Comparison of High-Dose With Low-Dose Aspirin in Patients With Mechanical Heart Valve Replacement Treated With Oral Anticoagulant
Background There are no reported studies on the safety and efficacy of low-dose aspirin with low-intensity oral anticoagulation in patients with heart valve replacement. In this study, we compared the use of 100 mg/d aspirin with 650 mg/d aspirin in the prevention of systemic embolism and vascular death in patients with heart valve replacement who were being treated with oral anticoagulants with a target international normalized ratio (INR) of 2.0 to 3.0.
Methods and Results Four hundred nine of 416 consecutive patients who had cardiac valve replacement were randomized in open allocation into one of two groups; both groups were treated with oral anticoagulant therapy with a target INR of 2.0 to 3.0. Two hundred seven patients who received 100 mg/d aspirin for an average of 24.1 months were compared with 202 patients who received 650 mg/d aspirin for an average of 21.7 months in a randomized-treatment, open-allocation study. There were no significant differences in systemic embolism, vascular death, or total death rates between the low- and high-dose aspirin treatment groups (0.5 and 1.1, 1.2 and 0.5, and 4.6 and 2.5 per 100 patients/y, respectively). The total number of hemorrhagic events was 13.4 per 100 patients/y in the high-dose aspirin group and 7.9 per 100 patients/y in the low-dose aspirin group (P=.035), but the rate of bleeding was influenced by dipyridamole in the 650-mg aspirin group.
Conclusions In patients with mechanical heart valve replacements, low-dose aspirin (100 mg/d) in conjunction with oral anticoagulants at an INR of 2.0 to 3.0 is as effective as the use of high-dose aspirin (650 mg/d) in the prevention of systemic embolism.
Thromboembolism remains a serious complication after prosthetic heart valve replacement despite improvements in valve design and manufacture. To reduce the risk of thromboembolism, lifelong anticoagulant therapy is widely recommended for such patients. Guidelines published by the American College of Chest Physicians1 and the British Society for Haematology2 recommend the use of oral anticoagulant therapy with a relatively high international normalized ratio (INR). A number of studies have shown that the addition of antiplatelet agents to oral anticoagulants reduces the risk of thromboembolism in heart valve replacement patients in comparison with the use of oral anticoagulants alone.3 4 5 6 7
In one study,8 aspirin given in a low dosage of 100 mg/d to patients with mechanical heart valves or to high-risk patients with tissue valves who were receiving warfarin with a target INR of 3.0 to 4.5 reduced mortality as well as major systemic embolism but increased the bleeding risk. High-dose antiplatelet therapy with aspirin (660 mg) plus dipyridamole (150 mg/d) in conjunction with a less-intense anticoagulant regimen at a target INR of 2.0 to 3.0 has been shown to be effective in reducing the risk of thromboembolic complications but with less bleeding than was found with the high-intensity anticoagulant regimen.7 The new guidelines published recently by the American College of Chest Physicians state that the addition of aspirin (100 mg/d) to warfarin offers additional protection in preventing systemic embolism and vascular death in patients with mechanical prosthetic cardiac valves.9 10 Aspirin (100 mg/d) plus an oral anticoagulant at a target INR of 2.5 to 3.5 is recommended for patients with mechanical heart prostheses who have systemic embolism despite adequate therapy with oral anticoagulant.10
The important question that remains is whether low-dose aspirin would be as effective as high-dose aspirin in combination with a less-intense anticoagulant regimen with a further reduction in the risk of bleeding. We therefore performed a randomized trial in patients with mechanical heart valve replacement who were treated with oral anticoagulants at a target INR of 2.0 to 3.0 to determine whether aspirin given in a dosage of 100 mg/d was as effective in the prevention of systemic embolism with less bleeding as aspirin given in a dosage of 650 mg/d with or without dipyridamole.
The study was approved by the Institutional Review Board of the Sanatorio Gu¨emes, Buenos Aires, and was carried out between July 1988 and February 1992. Consecutive patients with mechanical heart valve replacement were eligible for entry into the study. Patients were excluded if they had a contraindication to aspirin or oral anticoagulant or if they were unwilling to give informed consent.
