSS Advanced Models 10.0 statistical software) served
tors). Study drugs could be temporarily interrupted for major                     for the statistical analyses.
illnesses, elective surgeries, and dental treatment. The study code
was broken only when it was essential for optimal patient                                                           Results
                                                                                  The baseline clinical characteristics of patients in the 3 study
Other Therapies                                                                   groups are shown in Table 1. Mean age, left ventricular
Concomitant medications could be prescribed at the discretion of the              systolic function (left ventricular ejection fraction [LVEF]),
attending physicians, but drugs known to interact with warfarin or to             and time elapsed after CABG were similar among the
increase the bleeding risk (barbiturates, NSAIDs, and salicylates)                patients. Although no statistically significant differences
were prohibited during the whole study period. All patients received
                                                                                  existed between the 3 groups, there were more patients that
aspirin at the end of the 12-month study period, when study drugs
were discontinued.                                                                were female and had prior MI, diabetes, and functional
                                                                                  impairment by angina in the group assigned to
Study End Points                                                                  aspirin⫹warfarin and fewer patients with depressed LVEF
The primary study end point was a composite end point of any-cause                (⬍40%) in the group assigned to aspirin⫹placebo than in the
death, myocardial infarction, or unstable angina requiring a new
                                                                                  other 2 groups. More than 60% of the patients had prior
hospitalization. Other end points were performance of reperfusion
procedure (either percutaneous or open chest). Myocardial infarction              myocardial infarction. Cardiac catheterization was performed
was defined as typical chest pain or discomfort lasting ⱖ30 minutes               in 60% of patients during the index hospitalization. The time
Huynh et al             Aspirin and Warfarin in ACS With Prior CABG                     3071

                          TABLE 2.       Baseline Medical Therapy of Study Patients
                                                          Warfarin⫹Placebo      Aspirin⫹Placebo     Warfarin⫹Aspirin
                          Medications                          (n⫽45)                (n⫽46)             (n⫽44)            P
                          ␤-Blockers, %                           57.8                71.7                 70.5          0.30
                          Calcium channel blockers, %             55.6                52.2                 54.5          0.95
                          Nitrates, %                             64.4                63.0                 75.0          0.95
                          Lipid-lowering agents, %                42.2                37.0                 36.4          0.82
                          ACE inhibitors, %                       35.6                15.3                 22.7          0.07
                          Antiarrhythmic agents, %                 6.8                   2.2                4.6          0.53
                          Diuretics, %                            26.7                11.0                 34.1          0.03

elapsed between the last CABG and occurrence of the index                      Patients using diuretics at the time of entry into the study had
acute coronary event averaged 7.5 years.                                       rates of 15.6%, 6.5%, and 13.6%, respectively, and those
   Table 2 shows the medical treatment received at the time of                 using ACE inhibitors had rates of 13.3%, 8.7%, and 9.1%,
randomization. Patients in the warfarin⫹placebo group more                     respectively.
often received ACE inhibitors and lipid-lowering drugs and                        Table 4 shows the distribution of complications among the
less often received ␤ -blockers; patients assigned to                          3 treatment arms. Minor bleeds were more frequent in
aspirin⫹placebo less often received diuretics and ACE inhib-                   patients receiving warfarin (P⫽0.02), either as single therapy
itors; and patients in the combination group more often                        (8.9%) or in combination (9.8%), than in those receiving only
received nitrates.                                                             aspirin (1.5%). Major bleeding occurred only in patients
   Table 3 presents the study end points for the 3 treatment                   receiving the anticoagulant. Two patients on warfarin⫹
arms, and the Figure presents the cumulative event rates. No                   placebo and 2 patients on warfarin⫹aspirin required blood
statistically significant differences existed between groups in                transfusions. In the present study, compared with warfarin
the incidence of the primary end point. End points regrouped                   alone, the addition of 80 mg of aspirin to warfarin treatment
or considered individually and secondary end points were all                   resulted in no excess bleeding. No significant excess in
less frequent in the aspirin-alone arm than in the 2 other arms.               clinical events occurred in the month after cessation of the
A percutaneous procedure during follow-up was also per-                        study drugs, with all patients administered open-label aspirin
formed more frequently in the 2 groups of patients random-                     (Figure).
