Author + information
- Received January 2, 1997
- Revision received March 17, 1997
- Accepted April 16, 1997
- Published online August 1, 1997.
- Michael J.A. Williams, MB, ChB, FRACPAB,
- Ian M. Morison, MB, ChB, FRCPAAB,
- Josie H. Parker, MB, ChBAB and
- Ralph A.H. Stewart, MD, FRACPAB,* ()
- ↵*Dr. Ralph A. H. Stewart, Department of Medicine, University of Otago, PO Box 913, Dunedin, New Zealand.
Objectives. This study assessed whether combination therapy with aspirin and warfarin for 10 weeks reduces the risk of progression or reocclusion of the unstable coronary artery lesion.
Background. Reocclusion of the culprit coronary artery occurs in up to one third of patients during the 3 months after myocardial infarction (MI) or unstable angina and is associated with increased morbidity and mortality.
Methods. Fifty-seven patients presenting with unstable angina or MI who had an identifiable culprit lesion at coronary angiography were randomized in double-blind manner to receive warfarin (target international normalized ratio [INR] 2.0 to 2.5) or placebo in addition to aspirin (150 mg daily). Changes in the culprit lesion were assessed by quantitative angiography in 50 patients after 10 weeks of therapy or after a clinical event. Progression of the culprit lesion was defined as a decrease in minimal lumen diameter >0.4 mm or a new total occlusion. Regression was defined as an increase in minimal lumen diameter >0.4 mm.
Results. In subjects randomized to receive warfarin, the culprit lesion was less likely to progress (1 [4%] vs. 8 [33%]) and more likely to regress (5[19%] vs. 2[9%]) than in subjects receiving placebo (p = 0.02). Recurrent MI or a new occlusion at angiography occurred in 2 (7%) of 29 patients receiving warfarin versus 11 (39%) of 28 patients receiving placebo (p = 0.005).
Conclusions. In patients with an acute coronary syndrome, combined therapy with aspirin and warfarin with a target INR of 2.0 to 2.5 for 10 weeks reduces the risk of progression or reocclusion of the culprit coronary lesion.
There is an increased risk of recurrent coronary events during the first few weeks after an acute ischemic event . After the acute event, the ruptured atherosclerotic plaque remains potentially unstable with angiographic evidence of overlying thrombus and continued activation of procoagulant and inflammatory mediators . This condition predisposes to further thrombus formation and a risk of progression or reocclusion of the culprit coronary lesion. Both antiplatelet [3, 4]and antithrombin [5–7]therapies reduce the risk of myocardial infarction (MI) when given separately. By simultaneously inhibiting different pathways of thrombogenesis, combined antiplatelet and antithrombin therapy has the potential to more effectively reduce new thrombus formation over a fissured plaque. The aim of this study was to determine whether combination treatment with low dose aspirin and warfarin or administration of aspirin alone for 10 weeks after an episode of unstable coronary disease more favorably influenced the culprit lesion.
1.1 Study patients.
Patients were eligible for inclusion in the study if 1) they had had an acute MI or unstable angina with onset of symptoms <48 h before admission to the hospital; 2) the culprit coronary lesion could be identified at angiography performed 36 h to 8 days after admission; 3) the patient’s clinical status was stable and the patient’s physician did not consider early revascularization indicated; and 4) there was no recognized contraindication to anticoagulation with warfarin. Participants gave written informed consent and the study protocol was approved by the institutional ethics committee.
Myocardial infarction was defined as 1) acute onset ischemic chest pain of >20 min duration, 2) new Q waves (Q wave MI) or isolated T wave inversion (non-Q wave MI) on the electrocardiogram; and 3) an increase in creatine kinase >2 times the upper limit of normal. Patients with ST segment elevation ≥1 mV in at least two leads (≥2 mV in precordial leads) were treated with intravenous streptokinase or tissue plasminogen activator. Unstable angina was defined as 1) new onset ischemic chest pain of >20 min duration; 2) transient ST segment depression ≥0.1 mV or T wave inversion; and 3) serum creatine kinase <2 times the upper limit of normal. All patients received intravenous heparin to maintain activated partial thromboplastin time at 2 times the control level and oral aspirin (150 to 300 mg daily).
1.2 Study protocol.
Within 6 h of coronary angiography subjects were randomized in a double-blind manner to receive warfarin or placebo. Stratified randomization with randomly generated numbers was used to maintain similar proportions of patients with unstable angina and MI. The dose of warfarin was 10 mg for 2 days, then adjusted to the target international normalized ratio (INR) of 2.0 to 2.5. To maintain blinded treatment, INR tests were performed on all study patients and the dose was adjusted by a physician not involved with clinical management or angiographic analysis. Matched dose adjustment was undertaken in patients randomized to placebo, and INR results were not reported. After randomization all patients received subcutaneous heparin (12,500 U twice daily) for 3 days and aspirin (150 mg daily) for the duration of the study. Other medications were adjusted as clinically indicated.