All patients received oral anticoagulant therapy (acenocoumarol) monitored to give an INR of 2.0 to 3.0. The study was an open trial in which patients were randomized to receive low-dose aspirin (100 mg/d; group A) or high-dose aspirin (650 mg/d; group B). In group B only, patients could also receive dipyridamole (150 mg/d) according to the preference of the treating physicians. Randomization was performed in groups of 20 to ensure an even distribution of patients among treatment groups. Patients were assigned to each group according to a computer-generated random assignment.
The principal outcome events of the study, which were adjudicated by a panel of experts who were blinded to the treatment allocation, were systemic embolism, death, and major hemorrhage. Cerebral embolism was defined as a sudden, temporary, or permanent neurological disorder, including memory loss, focal limb weakness, or visual disturbance. Routinely, a computed tomography scan was indicated in the presence of an intracranial event to confirm infarction or hemorrhage.
Major hemorrhage was defined as hemorrhage that required temporary discontinuation of anticoagulant therapy with or without the need for blood transfusion or hospitalization. All other hemorrhages were classified as minor. The present definition was used in a previous trial and found to be clinically appropriate.7
Prothrombin time (Quick method), reported as the INR, using locally produced human brain thromboplastin standardized with the Argentine reference thromboplastin and activated partial thromboplastin time were determined daily during the first 5 to 7 days according to standard methods.
After discharge from the hospital, patients were monitored to control anticoagulant therapy at 15- to 20-day intervals with adjustment of the acenocoumarol dose to maintain the INR at 2.0 to 3.0 and were examined for clinical outcomes at 4- to 6-month intervals. In addition, patients were instructed to immediately report any outcomes to study personnel. Compliance for antiplatelet drugs was tested every 3 to 4 months by monitoring platelet aggregation with arachidonic acid, sodium salt (90% pure; Sigma Chemical Co) (0.75 mmol/L) as the aggregating agent. Complete inhibition of platelet aggregation was found in 97% of the tests of patients in group A and 100% of patients in group B.
Analysis was performed on an intention-to-treat basis. For thromboembolic episodes and death, Mantel-Cox, Taron-Ware, and generalized Wilcoxon (Breslow) methods were used; for bleeding events, statistical analyses were performed with the χ2 test. Overall survival rates were calculated with the Kaplan-Meier product-limit method. Statistical analyses were performed on the rates of vascular death, total death, and thromboembolic complications; on the composite outcome of thromboembolic episodes plus mortality; and on thromboembolic events plus vascular mortality. All tests of significance were two sided.
Four hundred sixteen consecutive patients were evaluated for eligibility to enter the trial; 7 were excluded for the following reasons: contraindication to aspirin (1 patient), contraindication to oral anticoagulants (5 patients), and refused consent (1 patient). Of the 409 patients included in the study, 207 were randomized to receive 100 mg/d aspirin (group A) and 202 to receive 650 mg/d aspirin (group B). One hundred thirty-five patients in group B who received 650 mg/d aspirin also received 150 mg/d dipyridamole. The baseline characteristics of the patients, which were comparable in the two groups, are given in Table 1⇓. Twenty-eight patients in group A and 33 patients in group B had associated surgical procedures, including 22 patients in group A and 25 in group B who had coronary artery bypass graft procedures. Fourteen patients in group A and 12 in group B had had previous valve surgery.
Four patients in group A and 7 patients in group B failed to attend follow-up. Two patients in group B failed to comply with the assigned regimen; 2 patients in group B were withdrawn by their attending physicians; and 1 patient in group B developed endocarditis. In addition, 13 patients in group B were intolerant of the antiplatelet regimen, which was discontinued within 10 days of the start of treatment.
Patients were followed for a maximun of 44 months in group A and 40 months in group B. Patients in group A had a follow-up of 4994 months (average, 24.1 months per patient); in group B, the follow-up was 4394 months (21.7 months per patient). In the 135 group B patients who were treated with aspirin plus dipyridamole, follow-up was for 3294 months (average, 24.4 months per patient); for the group B patients treated with aspirin alone, the follow-up was for 1100 months (average, 16.6 months per patient). Adequacy of anticoagulant therapy was equal in both groups, with the INR in the therapeutic range of 70% in group A and 69.5% in group B. In 17% of group A and 16.8% of group B patients, the INR values were lower than the target therapeutic range; in 13% of group A and 13.3% of group B, INR values were higher than the target range.
