Previous Aspirin Use May Attenuate the Severity of the Manifestation of Acute Ischemic Syndromes
Background The present study was designed to investigate whether the prior use of aspirin could influence the severity of the manifestation of acute coronary artery syndromes, given the well-documented observations that aspirin can prevent myocardial infarction, stroke, and death in cardiovascular disease.
Methods and Results A series of 539 consecutive patients admitted to the Coronary Care Unit of a General Hospital was carefully characterized in a study with an ambidirectional design, with regard to previous medical history, aspirin use, and subsequent hospital diagnosis. Among the 214 patients previously taking aspirin, the hospital diagnosis was myocardial infarction in 24% and unstable angina in 76% compared with 54% and 46%, respectively, among the 325 not taking aspirin (P<.0001), for a reduction in the odds ratio of myocardial infarction with aspirin of 72% (95% CI, 59% to 90%). The decrease in odds was homogeneous in all subsets studied and independent of age, sex, previous angina, or previous myocardial infarction. The myocardial infarction was of a Q-wave type in 62% of aspirin users compared with 76% of nonusers (P<.05). By multivariate analysis, previous aspirin use was a strong predictor of unstable angina versus myocardial infarction and the only independent predictor of non–Q-wave versus Q-wave myocardial infarction.
Conclusions This study, thus, suggests a shift to less severe manifestation of acute coronary syndromes with aspirin use, implying that the failure of the drug in many patients with an acute coronary syndrome is only partial and that aspirin has the potential of attenuating the severity of the underlying acute thrombotic disease process.
Numerous clinical trials1 2 3 4 5 supported by a large meta-analysis6 7 have documented that antiplatelet therapy with aspirin modifies the natural history of coronary artery disease. These benefits apply to acute coronary syndromes to prevent the complications of unstable angina8 9 and of myocardial infarction.10
Therapy with aspirin, however, often fails to prevent cardiac events, and many patients on chronic aspirin therapy die or are hospitalized for acute coronary syndromes. These patients were compared with patients not previously taking aspirin in subanalyses of a few clinical trials.11 12 13 In general, no differences were observed in clinical findings, presenting symptoms, and disease severity. These observations contrast with the benefit of aspirin to prevent myocardial infarction in primary prevention trials4 and myocardial infarction and death in secondary prevention trials.1 2 3 4 5 6 7 The present study was specifically designed to investigate whether the previous use of aspirin could in some way influence the manifestation of acute coronary syndromes when these occur. For this purpose, a large and consecutive series of patients admitted to the Coronary Care Unit of a General Hospital was carefully characterized with regard to previous aspirin use, clinical characteristics, and in-hospital diagnosis.
All patients admitted to Hospital General Universitari Vall d’Hebron between October 1991 and March 1993 for suspected unstable angina or myocardial infarction and <81 years old were prospectively entered into the study registry. A total of 648 patients were screened. By predefined entry criteria, 37 patients using oral anticoagulants or antiplatelet drugs other than aspirin and 72 patients with a final diagnosis by the treating physician of noncardiac chest pain based on clinical investigation helped by coronary angiography and/or exercise treadmill testing were excluded from the primary analysis. The final diagnosis in the 539 enrolled patients was unstable angina in 312 (58%) and myocardial infarction in 227 (42%), including 62 patients with non–Q-wave myocardial infarction and 165 with Q-wave infarction. A secondary analysis including patients with noncardiac chest pain was also performed.
A standardized questionnaire with 87 variables was completed directly with the patient within 72 hours of admission by one of two study investigators. This questionnaire described demographic variables, risk factors, previous medical history, clinical events that precipitated the hospitalization, and a list of all medications consumed by the patients in the previous week, month, and year. The part of the questionnaire relating to previous use of aspirin was obtained by a particularly detailed standard questionnaire. It included a list of all pharmaceutical preparations containing aspirin available in Spain. The weekly dose and the duration of use before the event were recorded, as well as reasons for taking the medication. Confirmation of the data and additional information, if needed, were obtained from close relatives of the patients. These data were entered into the database before the clinical chart of the patient was consulted, and the final diagnosis of the index event was entered at a different time.
A 12-lead ECG was obtained at admission and daily during the hospital stay. Creatine kinase (CK) and isoenzyme plasma levels were determined at admission and after 8 hours and, when elevated, every 8 hours until normalization. The ECGs and the serial enzyme determinations were repeated whenever chest pain recurred. A 99mTc pyrophosphate scan was obtained in patients consulting more than 24 hours after an episode of prolonged pain when no new Q waves were documented.