Trial therapy was discontinued early for bleeding, clinical events (including myocardial infarction, death, coronary angioplasty, coronary artery bypass grafting) or withdrawal of patient consent. Trial medication was discontinued 2 days before follow-up angiography planned for 10 weeks after randomization.
1.3 Angiographic analysis.
Antianginal therapy, contrast media and the sequence of angiographic projections were documented at the baseline angiogram; these conditions were duplicated at the follow-up angiogram. All patients received sublingual glyceryl trinitrate spray, 800 μg, before angiography was performed. Paired films were later evaluated by a cardiologist who had no knowledge of treatment by using dual projectors to match frames for orientation and vessel filling. The projection that optimally visualized the maximal stenosis of the culprit lesion was chosen for analysis. Three sequential end diastolic frames were marked for digitization. Quantitative coronary angiography was then performed without knowledge of treatment, patient identity or temporal order by using the Ancor coronary analysis system (Siemens). The coronary catheter was used for calibration to determine absolute measurements in millimeters. After selection of a segment, the arterial borders were defined by an automatic edge detection algorithm and measurements displayed automatically. For each lesion the mean of measurements made on three end-diastolic frames was calculated .
Variability between baseline and follow-up angiograms was assessed in 34 nonculprit lesions. The standard deviation of the difference was 0.08 mm for repeat measurements of the same view and 0.18 mm for equivalent views obtained 2 months apart. Lesion progression or regression was defined as a change >0.4 mm (>2 SD of the 2-month variability) or a change from a patent (Thrombolysis in Myocardial Infarction [TIMI] flow grade 2 or 3) to an occluded (TIMI grade 0 or 1 flow) vessel. These criteria are similar to those used in previous reports [9–12].
1.4 Analysis and statistics.
The primary end point of the study was the mean change in the minimal lumen diameter of the culprit coronary lesion between baseline and follow-up angiography. The sample size of 50 paired angiograms was chosen to detect a 0.30-mm difference between the two treatments assuming an SD of 0.36 mm with a power of 0.80 and alpha 0.05. Change in percent stenosis of the culprit lesion and the proportion of patients with progression/regression or progression to occlusion of the culprit lesion were secondary end points of the study. The combined end point of progression to occlusion on angiography or MI was not defined prospectively. Two-sided ttests were used to compare the two treatment groups for continuous variables. The Fisher exact test was used to compare the proportions of categoric variables in each group. Statistical analysis was performed by using SPSS. A p value <0.05 was regarded as statistically significant.
For patients randomized to warfarin, the proportion of time spent in each INR range was calculated by using a computer program that assumes a linear change between two consecutive INR measurements .
Fifty-seven patients were randomized to receive warfarin (i.e., aspirin plus warfarin, n = 29) or aspirin alone (i.e., aspirin plus placebo, n = 28). Baseline characteristics for the two treatment groups were similar (Table 1). For the patients assigned to warfarin the median INR was 2.0. The percent of patient time spent within each INR range is presented in Fig. 1.
2.1 Clinical events.
Table 2provides details of the clinical events in each treatment group. One patient randomized to aspirin alone died after recurrent MI 5 days after randomization. Including this event, one patient (3%) randomized to warfarin and six patients (21%) randomized to aspirin alone had recurrent MI (p = 0.052). Differences in the incidence of death, MI or revascularization between patients randomized to warfarin and patients receiving aspirin alone were not statistically significant (5 [17%] vs. 9 [32%], p = 0.23).
Three patients randomized to warfarin discontinued therapy because of bleeding. One patient had major bleeding (INR = 2.5) requiring transfusion and oversewing of a duodenal ulcer 5 days after randomization. The other two patients had minor bleeding.
2.2 Angiographic results.
Baseline angiography was performed a median of 3 days (range 1.5 to 8) after admission; follow-up angiography was performed 64 days (range 3 to 78) after randomization. In 11 patients angiography was performed early because of a clinical event (median 40 days [range 3 to 69]). Follow-up angiography was not performed in seven patients (four randomized to aspirin alone and three to warfarin, Table 2).
At baseline, the mean minimal lumen diameter and percent stenosis of the culprit lesion were similar for the two treatment groups (Table 3). ⇓At follow-up, there was a decrease in mean minimal lumen diameter and an increase in the percent diameter stenosis in patients randomized to aspirin alone. The culprit lesion had regressed in two of these patients but had progressed in eight (in seven to total occlusion). For patients randomized to warfarin, the mean minimal lumen diameter did not decrease during follow-up; only one patient had progression to total occlusion and five had regression of the culprit lesion (Table 4). Differences between treatment groups in mean minimal lumen diameter and percent diameter stenosis were statistically significant.