Deaths and Systemic Embolism
There were 19 deaths among the 207 patients in the low-dose aspirin group (4.6 per 100 patients/y) compared with 9 deaths among the 202 patients in the high-dose aspirin group (2.5 per 100 patients/y). The overall survival curves for groups A and B are shown in the Figure⇓. The difference in mortality rates between the two groups was not statistically significant. The classification of death for the two treatment groups is shown in Table 2⇓. Five patients in group A (1.2 per 100 patients/y) and 2 patients in group B (0.5 per 100 patients/y) died from vascular causes. In group A, the vascular deaths included 1 sudden death, 2 acute heart failure deaths, and 2 deaths due to acute valvular failure; in group B, there was 1 sudden death and 1 acute ischemic stroke. Nine patients in group A and 3 patients in group B died from nonvascular causes, and the cause of death was unknown for 3 patients in each group. Two patients in the low-dose and 1 patient in the high-dose group died as a result of hemorrhage.
Two patients (0.5 per 100 patients/y) in group A and four patients (1.1 per 100 patients/y) in group B had nonfatal systemic emboli (Table 2⇑). There were no statistically significant differences in the composite outcomes of death and nonfatal systemic embolism or vascular death and nonfatal embolism (Table 2) in the two treatment groups.
Nonfatal Hemorrhagic Complications
Thirty-three (7.9 per 100 patients/y) and 49 (13.4 per 100 patients/y) nonfatal events were observed in groups A and B, respectively. The difference between the two groups in the frequency of all bleeding events was statistically significant (P=.035). There was no difference in the frequency of major bleeding events between the two groups (Table 3⇓). The addition of dipyridamole to 650 mg/d aspirin did not reduce the rate of thromboembolism in group B but resulted in an increase in minor bleeding events (P=.04) (Table 3⇓).
There were 15 major hemorrhagic events in 207 patients of group A (3.6 per 100 patients/y) and 19 in 202 patients of group B (5.1 per 100 patients/y) (Tables 3⇑ and 4⇓). In 19 patients (10 of group A and 9 of group B), the INR was measured at the time of the major bleeding event; in 4 patients of group A and 5 of group B, the hemorrhagic events occurred when the INR was >3.0.
The pathogenesis of thromboembolic complications in patients with prosthetic heart valve replacement involves both platelet activation and blood coagulation. Although anticoagulants significantly reduce the risk of embolic complications, inhibition of both coagulation and platelet function may improve the efficacy of antithrombotic therapy in patients with prosthetic heart valves.
The use of high-intensity oral anticoagulant (INR, 3.0 to 4.5) with dipyridamole4 or with high-dose aspirin (1000 mg/d)5 in patients with prosthetic heart valve replacement has been shown to be more effective in the prevention of systemic embolism than the use of anticoagulants alone, but there have been significant bleeding rates. A reduced-intensity anticoagulation (INR, 2.0 to 3.0) with the use of high-dose aspirin (650 mg) plus dipyridamole (150 mg) has resulted in fewer bleeding episodes compared with what occurred with a higher-intensity anticoagulation regimen.7 Turpie et al8 found that the addition of 100 mg aspirin to warfarin therapy reduced mortality and major systemic embolism in patients with mechanical heart valves and in high-risk patients with prosthetic tissue valves. In that trial, patients were maintained at a higher intensity of anticoagulation (INR, 3.0 to 4.5) in comparison to the present study. The mean INR was close to 3.0 in both groups of patients (3.1 in the aspirin group and 3.0 in the placebo group), and 37% of the time was below the target INR, indicating that a relatively lower anticoagulant intensity might be sufficient. However, in that study, patients who received the combined regimen of low-dose aspirin with high-intensity anticoagulation had more total hemorrhages than patients on anticoagulants alone, although major hemorrhage was similar in the two groups.