The in-hospital treatment included, unless contraindications were present, routine intravenous heparin and oral aspirin in all patients and thrombolysis in patients with ST-segment elevation admitted within 6 hours after onset of pain. The hospital diagnosis of the qualifying event was validated by a consensus of two study investigators using the objective information available in the medical chart.
Definition of Terms
Previous use of aspirin was defined as an intake of a total of ≥500 mg of acetylsalicylic acid in the week preceding the qualifying clinical event. Secondary analyses were also performed by aspirin dose of <500, 500 to 1500, and >1500 mg/wk and by the duration of use, from any use in the previous week, to use between 1 week and 1 month, and to use for >1 month.
Unstable angina was defined as a clinical history of accelerating angina with chest pain occurring at rest or at minimal exercise or by the presence of prolonged chest pain (≥20 minutes in duration) with elevation of the CK-MB values to <36 IU/L, representing 1.5 times the upper limit of normal. Myocardial infarction was diagnosed when the prolonged chest pain was associated with a typical rise in cardiac enzymes with new Q waves or persisting ST-T changes on the ECG or with a positive pyrophosphate scan. In addition to new Q waves, an increasing R wave to an R/S ratio >1 in leads V1 and V2 and a left bundle-branch block with cardiac enzyme elevation to more than three times the upper limit of normal were considered Q-wave myocardial infarction; the others were classified as non–Q-wave infarction.
The 87 variables collected per patient were stored in a database (DBASE III) and analyzed with spss/pc version 4.0 statistical software. Intergroup comparisons involving quantitative variables were performed by two-tailed t tests for independent samples and by χ2 tests for categorical data. Odds ratios (ORs) and 95% CIs were calculated. Variables showing significant differences or marked trends in univariate analyses or believed to be of clinical or biological relevance were entered into a logistic regression analysis to evaluate their independent influence. The criteria used to enter or to remove a variable from the model were a probability of the Wald statistic of P=.05 and of P=.02, respectively.
The characteristics of the total population and per group of patients who took or did not take a total of ≥500 mg of aspirin in the previous week are shown in Table 1⇓. The study population was relatively young (median, 62 years), with a predominance of men (75%) and of patients with previous coronary artery disease (68%). Fewer than half of the patients (39.7%), however, were aspirin users. A previous cardiac history, denoted by significantly more frequent previous angina, myocardial infarction, balloon angioplasty, and bypass surgery procedures, was present in 88% of aspirin users but was also encountered in 56% of nonusers. The reason for not taking aspirin in the presence of previous coronary artery disease was no prescription in 80% of the patients and side effects in 20%, with gastric intolerance in 75% of those with side effects. Smoking habits differed significantly, with fewer aspirin patients being current smokers and more former smokers. Age, sex distribution, and risk factors were similar in aspirin users and nonusers. Other therapy at admission included nitrates, a β-blocker, or a calcium antagonist in one third of the total population.
The final discharge diagnosis of the qualifying event was unstable angina in 312 patients (58%) and myocardial infarction in 227 (42%). Unstable angina was manifested by crescendo angina in 34% of patients and by prolonged chest pain in 66%; ECG changes were present in 82% of patients. The myocardial infarctions were distributed as 27% non–Q-wave and 73% Q-wave infarction, 53% involving an inferior site, 44% an anterior site, and 3% an undetermined site.
Unstable Angina Versus Myocardial Infarction
The qualifying event resulted in unstable angina in 162 of the 214 aspirin users (76%) and in 150 of the 325 nonusers (46%). Conversely, myocardial infarction was the index event in 52 (24%) and 175 (54%) of patients, respectively (P<.0001) (Fig 1⇓). With aspirin use, the OR for developing myocardial infarction was 0.28 (95% CI, 0.10 to 0.41) and for unstable angina, 3.63 (95% CI, 2.44 to 5.42). Table 2⇓ compares the clinical characteristics of patients with unstable angina and of patients with myocardial infarction. Age and sex were similar. Unstable angina was associated with higher plasma cholesterol levels, and myocardial infarction with more frequent active smoking habit. Patients with unstable angina more often had a previous history of angina and of previous myocardial infarction. They were also more frequently taking an antianginal drug.
The association between the previous use of aspirin and a shift to more frequent unstable angina and less myocardial infarction was statistically significant in 20 of the 24 subgroups considered, including younger or older age, male or female sex, previous myocardial infarction or not, and use or nonuse of an antianginal drug. The association was nonsignificant in patients with previous angina, in smokers, and in patients taking a nitrate or a β-blocker. The ORs and CIs for the most relevant subgroups are shown in Fig 2⇓.
