SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

Background : Clopidogrel causes significantly less symptomatic peptic ulcer disease and gastrointestinal bleeding than low-dose aspirin in average-risk patients. The gastrotoxicity of clopidogrel in patients with active peptic ulcer disease is unknown.

Aim : To compare the incidence of unhealed ulcers in patients receiving clopidogrel or aspirin.

Methods : Patients with aspirin-induced peptic ulcer disease treated with omeprazole (20 mg/day) were randomized to receive clopidogrel (75 mg/day) or to continue with low-dose aspirin. Success was defined as ulcer/erosion healing at the eighth week.

Results : One hundred and twenty-nine patients were recruited (69 received clopidogrel and 60 continued with aspirin). Thirty-one (45%) in the clopidogrel group and 25 (42%) in the aspirin group had a minor gastrointestinal bleed. No ulcer showed an adherent clot or visible vessel. The distributions of peptic ulcer disease were similar in the clopidogrel and aspirin groups (gastric ulcer: 41% vs. 40%; duodenal ulcer: 10% vs. 12%; gastric ulcer + duodenal ulcer: 6% vs. 3%; gastritis: 32% vs. 37%; duodenitis: 4% vs. 7%; gastritis + duodenitis: 0% vs. 2%). Clopidogrel and aspirin were re-started after 0.86 ± 1.79 and 0.44 ± 1.60 days, respectively (P = 0.170). Three (4%) patients stopped clopidogrel due to drug rash. Using per protocol analysis, the treatment success rates of clopidogrel and aspirin were 94% (62/66) and 95% (57/60), respectively.

Conclusions : In patients with aspirin-associated peptic ulcer disease of low to moderate grade, both early conversion from aspirin to clopidogrel and continuation of aspirin are safe.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

Antiplatelet therapy is clearly recognized to reduce the incidence of cerebral and cardiovascular accidents and mortality from vascular causes in individuals with symptomatic atherothrombotic disease.1 Its use is frequently limited by gastrointestinal side-effects; these ranged from dyspepsia (31%) to life-threatening bleeding or perforation of gastroduodenal ulcers (3.1%) over a period of 4 years in the UK transient ischaemic attack study.2 The elderly are regarded as a high-risk group. In a study conducted by Silagy et al. clinically evident gastrointestinal bleeding occurred in 3% of elderly patients (≥ 70 years of age) receiving 100 mg of aspirin daily for 12 months.3

What is the best initial treatment of aspirin-induced symptomatic peptic ulcer? Unfortunately, no data are available on this. By analogy with trials using full-dose conventional non-steroidal anti-inflammatory drugs (NSAIDs), our current practice is to prescribe a proton pump inhibitor whilst continuing aspirin in patients without a massive gastrointestinal bleed. Although the discontinuation of aspirin during the period of ulcer healing may offer a theoretical advantage, there is always the potential to precipitate an ischaemic vascular event, particularly in high-risk patients with unstable angina.4

Recently, clopidogrel (Plavix, Bristol-Myers Squibb Co., New York, NY, USA), a new antiplatelet agent, has been approved by the Food and Drug Administration for use in the secondary prevention of heart attacks and stroke.5 Clopidogrel, a thienopyridine derivative similar to ticlopidine, differs from aspirin in the mechanism by which it inhibits platelet aggregation. Aspirin inhibits platelet aggregation by irreversibly blocking the enzyme cyclo-oxygenase, essential for the synthesis of thromboxane A2, a substance that causes both vasoconstriction and amplifies the platelet activation process leading to platelet aggregation.6 By contrast, the thienopyridines inhibit platelet aggregation by irreversibly inhibiting the binding of adenosine diphosphate, a substance that is released in platelets during activation and amplifies the aggregation process. This agent does not impair the prostaglandin-dependent mucosal protective mechanism, which is a side-effect of aspirin. The clinical efficacy of clopidogrel in the secondary prevention of coronary heart disease, peripheral vascular disease and ischaemic stroke has been demonstrated to be marginally better than that of aspirin in a randomized controlled clinical trial [Clopidogrel vs. Aspirin in Patients at Risk of Ischaemic Events (CAPRIE)].7 Furthermore, the incidence of gastrointestinal complications is significantly lower for clopidogrel than for aspirin (dyspepsia: 0.97% vs. 1.22%, P < 0.05; gastrointestinal haemorrhage: 0.52% vs. 0.72%, P < 0.05). Therefore, clopidogrel may be safer than aspirin in patients with low-dose aspirin-induced peptic ulcers treated by proton pump inhibitors.