2.3 Relation between recurrent coronary events and angiographic changes.
Of the eight patients with progression to occlusion at follow-up angiography, two had had an MI and two unstable angina. The other four patients had asymptomatic occlusion at the site of the culprit coronary lesion.
Follow-up angiography was performed in six of the seven patients who had recurrent MI (one patient died). Progression to occlusion was confirmed in two patients. Although the other four patients had patent arteries, they underwent angiography after receiving a thrombolytic drug or heparin, or both. Recurrent MI or angiographic evidence of occlusion occurred in 2 (7%) of 29 patients randomized to warfarin versus 11 (39%) of 28 randomized to aspirin alone (p = 0.005).
3.1 Prevention of late reocclusion with warfarin plus aspirin.
In this study, combination treatment with warfarin and aspirin was more successful in maintaining the average minimal lumen diameter and in reducing the incidence of progression and new occlusion of the culprit coronary lesion than was conventional therapy with aspirin alone. The study was designed to detect both progression and regression of the culprit lesion, but the most frequent change was progression from a patent to an occluded artery; an observation also made in other studies . This finding implies that in most cases lesion progression is due to recurrent thrombosis rather than to gradual remodeling and that the combination of warfarin and aspirin reduces the risk of new thrombus formation.
The rate of reocclusion in this study of 29% for subjects receiving aspirin alone is similar to the reocclusion rate at 3 months of 25% to 32% reported in other studies [15, 16]. In previous angiographic studies, treatment with warfarin alone or with combined aspirin and dipyridamole did not differ from placebo [15, 17]in reducing the rate of reocclusion. In this study, the reocclusion rate for patients receiving warfarin and aspirin was only 4% and there was a significant reduction in the combined end point of either progression to occlusion or MI.
3.2 Relation between angiographic progression and clinical events.
The importance of angiographic progression to occlusion has been well documented. After MI, left ventricular function is worse in patients with reocclusion without reinfarction than in those with a patent artery [18, 19]. Reocclusion of the culprit coronary artery is also associated with increased morbidity and mortality [19, 20]. It is therefore likely that preventing reocclusion will improve long-term prognosis even for patients in whom reocclusion is clinically silent.
Angiographic and clinical measures of progression were not always concordant. It is known that progression to occlusion can occur in patients whose clinical status remains stable [14, 16, 18]. In addition, several of our patients had recurrent MI during follow-up but had no change in the culprit lesion at subsequent angiography. In these patients it is likely that recurrent thrombosis at the culprit lesion was followed by thrombolysis.
3.3 Comparison with previous studies.
In previous studies, the combination of heparin and aspirin for 5 days in patients with unstable coronary disease either showed no significant benefit over heparin or aspirin alone or resulted in fewer events during treatment but no difference during longer follow-up . Although these observations may be partly explained by a rebound increase in events after withdrawal of heparin , they also emphasize the need for a longer duration of combination therapy to cover the period when the risk of recurrent events remains high.
The value of combined antiplatelet and antithrombin therapy suggested by this study is supported by recent studies with clinical end points. In two studies [23, 24]low molecular weight heparin and aspirin taken for 3 months after an acute coronary event significantly reduced the number of recurrent events. The Antithrombotic Therapy in Acute Coronary Syndromes (ATACS) study suggested a benefit of combined warfarin (INR 2 to 3) and aspirin (162.5 mg daily) commenced soon after hospital admission with a reduction in total ischemic events in the combination group versus aspirin alone (13% vs. 25%).
These results appear to conflict with a preliminary report from the Coumadin Aspirin Reinfarction Study (CARS) that found no benefit for combined Coumadin and aspirin compared with aspirin alone.1There are several possible explanations for this difference. In the CARS study the dose of aspirin was lower (80 mg/day) for patients randomized to Coumadin than for those randomized to placebo (160 mg/day). A fixed dose of Coumadin (1 mg or 3 mg) was used and a lower intensity of anticoagulation was achieved than in the current study or in earlier studies in which long-term anticoagulation reduced the risk of MI [5–7]. In the CARS study treatment was started up to 3 weeks after the acute coronary event, but it is known that the greatest risk for recurrent events is during the 1st 2 to 3 weeks after the index event [1, 23]. Early introduction of combination therapy and rapid achievement of target levels of anticoagulation may be needed to exert a significant impact on the culprit coronary lesion.