The clinical manifestation of unstable angina was unaffected by the previous use of aspirin. It had a predominant pattern of accelerated angina in 34% of aspirin users and in 33% of the nonusers and of prolonged rest pain in 67% and 64% of patients, respectively. ST-segment changes were also observed as frequently in aspirin users and nonusers, with ST-segment depression in 25% and in 22% and ST-segment elevation in 16% and 14%, respectively.
The multiple logistic regression retained the use of aspirin and a history of previous angina as the strongest predictors of the shift to more frequent unstable angina and less frequent myocardial infarction. Not smoking and higher cholesterol levels were also retained as independent predictors of the shift (Table 3⇓).
Effects of Duration of Previous Aspirin Use and of Dosage
Acute myocardial infarction was the clinical event in 43% of the 36 patients who had used aspirin only during the preceding week, in 25% of the 24 patients who had used it during the preceding month, and in 19% of the 152 patients who had used it for >1 month. Doses of aspirin of <500 mg/wk were associated with myocardial infarction in 47% of patients, doses of 500 to 1500 mg in 16%, and doses of ≥1500 mg in 23% (Fig 3⇓).
Noncardiac Chest Pain
Patients with a diagnosis of noncardiac chest pain at hospital discharge were excluded from the main analysis. These patients were less often aspirin users (26% versus 40% of included patients), were younger (58 versus 63 years old), were more frequently female (42% versus 25%), and less often had a prior history of previous angina (49% versus 61%). Risk factors, however, were similar. Including these patients in the analysis did not modify the results, with 22% of aspirin users and 46% of nonusers experiencing a myocardial infarction (P<.00006); previous aspirin use remained a strong predictor of unstable angina versus myocardial infarction by multivariate analysis (P=.0006).
Non–Q-Wave Versus Q-Wave Myocardial Infarction
Of the 227 patients with acute myocardial infarction, 62 (27%) had a non–Q-wave and 165 (63%) a Q-wave myocardial infarction. The characteristics of these patients are shown in Table 4⇓. Peak CK elevation was 1008±1123 IU/mL for non–Q-wave and 1918±1711 IU/mL for Q-wave infarction (P<.0001) and peak CK-MB elevation 100±81 versus 205±181 IU/mL (P<.00001). Peak CK and CK-MB values were smaller in patients with previous aspirin use: 1121±1199 versus 1827±1682 IU/mL, P=.006, and 115±101 versus 196± 177 IU/mL, respectively. Age, sex, and other risk factors were equally distributed among the two groups except for trends to less frequent active smoking and higher plasma cholesterol levels in non–Q-wave infarction, as was also observed in the comparison of unstable angina versus myocardial infarction. Although a previous myocardial infarction was present in the same number of patients, more patients with non–Q-wave infarction had previous angina. This was associated with more frequent use of nitrates and, to a lesser extent, of calcium antagonists and β-blockers.
The previous use of aspirin resulted in significantly more non–Q-wave infarction (38% versus 24%) and its nonuse in significantly more Q-wave infarction (76% versus 62%) (P<.05) (Fig 1⇑). The OR for developing a Q-wave versus a non–Q-wave myocardial infarction with aspirin use was 0.51 (95% CI, 0.25 to 1.03). The multivariate analysis retained the previous use of aspirin as the only independent predictor of non–Q-wave versus Q-wave infarction.
The present study was specifically designed to examine whether the previous use of aspirin influences the severity of manifestations of acute coronary artery syndromes, based on the well-documented benefits of the drug in the prevention of myocardial infarction, stroke, and death in cardiovascular disease.7 A large and consecutive series of patients was prospectively included in this ambidirectional study, with collection of antecedent data on medical history at entry and subsequent determination of outcome diagnosis. The results suggest that patients who developed a cardiac event while taking aspirin were much more likely to experience unstable angina than myocardial infarction and a non–Q-wave infarction rather than a Q-wave infarction. The apparent failure of aspirin in patients with acute coronary syndromes may thus be only relative or partial, since the drug still has the potential of modifying the natural history of the disease by producing a shift to less severe acute manifestation. No previous studies have specifically addressed this issue. In a small secondary analysis of the usefulness of fibrinolysis in unstable angina, Schreiber et al11 described more frequent three-vessel disease in 31 patients who developed unstable angina while on aspirin. Cohen et al,12 on the other hand, described no differences in clinical characteristics and in severity of coronary artery disease analyzed by quantitative coronary angiography in 93 patients with unstable angina or non–Q-wave myocardial infarction. The same group of investigators subsequently enrolled 358 such patients in a prospective trial of antithrombotic therapy.13 The admission diagnosis in this study was unstable angina in 45% of prior aspirin users and non–Q-wave myocardial infarction in only 17%, figures very similar to our observations. In the Physicians’ Health Study of primary prevention among 22 071 male physicians, aspirin compared with placebo reduced the incidence of myocardial infarction by 44%; the size of an infarction, when it occurred, its location, its ECG features, and the resulting ejection fraction, however, were not influenced by the previous use of aspirin.14 Patients catheterized in that study also had the same extent of coronary artery disease whether they received aspirin or placebo. The authors emphasized the lack of adequate information in some patients and of statistical power of the analysis; also, unstable angina was not considered in the diagnosis.