ized to receive warfarin.                                                         The rate of completion of protocol was high among the 3
   Baseline characteristics associated with an impaired prog-                  groups. The reasons for premature cessation of study medi-
nosis were as follows: age ⱖ65 years (relative risk [RR] 1.65,                 cations were diverse; most of the reasons for cessation were
P⫽0.19), male sex (RR 2.32, P⫽0.11), diabetes (RR 2.23,                        medical conditions requiring nonstudy use of warfarin or
P⫽0.07), and Canadian Cardiovascular Society (CCS) angina                      aspirin. Three patients had to stop the study drugs because of
class ⱖ3 (RR 1.83, P⫽0.31). Treatment effects in all sub-                      bleeding. Nonstudy use of warfarin was required in 5 patients
groups showed no trends to a benefit with Coumadin alone or                    on warfarin⫹placebo, 2 patients on aspirin⫹placebo, and 2
in combination with aspirin over aspirin alone. Thus, the                      patients on aspirin⫹warfarin. Three patients (1 in each
event rates in the warfarin-alone, aspirin-alone, and                          treatment group) were lost to follow-up.
warfarin⫹aspirin groups were, respectively, as follows: in
patients aged ⬍65 years, 40%, 22.7%, and 19%; in patients                                                       Discussion
aged ⱖ65 years, 41.7%, 34.8%, and 36.8%; among women,                          No benefit of warfarin alone or of warfarin⫹aspirin com-
9%, 0%, and 10%; among men, 15.7%, 15.3%, and 13.8%;                           pared with aspirin alone could be documented for the sec-
among diabetics, 13%, 18%, and 30%; among patients with                        ondary prevention of ischemic events in the present study of
LVEF ⬍40%, 11%, 8%, and 12%; and among patients in                             patients with previous CABG and an acute coronary syn-
CCS class 3 or 4 at randomization, 6.3%, 0%, and 9.1%.                         drome (ACS). The composite end point and the various

                          TABLE 3.       End-Point Events According to Treatment
                                                     Warfarin⫹Placebo       Aspirin⫹Placebo       Warfarin⫹Aspirin
                          Events                          (n⫽45)                 (n⫽46)               (n⫽44)              P
                          Primary end point, %             14.1                   11.5                   11.4            0.76
                          Death, %                          0.7                    0.0                    1.6            0.34
                          MI, %                             3.0                    0.8                    2.9            0.43
                          UA, %                            12.6                   11.5                   10.6            0.88
                          PCI, %                            4.4                    0.8                    3.3            0.12
                          Repeat CABG, %                    1.5                    1.5                    1.6            0.99
                            UA indicates unstable angina requiring rehospitalization; PCI, percutaneous coronary intervention;
                          and MI, myocardial infarction. Primary end point is any-cause mortality, MI, or UA requiring
3072       Circulation       June 26, 2001

                          TABLE 4.     Complications and Adherence to Protocol by Patients
                                                   Warfarin⫹Placebo   Aspirin⫹Placebo   Aspirin⫹Warfarin
                          Complications                 (n⫽45)             (n⫽46)            (n⫽44)         P
                          Minor bleeding, %               8.9               1.5                9.8         0.02
                          Major bleeding, %               7.4               0.0                2.4         0.15
                          Blood transfusions, %           4.4               0                  4.5         0.34
                          Compliance, %                  90.0              86.7              86.1          0.66
                          Protocol completion, %         77.6              78.5              69.9          0.22

components of the end point were all less frequent in the               lipid-lowering agents, ACE inhibitors, and antithrombotic
aspirin-alone group. More frequent interventions were also              therapy. Yet, medical management in the study remained
performed during follow-up in patients receiving Coumadin,              suboptimal, with 20% of the hyperlipemic patients not being
although they were poor candidates for surgery, suggesting a            on lipid-lowering therapy at the time of randomization, 25%
lack of a benefit of the anticoagulation in preventing severe           of patients with depressed LVEF not being on ACE inhibi-
angina as well as recurrent ischemic events.                            tors, and 25% of the patients remaining active smokers.
   ACS in patients with previous CABG is usually associated                With the premature discontinuation of the study and an
with fibrin-rich blood clot,14 –16 more extensive coronary              observed event rate less than expected, the sample size of the
artery disease, and an impaired prognosis.2– 4 Other specific           present study was suboptimal. It provided 60% power to
features of high risk in the study population included mean             detect a 15% risk difference, 70% power to detect a 20% risk
age ⬎65 years, a high incidence of previous myocardial                  difference, and 88% power to detect a 25% difference
infarction and of diabetes mellitus, a low LVEF (30⫾2%),                between groups at a value of P⬍0.05. However, Coumadin
and, most important, selection of patients evaluated as poor            alone, compared with aspirin alone, was associated with an
candidates for a new revascularization procedure by coronary            increased absolute risk of 2.6%, and the combination of
angiography and, for noncatheterized patients, by a coronary            Coumadin with aspirin was associated with a minimal de-
anatomy known to be poorly suitable for revascularization.              crease of 0.1% (RR reduction of 0.87%), providing a low
   Despite these high-risk features, the primary end point in           probability of a significant gain with Coumadin. A sample
the present study was relatively infrequent, less than ex-
                                                                        size ⬎50 000 of patients would be required to document an
pected, and composed primarily of recurrent unstable angina
                                                                        eventual benefit of the combination Coumadin⫹aspirin over
with a relatively low rate of death or myocardial infarction.