The objective of this prospective study was to compare the incidence of unhealed ulcers in patients receiving clopidogrel or aspirin.

Method

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

Patients of either sex, 18–85 years of age, who required continuous aspirin at a daily dose in the range 80–160 mg for the secondary prevention of coronary heart disease, peripheral vascular disease and ischaemic stroke or transient ischaemic attacks were recruited. Patients who underwent upper endoscopy for dyspepsia or gastrointestinal bleeding and were found to have any of the following were invited to enrol into the study: ulcers of 3 mm or more in diameter and more than 10 erosions in the stomach or duodenum. Erosions were assessed by the modified Lanza scale.8, 9 Two antral biopsy specimens were obtained to determine the Helicobacter pylori status by histological examination. H. pylori infection was not treated before the completion of this study.

Major exclusion criteria were concurrent erosive or ulcerative oesophagitis, pyloric stenosis, major active gastrointestinal bleeding (shock, ulcer with adherent clot, visible vessels or arterial spurting requiring endoscopic haemostasis), thrombocytopenia, coagulopathy, including the use of warfarin, chronic diarrhoea or chronic skin disorder, known allergy to clopidogrel, aspirin or omeprazole, pregnancy or disorders that might modify the absorption of the study drugs. In addition, patients with an antiplatelet-free period of more than 7 days were also excluded.

This study was approved by the Ethics Committee of Ruttonjee Hospital. All patients provided written informed consent. All were treated with 20 mg/day omeprazole (Losec, Astra Hassle, Molndal, Sweden) for 8 weeks. They were randomly assigned to treatment with the original dose of aspirin or clopidogrel (Plavix, Bristol-Myers Squibb Co.) 75 mg/day for 8 weeks. Our study was not sponsored by pharmaceutical companies. With limited funding, we performed a single-blind study.

Follow-up

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

Patients were asked standardized questions about their overall upper gastrointestinal symptoms (epigastric or abdominal pain, heartburn, nausea, vomiting, upper abdominal bloating and empty feeling in the stomach) during the preceding 7 days. The overall symptoms and individual symptoms were graded as absent (grade 0), mild (easily tolerated) (grade 1), moderate (interfering with normal activities) (grade 2) or severe (incapacitating; leaving the patient unable to perform normal activities) (grade 3). Safety assessments were based on the reported symptoms, adverse events and the results of complete blood count and renal and liver biochemical tests. All patients were asked to bring all their medication bags and remaining tablets to the clinic visit for the assessment of the unintentional co-prescription of NSAIDs or warfarin and pill count. The physicians responsible for the clinic visits were blind to the drug regimens.

Upper endoscopy was performed at the eighth week. The study finished at this point. Treatment success was defined as the disappearance of ulcers and the presence of less than five erosions in the stomach or duodenum. The endoscopists were blind to the drug regimens.

Statistical analysis

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

The rates of successful treatment in the two groups and the categorical variables were compared using chi-squared test or Fisher's exact test when appropriate. All continuous variables and upper gastrointestinal symptoms (dyspepsia grade) were compared using Student's t-test.

The null hypothesis of this study was the non-superiority of clopidogrel over aspirin in the ulcer healing rate. The study was designed to have a power of at least 80% for a one-sided test at the 0.05 level of significance. The primary efficacy analysis used an intention-to-treat approach that included all patients meeting the major entry criteria who had taken at least one dose of medication. According to Yeomans et al., the healing rate of the aspirin group was assumed to be 80%,10 whereas the healing rate of non-NSAID ulcers was ∼ 95% at the eighth week. As clopidogrel did not inhibit cyclo-oxygenase-2, the healing rate of the clopidogrel group was assumed to be 95%, similar to that of non-NSAID ulcers. Fifty-five patients were required in each treatment protocol in order to detect a 15% difference.