A bias toward more frequent coronary artery disease and possibly toward more severe disease, better care, and closer supervision in patients using aspirin cannot be avoided with our study design and certainly had an impact on the results. Patients known to suffer the disease are also more likely to use cardiac medications, and some may consult earlier, when symptoms first manifest. Indeed, in this study, patients with previous angina were more frequent aspirin users and patients with noncardiac chest pain less frequent users. The pattern of chest pain, however, is often much alike in non–Q-wave and Q-wave myocardial infarction. A large sample size was used in this study, with as much as 20% of patients with a previous myocardial infarction and 50% of patients with previous angina not using prophylactic aspirin; 12% of patients without any previous cardiac history, on the other hand, were using it. The relatively high rate of no aspirin use in patients with coronary artery disease may be surprising, but it is similar to that observed in other studies.13 The main reason was nonprescription, possibly because the beneficial effects of aspirin in secondary prevention of coronary artery disease were documented only recently, after the documentation of the presence of coronary artery disease in our patients; only 20% of patients were not taking aspirin because of intolerance. The multivariate analysis retained the previous use of aspirin as a powerful independent predictor of the occurrence of unstable angina versus myocardial infarction and of non–Q-wave myocardial infarction versus Q-wave infarction. Also, consistent results were observed in all subsets of patients analyzed, including the subsets of patients without a previous myocardial infarction and without previous angina and of patients not taking an antianginal drug; this could suggest that the shift observed toward less severe disease manifestation could also apply in primary prevention. The current worldwide trend toward increased use of aspirin in primary and secondary prevention will, in the future, hamper the reproduction of our observations.
Aspirin use was not verified in this study by objective assessment of the degree of inhibition of thromboxane A2 generation, and a recall bias could have been present in many patients. This bias, however, most likely did not influence the results, since ascertainment of aspirin intake was made very shortly after the event with an oriented and standard questionnaire by a study investigator; there is also no reason why the bias would have occurred more frequently in myocardial infarction then in unstable angina.
The different benefits observed with various doses of aspirin and with the duration of use before the events are of some interest. The data suggest that weekly doses of aspirin of 500 to 1000 mg are adequate and that maximal benefit is achieved after 1 week of administration. Doses of <500 mg/wk and use for <1 week were associated with lesser benefit. Caution is needed in the interpretation of these results, considering that the study was uncontrolled, that some data were collected retrospectively, and that the data apply to smaller subsets of patients. One explanation of the findings could be that the disease process was already engaged in many patients who received aspirin in the previous week only and that the drug was added too late; studies performed during the acute phase of unstable angina8 and myocardial infarction,10 however, have clearly demonstrated the benefits of aspirin. Alternative hypotheses could be other antithrombotic15 or anti-inflammatory effects of aspirin16 or delayed benefit on plaque progression17 or structure18 with more prolonged exposure. The Physicians’ Health Study, however, did not document benefit on prevention of future angina pectoris.19 Clinical studies have also now well established the effectiveness of small doses of aspirin: 30 mg,20 75 mg,2 3 9 and 325 mg on alternate days.4 Doses of aspirin of 325 mg 1 to 3 times per week or 4 to 6 times per week in the Nurses’ Health Study cohort were equally effective.21 Aspirin benefits are probably explained by its preventing thrombus formation by inhibition of the cyclooxygenase pathway; this is achieved within 60 minutes after a single dose of 100 mg or chronically with daily use of 30 mg.1 Cigarette smoking could modify the useful dose of aspirin because of increased platelet turnover. This could not be assessed in this study because only 10 of the 134 smokers used large doses of aspirin; the incidence of myocardial infarction versus unstable angina in these patients was 50% compared with 58% in smokers with any aspirin dose and 68% in smokers with no aspirin.