                                                                        aspirin for a high futility index, assuming that the observed
These figures are in contrast to those of previous studies,
                                                                        event rates were not confounded by an imbalance in the
including one from our center, in which rates as high as 30%
                                                                        baseline characteristics of patients. Differences were indeed
to 40% were reported.2,4 The improved prognosis possibly
                                                                        observed between groups, although they were not statistically
reflects a more aggressive use of conventional and investiga-
                                                                        significant. Subanalysis of the results by variables influencing
tive intervention procedures in patients with severe angina as
                                                                        prognosis did not influence results, and no favorable trends
well as the progress made in medical therapy in recent years,
with emphasis on drugs that achieve plaque passivation and              emerged in any end point favoring Coumadin alone or in
better control of the disease process. Such drugs include               combination. However, it remains possible that the differ-
                                                                        ences observed, a combination of various differences, or
                                                                        other undetected differences could have had an impact on the
                                                                        results. Indeed, the univariate and multivariate analyses of
                                                                        determinants of prognosis suffered a similar lack of power as
                                                                        the outcome end points. Therefore, the present results should
                                                                        be interpreted as highly suggestive but not conclusive.
                                                                           Although no studies with aspirin and Coumadin have
                                                                        previously specifically addressed the specific population of
                                                                        patients with previous CABG developing an ACS, a large
                                                                        body of experience has been gained with aspirin, warfarin,
                                                                        and the combination in unstable angina and in CABG in
                                                                        general. The efficacy of aspirin in the prevention of death and
                                                                        myocardial infarction in patients with unstable angina has
                                                                        been well documented, as has the efficacy of aspirin in
                                                                        maintaining venous graft patency when initiated at the time of
                                                                        surgery.5–12 However, long-term benefits to maintain graft
Cumulative rates of death, myocardial infarction, or recurrent          patency are less well documented. Warfarin used alone can
unstable angina requiring hospitalization during 12-month treat-        prevent recurrence of unstable angina13 but has not been
ment period (months 0 to 12) and during month after discontin-
uation of study drugs and therapy with open-labeled aspirin             shown to be more effective than aspirin in maintaining graft
(ASA) (months 12 to 13).                                                patency.8 –12
Huynh et al           Aspirin and Warfarin in ACS With Prior CABG                               3073

   The combination of antiplatelet and antithrombotic therapy         lowering of LDL cholesterol levels.18,20 Therefore, the treat-
has been extensively studied. In a primary prevention trial in        ment currently recommended for the management of ACS in
high-risk men,17 low-intensity oral anticoagulation (INR 1.5)         high-risk patients with previous CABG is aspirin, clopi-
and low-dose aspirin (75 mg daily) had additive benefit.              dogrel, ACE inhibitors, and an aggressive control of risk
However, a low-intensity regimen was ineffective in large             factors aided by lipid-lowering drugs.
secondary prevention trials addressing patients with previous
CABG18 and patients with a recent myocardial infarction,19                                   Acknowledgments
although post-CABG patients appeared to have derived some             The authors wish to thank Susan Nguyen for the statistical analyses,
benefit, years after the discontinuation of Coumadin.20 The           Marie-Andrée Séguin for data compilation, and the research coordi-
combination therapy with moderate-intensity anticoagulation           nators for patient recruitment and follow-up.
to INR 2 to 3 might be more interesting. In a small
angiographic study, Williams and Stewart21 suggested less                                           References
                                                                       1. Lambert M, Kouz S, Campeau L. Preoperative and operative predictive
progression in the severity of the culprit coronary lesions with          variables of late clinical events following saphenous vein coronary artery
combination therapy compared with aspirin alone. Another                  bypass graft surgery. Can J Cardiol. 1989;5:87–92.
small study by Cohen et al22 showed a trend to a reduction in          2. Waters D, Walling A, Roy D, et al. Previous coronary artery bypass
coronary events with the combination therapy. Favorable                   grafting as an adverse prognostic factor in unstable angina pectoris. Am J
                                                                          Cardiol. 1986;58:465– 469.