According to the CAPRIE trial,7 the incidence rates of clopidogrel-induced skin rash and diarrhoea were 6.0% and 4.6%, respectively. Adjusting for these two adverse reactions specific to clopidogrel, the ratio of the sample size of the aspirin group to the clopidogrel group was set at 1.0 : 1.1. Assuming a 10% significant protocol violation rate, at least 65 and 60 patients had to be recruited to the clopidogrel and aspirin groups, respectively.

In the analysis, patients with clopidogrel-induced skin rash and diarrhoea were included in the intention-to-treat analysis, but were excluded in the per protocol analysis. A 95% confidence interval (95% CI) was constructed for the treatment success rates.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

One hundred and thirty-nine patients were recruited. Seventy-four received clopidogrel and 65 received aspirin. Five patients in each group were excluded after randomization. In the clopidogrel group, the reasons for exclusion were concomitant carcinoma of the colon (one), cardiac death due to myocardial infarction at the third week (one), unstable angina requiring a combination of aspirin and clopidogrel therapy at the fourth week (one), cancellation of follow-up endoscopy due to severe exacerbation of chronic obstructive pulmonary disease (one) and refusal of follow-up upper endoscopy (one). In the aspirin group, the reasons were cholecystectomy for biliary colic at the fifth week (one), carcinoma of the stomach confirmed by histological examination (two), unstable angina requiring a combination of aspirin and clopidogrel therapy at the third week (one) and refusal of follow-up upper endoscopy (one). At follow-up at the eighth week, the last three patients in the clopidogrel group and the last two patients in the aspirin group did not show any dyspepsia and the haemoglobin level remained stable.

One hundred and twenty-nine patients finally completed the study: 69 received clopidogrel plus omeprazole and 60 received aspirin plus omeprazole. The demographic characteristics, risk factors for peptic ulcer disease, previous history of peptic ulcer disease and the dosage or duration of aspirin used were similar at baseline between the two groups (Table 1). There was no significant difference in the degree of dyspepsia, incidence of gastrointestinal bleeding, location and type of peptic disease, ulcer size, grading of erosions, H. pylori status and the duration of antiplatelet-free period between the two groups (Table 2).

Table 1.  Base-line characteristics of the patients
 Clopidogrel group n = 69Aspirin group n = 69P
Age (year) 74.7 (9.3) 71.1 (12.6)0.098
Male sex 42 (60.9%) 39 (65.0%)0.628
Current personal habits
 Smoking  5 (7.2%)  6 (10.0%)0.576
 Alcohol drinking   1 (1.4%)  1 (1.7%)1.000
Risk factor for peptic ulcer
 Renal failure (serum creatinine> 400 umol/L)  7 (10.1%)   4 (6.7%)0.542
 Liver cirrhosis  0  0 
 Concomitant corticosteroid  0  0 
 Non-aspirin NSAID  0  0 
Previous history of peptic ulcer disease
 No. of patients 11 (11.9%)  9 (15.0%)0.883
 Gastrointestinal bleeding  4 (5.8%)  2 (3.3%)0.684
Indication for anti-platelet drug – no. (%)
 Ischemic heart disease 57 (82.6%) 50 (83.3%)0.924
 Ischemic stroke 22 (31.9%) 14 (23.3%)0.280
 Peripheral vascular disease  2 (2.9%)  0 (0.0%)0.499
Previous use of aspirin before randomization
 Dose – mg122 ± 35.5125.3 ± 34.80.589
 Duration – months 31.3 ± 35.6 28.8 ± 28.30.669
Anti-ulcer drug before randomization
 Nil 53 (76.8%) 46 (76.7%) 
 Antacid 11 (15.9%) 12 (20.0%) 
 H2-receptor antagonist   5 (7.2%)  2 (3.3%) 
Table 2.  Details of current peptic ulcer diseases
 Clopidogrel group n = 69Aspirin group n = 69P
  • * 

    Plus – minus values are means ± S.D.