3.4 Study limitations.
The study group was a selected cohort of patients who had early angiography but were not referred for early angioplasty or coronary artery bypass surgery. In our center ∼30% of patients admitted with unstable angina or MI have early coronary angiography, and 15% to 20% undergo early revascularization by angioplasty or surgery. In New Zealand publicly funded access to coronary angioplasty and bypass surgery is restricted, with priority given to patients with ongoing ischemia, an early positive exercise test result or more extensive coronary artery disease.
Our study included patients with both MI and unstable angina. Differences in the use of thrombolytic therapy and the timing of the first angiogram may influence the severity of the culprit stenosis. The importance of these factors in determining the response to combination antithrombotic therapy is uncertain and could not be determined from the current study, which is too small to allow a useful subgroup analysis.
The study was not designed and had insufficient power to detect clinically important differences in morbidity or mortality between treatment allocations. Quantitative angiographic measures were chosen as the primary end points because they provide a more precise estimate of the effects of treatment on the culprit coronary lesion than do clinical events alone. The sample size was small, but it was sufficient to detect a statistically significant difference in progression of the culprit lesion with the use of more than one angiographic measure.
The combined clinical and angiographic end point of reocclusion or MI was not prospectively defined. This combined end point was included because several patients had clinical evidence of reocclusion of the culprit coronary artery but no change either at the culprit lesion or at other sites on subsequent angiography. Including an assessment of clinical as well as angiographic end points may also reduce the potential for bias that could occur if patients without a follow-up angiogram were excluded from the analysis.
The current study was too small to reliably evaluate the safety of combined warfarin and aspirin. In previous larger studies [25, 26]the risk of minor bleeding was increased for patients treated with this combination. However, the risk of major bleeding is not increased provided the INR is between 1.5 and 3.0 and aspirin dose is ≤150 mg daily [26–28]. Combined aspirin and warfarin with a target INR of 2.0 to 2.5 for 10 weeks may have a net benefit despite a small increase in bleeding risk because of the much higher absolute risk of recurrent cardiac events in the early period after presentation with unstable coronary artery disease.
Combined therapy with aspirin (150 mg daily) and warfarin with a target INR of 2.0 to 2.5 for 10 weeks after unstable angina or acute MI reduces the risk of progression or reocclusion of the unstable coronary lesion. This finding suggests that patients with an acute coronary syndrome may benefit from treatment with a combination of warfarin and aspirin that is started early and continued for several weeks or months. Because our study group was small, these conclusions are preliminary and further studies are needed to determine whether this therapeutic approach improves clinical as well as angiographic outcome.
We thank Deborah Scott, EN, and the staff of the cardiology service for their assistance with the study. We are grateful to F.R. Rosendaal, MD, University Hospital, Leiden who supplied the computer program for calculating patient INR times; Peter Herbison, MSc for statistical advice; and Harvey White, DSc, FRACP and John French, PhD, FRACP for valuable advice.
↵1 Fuster V. Low-dose Coumadin plus low-dose aspirin following myocardial infarction (CARS trial). Presented at the 45th Annual Scientific Session of the American College of Cardiology, March 1996, Orlando, Florida.
☆ The study was supported by grants from the Scott Trust, Dunedin, and Glaxo Pharmaceuticals Ltd., Auckland, New Zealand. Dr. Williams was supported by the Southland Medical Foundation, Invercargill, New Zealand as the W. & G.S. Dick Research Fellow.
- Coumadin Aspirin Reinfarction Study
- international normalized ratio
- myocardial infarction
- Thrombolysis in Myocardial Infarction
- Received January 2, 1997.
- Revision received March 17, 1997.
- Accepted April 16, 1997.
- The American College of Cardiology
- Fuster V
- Chesebro JH,
- Knatterud G,
- Roberts R,
- et al.
- Ambrose JA,
- Winters SL,
- Arora RR,
- et al.
- Waters D,
- Lesperance J,
- Craven TE,
- Hudon G,
- Gillam LD
- Chen L,
- Chester MR,
- Redwood S,
- Huang J,
- Leatham E,
- Kaski JC
- Meijer A,
- Verheugt FW,
- Werter CJ,
- Lie KI,
- van der Pol JM,
- van Eenige MJ
- White HD,
- French JK,
- Hamer AW,
- et al.
- Meijer A,
- Verheugt FW,
- van Eenige MJ,
- Werter CJ
- Ohman EM,
- Califf RM,
- Topol EJ,
- et al.
- White HD,
- Cross DB,
- Elliott JM,
- Norris RM,
- Yee TW
- Cohen M,
- Adams PC,
- Parry G,
- et al.
- Cohen M,
- Adams PC,
- Hawkins L,
- Bach M,
- Fuster V