trends were also found in the Organization to Assess Strate-           3. Wiseman A, Waters DD, Walling A, et al. Long-term prognosis after
gies for Ischemic Syndromes (OASIS) program, with fewer                   myocardial infarction in patients with previous coronary artery bypass
deaths, myocardial infarctions, strokes, and rehospitalizations           surgery. J Am Coll Cardiol. 1988;12:873– 880.
in patients receiving the combination therapy compared with            4. Kleiman NS, Anderson HV, Rogers WJ, et al. Comparison of outcome of
                                                                          patients with unstable angina and non-Q-wave acute myocardial
those receiving aspirin alone.23,24 All together, the results of          infarction with and without prior coronary artery bypass grafting (Throm-
the previous studies and of the present study do not suggest a            bolysis in Myocardial Ischemia III Registry). Am J Cardiol. 1996;77:
major benefit of Coumadin in ACS. One explanation could be                227–231.
a deficiency in important natural anticoagulants such as tissue        5. Chesebro JH, Clements IP, Fuster V, et al. A platelet-inhibitor drug trial
                                                                          in coronary artery bypass operations: benefit of perioperative dipyridam-
plasminogen activator/inhibitor and thrombomodulin in the                 ole and aspirin therapy on early postoperative vein-graft patency. N Engl
coronary circulation and in vein grafts submitted to arterial             J Med. 1984;307:73–78.
flow that contribute to create a thrombogenic state.25–28              6. Chesebro JH, Fuster V, Elveback LR, et al. Effect of dipyridamole and
                                                                          aspirin on late vein graft patency after coronary bypass operation. N Engl
                                                                          J Med. 1984;310:209 –214.
Clinical Implications                                                  7. Rajah SM, Salter MC, Donaldson DR, et al. Acetylsalicylic acid and
Although pathophysiological considerations make a strong                  dipyridamole improve the early patency of aorta-coronary bypass grafts:
case for the prolonged use of combined antiplatelet and                   a double-blind, placebo-controlled, randomized trial. J Thorac Car-
anticoagulant drugs, the various attempts made so far in that             diovasc Surg. 1985;90:373–377.
                                                                       8. Goldman S, Copeland J, Moritz T, et al. Starting aspirin therapy after
direction have not been rewarding. The present study, al-
                                                                          operation: effects on early patency. Circulation. 1991;84:520 –526.
though not conclusive because it was underpowered by a low             9. Brown BG, Cukingnan RA, DeRouen T, et al. Improved graft patency in
event rate, was composed of high-risk patients with venous                patients treated with platelet-inhibiting therapy after coronary bypass
graft disease most likely to benefit from anticoagulation. The            surgery. Circulation. 1985;72:138 –146.
                                                                      10. Pfisterer M, Jockers G, Regenass S, et al. Trial of low-dose aspirin plus
prolonged administration of low-molecular-weight heparins                 dipyridamole versus anticoagulants for prevention of aortocoronary vein
recently tested in 4 trials yielded no benefit,28 although 1 of           graft occlusion. Lancet. 1989;2:1–7.
the trials showed a benefit in the first few months of treatment      11. Van de Meer J, Hillege H, Kootstra G, et al. Prevention of one-year
that was not maintained, however, at the 6-month follow-                  vein-graft occlusion after aortocoronary-bypass surgery: a comparison of
                                                                          low-dose aspirin, low-dose aspirin plus dipyridamole and oral antico-
up.29 Specific inhibitors of coagulation factor Xa, oral hepa-            agulants. Lancet. 1993;342:257–264.
rins, and oral antithrombin drugs represent new therapeutic           12. Weber MA, Hasford J, Taillens C, et al. Low-dose aspirin versus anti-
opportunities that need to be explored. However, more potent              coagulants for prevention of coronary graft occlusion. Am J Cardiol.
antiplatelet therapy may be a better approach to secondary                1990;66:1464 –1468.
                                                                      13. Williams DD, Kirby MG, McPherson K, et al. Anticoagulant treatment in
prevention. Oral glycoprotein IIb/IIa receptor antagonists                unstable angina. Br J Clin Pract. 1986;40:114 –116.
have failed in this regard, but the Clopidogrel in the Unstable       14. Chen L, Theroux P, Lesperance J, et al. Angiographic features of vein
Angina to Prevent Recurrent Ischemic Events (CURE) trial                  grafts versus ungrafted coronary arteries in patients with unstable angina
recently showed that combination of aspirin and clopidogrel               and previous bypass surgery. J Am Coll Cardiol. 1996;28:493– 499.