Clinical severity
 Dyspeptic score – mean (s.d.) 1.52 ± 1.08 1.73 ± 0.950.244
 Minor gastrointestinal bleeding31 (44.9%)25 (41.7%)0.709
 Shock 0 (0%) 0 (0%)
Hem globulin (Hb)
 Lowest Hb – gm /dL11.0 ± 2.711.5 ± 2.90.239
 Hb <10 gm/dL – no. of patients26 (37.7%)19 (31.7%)0.435
Transfusion required – no. of patients11 (16.7%) 5 (8.3%)0.180
 1 unit 1 3 
 2 units  8 0 
 3 units  2 2 
Location and type of gastrointestinal disease
 Gastric ulcer28 (40.6%)24 (40.0%) 
 Duodenal ulcer 7 (10.1%) 7 (11.7%) 
 Gastric and duodenal ulcer  4 (5.8%)  2 (3.3%) 
 Gastric erosions27 (39.1%)22 (36.7%) 
 Duodenal erosions  3 (4.3%) 4 (6.7%) 
 Gastric erosion and duodenal erosions 0 (0%) 1 (1.7%) 
Characteristic of gastric or duodenal ulcers
 Ulcer diameter – cm 0.79 ± 0.63 0.85 ± 0.480.745
 Ulcer ≥ 2 cm in diameter  0 (0%) 1 (2.6%)1.000
 No. of ulcers 1.5 ± 1.2 1.8 ± 1.20.395
 Ulcer with active bleeding or non-bleeding visible vessels 0 (0%)0 (0%)
Characteristic of gastric or duodenal erosions  Lanza's grade 3.10 ± 0.84 2.96 ± 0.900.555
Helicobacter pylori infection – no. of patients31 (44.9%)33 (55.0%)0.254
Anti-platelet free period before randomization (range) – days0.86 ± 1.79 (0–7)0.44 ± 1.60 (0–7)0.170

Compliance to omeprazole was over 90% in all patients in both groups (Table 3). Clopidogrel was withdrawn in three (4.3%) patients due to the occurrence of a skin rash at the fourth week in two patients and at the fifth week in one patient. Two were switched to low-dose aspirin plus omeprazole. Follow-up endoscopy at the eighth week demonstrated complete ulcer healing. All the remaining patients were compliant to clopidogrel. In the aspirin group, one (1.7%) patient was non-compliant to aspirin (100 mg daily) due to persistent dyspepsia. Follow-up endoscopy at the eighth week demonstrated unhealed gastritis of Lanza grade 3.

Table 3.  Follow-up and outcome
 Clopidogrel group n = 69Aspirin group n = 69P
  • *

     Three patients with drug rash in clopidogrel group were excluded. The patient with aspirin non-compliance was classified as treatment failure.

  • ** 

    The three patients with drug rash in the clopidogrel group and the patient with aspirin non-compliance were classified as treatment failure.

Compliance of anti-platelet drug ≤90% 3 (4.3%) 1 (1.7%)0.623
Compliance of omeprazole ≤90% 0 (0%) 0 (0%)
Dyspepsia grade ≥ 2
• At 4th week – no. of patients 1 (1.4%) 2 (3.3%)0.597
• At 8th week – no. of patients 1 (1.4%) 1 (1.7%)1.000
Clinical adverse event
 Gastrointestinal bleeding 0 0 
 Perforation 0 0 
 Obstruction 0 0 
Treatment success
 Per-protocol analysis*62/66 (93.9%)57/60 (95.0%)1.000
 Intention-to-treat analysis**62/69 (90.0%)57/60 (95.0%)0.337
Details of treatment failure in intention-to-treat30
 Drug rash 4 3
 Non-healing of ulcer/erosions
  Gastric ulcer 1 0 
  Duodenal ulcer 0 1 
  Gastritis 3 2 

The overall assessment of dyspepsia symptoms showed no difference between the two groups at the fourth and eighth weeks. There was no re-bleeding or perforation of peptic ulcers.