In the present study, a low-intensity anticoagulant regimen (INR, 2.0 to 3.0) was used in conjunction with either a low dose (100 mg/d) or high dose (650 mg/d) of aspirin to determine whether there was similar efficacy with less bleeding with the low-dose regimen. The results show that both doses of aspirin gave similar protection from the major outcomes of vascular death or systemic embolism, with significantly less total bleeding in the low-dose group; however, bleeding in the high-dose aspirin group could be attributed to the dipyridamole administered with aspirin in 135 patients. Although total mortality from all causes was higher in the group with low-dose aspirin, there was no statistical difference in the frequency of vascular deaths between the two groups. Because sample size was small and a β error cannot be excluded, the finding for total mortality should be interpreted with caution. However, with the observed mortalities of 4.6% and 2.5% in the low- and high-dose aspirin groups, respectively, it would require >1300 patients in each group to detect a statistically significant difference (P<.05) with a power of 80%. For there to have been this difference in the present sample size, the mortality in the low-dose aspirin group would have to have been 8.6%.
Nevertheless, the total number of vascular complications (vascular deaths plus systemic embolism), events that are supposed to be lowered by antithrombotic therapy, was similar in each group (aspirin 100 mg/d group, 1.7 per 100 patients/y; aspirin 650 mg/d group, 1.6 per 100 patients/y; P=NS) (Table 2⇑). The addition of dipyridamole to the high-dose aspirin regimen did not give further protection and produced a trend toward increased bleeding. In comparison with other trials,1 the present study reports comparatively low rates of both thromboembolic and hemorrhagic events.
The results of this study show that a less-intense anticoagulant regimen (INR, 2.0 to 3.0) plus low-dose aspirin (100 mg/d) causes low rates of thromboembolic complications, including vascular death or systemic embolism, comparable to rates with a high-dose aspirin (650 mg/d) regimen with or without dipyridamole. Dipyridamole did not add a further protection and increased the rise of minor bleeding events. Thus, the combination of low-dose aspirin (100 mg/d) with low-intensity oral anticoagulant (INR, 2.0 to 3.0) may be recommended for patients with mechanical heart valve replacement independent of the antecedent of systemic embolism or the presence of other risk factors.10
The benefit of the prevention of thrombotic events in coronary diseases by the combined use of low-intensity oral anticoagulant and a low dose of aspirin is a possibility that will be addressed in ongoing trials.13 14
- Received November 14, 1995.
- Revision received May 2, 1996.
- Accepted May 20, 1996.
- Copyright © 1996 by American Heart Association
Stein PD, Kantrowitz A. Antithrombotic therapy in mechanical and biological prosthetic heart valves and saphenous vein bypass grafts. Chest. 1992;102(suppl):445S-455S.
British Society for Haematology. Guidelines on oral anticoagulation. J Clin Pathol.. 1990;43:177-183.
Sullivan JM, Harken DE, Gorlin R. Pharmacologic control of thromboembolic complications of cardiac-valve replacement. N Engl J Med.. 1971;248:1391-1394.
Chesebro JH, Fuster V, Elveback LR, McGoon DC, Pluth JR, Puga FJ, Wallace RB, Danielson GK, Orszulak TA, Piehler JM, Schaff HV. Trial of combined warfarin plus dipyridamole or aspirin therapy in prosthetic heart valve replacement: danger of aspirin compared with dipyridamole. Am J Cardiol.. 1983;51:1537-1541.
Hirsh J, Dalen JE, Fuster V, Harker LB, Patrono C, Roth G. Aspirin and other platelet-active drugs: the relationships among dose, effectiveness, and side effects. Chest. 1995;108(suppl):247S-257S.
Stein PD, Alpert JS, Copeland J, Dalen JE, Goldman S, Turpie AGG. Antithrombotic therapy in patients with mechanical and biological prosthetic heart valves. Chest. 1995;108(suppl):371S-379S.
Hoogendijk EMG, ten Cate JW. Aspirin and platelets. Lancet.. 1980;1:93-94.
Jakubowski JA, Stampfer MJ, Vaillancourt R, Deykin D. Cumulative antiplatelet effect of low-dose enteric coated aspirin. Br J Haematol.. 1982;60:635-642.