                                                                      15. White C, Ramee S, Collins T, et al. Percutaneous angioscopy of
added significant benefit to aspirin alone in the prevention of           saphenous vein coronary bypass grafts. J Am Coll Cardiol. 1993;21:
recurrent ischemic events in patients with non–ST-elevation               1181–1185.
ACS.30                                                                16. Silva J, White C, Collins T, et al. Morphologic comparison of athero-
   Therapeutic measures addressing other pathophysiological               sclerotic lesions in native coronary arteries and saphenous vein grafts
                                                                          with intracoronary angioscopy in patients with unstable angina. Am
mechanisms of the disease are also very much rewarding. The               Heart J. 1998;136:156 –163.
low event rates observed in our population compared with              17. The Medical Research Council’s General Practice Research Framework.
historical controls support this hypothesis. In the Post Coro-            Thrombosis prevention trial: randomised trial of low-intensity oral anti-
nary Artery Bypass Graft trial of the National Heart, Lung,               coagulation with warfarin and low-dose aspirin in the primary prevention
                                                                          of ischemic heart disease in men at increased risk. Lancet. 1998;351:
and Blood Institute, the number of grafts with progression of             233–241.
atherosclerosis and the need for revascularization over a             18. The Post Coronary Artery Bypass Graft Trial Investigators. The effect of
4-year period were significantly reduced with aggressive                  aggressive lowering of low-density lipoprotein cholesterol levels and
3074           Circulation           June 26, 2001

      low-dose anticoagulation on obstructive changes in saphenous vein cor-       24. The Organization to Assess Strategies for Ischemic Syndromes (OASIS)
      onary artery bypass grafts. N Engl J Med. 1997;336:153–162.                      Investigators. Effects of long-term, moderate-intensity oral anticoagu-
19.   Coumadin Aspirin Reinfarction Study Investigators. Randomised                    lation in addition to aspirin in unstable angina. J Am Coll Cardiol.
      double-blind trial of fixed low-dose warfarin with aspirin after myo-            2001;37:475– 484.
      cardial infarction. Lancet. 1997;350:389 –396.                               25. Golledge J, Turner RJ, Gosling M, et al. Rapid changes in the coagulant
20.   Knatterud G, Rosenberg Y, Campeau L, et al. The Post CABG investi-               proteins on saphenous vein endothelium in response to arterial flow.
      gators: long-term effects on clinical outcomes of aggressive lowering of         Angiology. 1999;50:693–701.
      low-density lipoprotein cholesterol levels and low-dose anticoagulation in   26. Kauhanen P, Siren V, Carpen O, et al. Plasminogen activator inhibitor-1
      the Post Coronary Artery Bypass Graft Trial. Circulation. 2000;102:              in neointima of vein grafts: its role in reduced fibrinolytic potential and
      157–165.                                                                         graft failure. Circulation. 1997;96;1783–1789.
21.   Williams MJA, Stewart RAH. Coronary artery flow ten weeks after              27. Christie PD, Edelberg JM, Picard MH, et al. A murine model of myo-
      myocardial infarction or unstable angina: effects of combined warfarin           cardial microvascular thrombosis. J Clin Invest. 1999;104:533–539.
      and aspirin therapy. Int J Cardiol. 1999;69:19 –25.                          28. Eibelboom J, Anand S, Malmberg K, et al. Unfractionated heparin and
22.   Cohen M, Adams P, Parry G, et al. Combination antithrombotic therapy             low molecular weight heparin in acute coronary syndromes without ST
      in unstable rest angina and non-Q-wave infarction in nonprior aspirin            elevation: a meta-analysis. Lancet. 2000;355:1936 –1942.
      users: primary end points analysis from the ATACS trial: Antithrombotic      29. The FRISC-2 Investigators. Long-term low-molecular-mass heparin in
      Therapy in Acute Coronary Syndromes Research Group. Circulation.                 unstable coronary-artery disease: FRISC II prospective randomised mul-
      1994;89:81– 88.                                                                  ticentre study: FRagmin and Fast Revascularisation during InStability in
23.   Anand S, Yusuf S, Pogue J, et al. The OASIS Pilot Study Investigators:           Coronary artery disease. Lancet. 1999;354:701–707.
      long-term oral anticoagulant therapy in patients with unstable angina or     30. Yusuf S. Clopidogrel in the Unstable Angina to Prevent Recurrent ische-
      suspected non-Q wave myocardial infarction. Circulation. 1998;98:                mic Events (CURE). Presented at: Scientific Sessions, American College
      1064 –1070.                                                                      of Cardiology; March 18 –21, 2001; Orlando, Fla.
You can also read