In the intention-to-treat analysis, treatment success was achieved in 90.0% (95% CI, 82.5–97.2%) (62/69) of patients in the clopidogrel group and in 95.0% (95% CI, 89.3–100.0%) (57/60) of patients in the aspirin group (P = 0.337). The reasons for treatment failure were drug rash (three), unhealed gastric ulcer (one) and gastritis (three) in the clopidogrel group and unhealed duodenal ulcer (one) and gastritis (two) in the aspirin group. In the per protocol analysis, the three patients with clopidogrel-induced generalized skin rash were excluded and treatment success was achieved in 93.9% (95% CI, 88.0–99.8%) of patients in the clopidogrel group and 95.0% (95% CI, 89.3–100.0%) of patients in the aspirin group (P = 1.000).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References

Our previous retrospective study showed that no patient suffered from re-bleeding, perforation or obstruction when clopidogrel was co-prescribed with proton pump inhibitor in patients with peptic ulcer disease over a 1-year period, although the study was limited by the small sample size.11 In the present study, we tested the hypothesis that clopidogrel was superior to aspirin in the healing of gastroduodenal ulcers and erosions treated with omeprazole.

The rationale for this study was that clopidogrel caused significantly less gastrointestinal complications in the CAPRIE trial,7 and did not delay ulcer healing due to the absence of cyclo-oxygenase-2 inhibition.12, 13 If the conversion to clopidogrel was not shown to be superior to continuous aspirin, the latter would be the treatment of choice as clopidogrel is 100-fold more expensive than aspirin.

This study enrolled patients with small ulcers without an adherent clot or visible vessels and patients with moderately severe gastroduodenitis. Around 40% of patients in each group had a minor gastrointestinal bleed. Although the number of patients who fulfilled the exclusion criteria was not recorded prospectively, it is estimated that the reasons for exclusion were major gastrointestinal bleeding in 5% of patients and acute myocardial infarction or acute coronary syndrome in another 5% of patients. The latter group has a potential risk of provoking adverse cardiac events during follow-up endoscopy. Only patients with a low to moderate risk of ulcer complications were selected because of the lack of safety profile on clopidogrel used in patients with gastrointestinal bleeding. In order to sustain antiplatelet activity, the design of this study did not allow the antiplatelet drug-free period to be longer than 7 days, as the platelet life span is around 8–12 days.14 Clopidogrel and aspirin were re-started 0.86 and 0.44 days after upper endoscopy, respectively. This study demonstrated that the incidence of unhealed ulcer or erosions at the eighth week was similar in patients converted to clopidogrel plus omeprazole and in those continuing their original low-dose aspirin plus omeprazole (6% vs. 5%). These healing rates in both groups were similar to that in those who had stopped taking NSAIDs. Furthermore, no patient in either group had a re-bleed or a perforated peptic ulcer during the study period.

This prospective, single-blind, randomized controlled pilot study had several limitations. Firstly, the study was only powered to test the surrogate endoscopic end-point of ulcer healing, but not the standard clinical end-points of bleeding, perforation or obstruction. In our low- to moderate-risk patient group, the re-bleeding rate was estimated to be only 5%.15 Therefore, the estimated sample size is 400 for each group for a clinical outcome study. Secondly, patients with a high risk of ulcer re-bleeding (presence of an adherent clot, visible vessel or arterial spurting) were excluded. These ulcers have a higher re-bleeding rate of ∼10% even after initial endoscopic haemostasis.16 Subsequent studies need to address the safety of clopidogrel and aspirin in patients with a high risk of ulcer re-bleeding. This is particularly relevant in those with unstable angina and acute myocardial infarction when aspirin can significantly reduce cardiovascular mortality by 23% and 51%, respectively.17, 18 Furthermore, the additional benefit of the combination of clopidogrel and aspirin for patients with non-ST elevating acute coronary syndrome has been confirmed recently.19 The safety of dual clopidogrel and aspirin therapy in patients with active peptic ulcer also needs to be studied.