This study assumes a gradient in the severity of the manifestation of acute coronary artery disease from unstable angina to non–Q-wave myocardial infarction to Q-wave infarction. This assumption is strongly supported by pathophysiological considerations and clinical evidence. Transmural myocardial infarction is associated with a thrombotic occlusion in angiograms performed acutely in >90% of patients22 ; non–Q-wave infarction is associated with the presence of apparently nonocclusive thrombi in 47% of patients and unstable angina in 29%.23 This certainly suggests a stronger thrombogenic stimulation in Q-wave infarction and a lesser stimulation in unstable angina. Similarly, the impairment of left ventricular function is usually less in non–Q-wave infarction than in Q-wave infarction and is absent in unstable angina. Altogether, these observations support a gradient to a less severe thrombotic process from Q-wave to non–Q-wave infarction and unstable angina, a gradient favored by previous aspirin use. This original observation extends our knowledge on the role of aspirin in primary and secondary prevention to suggest that in addition to reducing the incidence of myocardial infarction by 40%,7 aspirin also converts 50% of potential myocardial infarction to unstable angina and 20% of Q-wave infarction to non–Q-wave infarction. These observations help define the role of aspirin in primary and secondary prevention and understand the modalities for its clinical benefit; they may, for example, partly explain the trend recently observed toward more admissions to Coronary Care Units for unstable angina than for myocardial infarction24 and possibly also contribute, along with other factors, to the continuing decreasing rates in cardiac mortality observed in many countries.25
This study was supported in part by grant F1S 93/1255 from the Fondo de Investigaciones Sanitarias de la Seguridad Social and Fundacion Uriach 1838. We acknowledge the helpful comments of Dr Permanyer-Miralda on the discussion of the epidemiological aspects of the study.
- Received March 9, 1995.
- Revision received April 24, 1995.
- Accepted April 25, 1995.
- Copyright © 1995 by American Heart Association
Juul-Möller S, Edvardsson N, Jahnmatz B, Rosen A, Sorensen S, Ömblus R, for the Swedish Angina Pectoris Aspirin Trial (SAPAT) Group. Double-blind trial of aspirin in primary prevention of myocardial infarction in patients with stable chronic angina pectoris. Lancet. 1992;340:1421-1425.
Peto R, Gray R, Collins R, Wheatley K, Hennekens C, Jamrozik K, Warlow C, Hafner B, Thompson E, Norton S, Gilliland J, Doll R. Randomized of prophylactic daily aspirin in British male doctors. Br Med J. 1988;296:313-316.
Antiplatelet Trialists’ Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J. 1988;296:320-331.
Antiplatelet Trialists’ Collaboration. Collaborative overview of randomized trials of antiplatelet therapy, 1: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Br Med J. 1994;308:81-106.
Cohen M, Xiong J, Parry G, Adams PC, Chamberlain D, Wieczorek I, Fox KAA, McBride R, Chesebro JM, Fuster J, and the Antithrombotic Therapy in Acute Coronary Syndromes Research Group. Prospective comparison of unstable angina versus non-Q wave myocardial infarction during antithrombotic therapy. J Am Coll Cardiol. 1993;22:1338-1343.
Ridker PM, Manson JE, Buring JE, Goldhaber SZ, Hennekens CH. Clinical characteristics of non-fatal myocardial infarction among individuals on prophylactic low-dose aspirin therapy. Circulation. 1991;84:708-711.
Hanson SR, Harker LA. Effects of platelet-modifying drugs on arterial thromboembolism in baboons: aspirin potentiates the antithrombotic actions of dipyridamole and sulfinpyrazone by mechanism(s) independent of cyclooxygenase inhibition. J Clin Invest. 1985;75:1591-1599.
Hansen GK. Immune antiinflammatory mechanism in the development of atherosclerosis. Br Heart J. 1993;69(suppl):S38-S41.
Ranke C, Hecker H, Creutzig A, Alexander K. Dose-dependent effect of aspirin on carotid atherosclerosis. Circulation. 1993;87:1873-1879.
Cleland JGF, Krikler DM. Modification of atherosclerosis by agents that do not lower cholesterol. Br Heart J. 1993;69(suppl):S54-S62.
The TIMI IIIA Investigators. Early effects of tissue-type plasminogen activator added to conventional therapy on the culprit coronary lesion in patients presenting with ischemic cardiac pain at rest: results of the Thrombolysis in Myocardial Ischemia (TIMI IIIA) trial. Circulation. 1993;87:38-52.
National Center for Health Statistics. Vital Statistics of the United States, 1988. DHHS publication (DHS)91-1101. Washington, DC: US Department of Health and Human Services; 1991.