In conclusion, both early conversion from aspirin to clopidogrel and the continuation of low-dose aspirin are safe in patients with low to moderate bleeding/re-bleeding risk from active peptic ulcer disease treated with proton pump inhibitors. The conversion to clopidogrel does not appear to be superior to the continuation of low-dose aspirin in this subset of patients. Future studies are required to address high-risk bleeding peptic ulcers. A clinical outcome study with an adequate sample size is also needed.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Method
  6. Follow-up
  7. Statistical analysis
  8. Results
  9. Discussion
  10. References
  • 1
    Collaborative overview of randomised trials of antiplatelet therapy, I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Br Med J 1994; 308: 81106.
  • 2
    UK-TIA Study Group. The United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: final results. J Neurol Neurosurg Psychiatry 1991; 54: 104454.
  • 3
    Silagy CA, NcNeil JJ, Donnan GA, Tonkin AM, Worsam B, Campion K. Adverse effects of low-dose aspirin in healthy elderly population. Clin Pharmacol Ther 1993; 54: 849.
  • 4
    Patrono C, Coller B, Dalen JE, et al. Avoidance of a major vascular event in 50 patients per 1,000 patients with unstable angina who has been treated for 6 months. Platelet-active drugs: the relationships among dose, effectiveness, and side effects. Chest 2001; 119: 39S63S.
  • 5
    Food and Drug Administration. Food and Drug Administration Web Site. Drug Approvals for November 1997. Available at: http://www.fda.gov.
  • 6
    Schafer AI. Antiplatelet therapy. Am J Med 1996; 101: 199209.
  • 7
    CAPRIE Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischemic event (CAPRIE). Lancet 1996; 348: 132939.
  • 8
    Ehsanullah RS, Page MC, Tildesley G, Wood JR. Prevention of gastroduodenal damage induced by non-steroidal anti-inflammatory drugs: controlled trial of ranitidine. Br Med J 1988; 297: 101721.
  • 9
    Robinson MG, Friffin JW, Bower H, et al. Effect of ranitidine on gastroduodenal mucosal damage induced by nonsteroidal antiinflammatory drugs. Dig Dis Sci 1989; 34: 4248.
  • 10
    Yeomans ND, Tulassay Z, Juhasz L, et al. Omeprazole compared with ranitidine for ulcers associated with non-steroidal anti-inflammatory drugs. N Engl J Med 1998; 338: 71926.
  • 11
    Ng FH, Wong SY, Chang CM, et al. High incidence of clopidogrel-associated gastrointestinal bleeding in patients with previous peptic ulcer disease. Aliment Pharmacol Ther 2003; 18: 4439.
  • 12
    Mizuno H, Sakamoto C, Matsuda K, et al. Induction of cyclooxygenase 2 in gastric mucosal lesions and its inhibition by the specific antagonist delays healing in mice. Gastroenterology 1997; 112: 28797.
  • 13
    Sihigeta J, Takahashi S, Okabe S. Role of cyclooxygenase-2 in the healing of gastric ulcers in rats. J Pharmacol Exp Ther 1998; 286: 138390.
  • 14
    Aster RH. The study of platelet kinetics with 51Cr-labeled platelets. In: PaulusJM, ed. Platelet Kinetics. Amsterdam: North Holland, 1971: 31723.
  • 15
    Rockall TA, Logan RFA, Devlin HB, et al. Risk assessment after acute upper gastrointestinal haemorrhage. Gut 1996; 38: 31621.
  • 16
    Wong SK, Yu LM, Lau JY, et al. Prediction of therapeutic failure after adrenaline injection plus heater probe treatment in patients with bleeding peptic ulcer. Gut 2002; 50: 3225.
  • 17
    ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. Lancet 1988; 2: 34960.
  • 18
    Cairns JA, Gent M, Singer J, et al. Aspirin, sulfinpyrazone, or both in unstable angina. Results of a Canadian multicenter trial. N Engl J Med 1985; 313: 136975.
  • 19
    Yusuf S, Zhao F, Mehta ST, et al. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. The Clopidogrel in Unstable Angina to Prevent Recurrent Events Trial Investigators. N Engl J Med 2001; 345: 494502.