The understanding of the pathogenesis of myocardial infarction has advanced substantially during the past 15 years. Pathologic, angiographic, and angioscopic observations have led to the concept that coronary occlusion begins with the rupture of a lipid-rich atherosclerotic plaque, which leads to the formation of a thrombus consisting of platelets, fibrin, and other blood elements and to local vasoconstriction [5]. Other observations in animal models and early clinical trials have suggested that reperfusion therapy was feasible and could limit both the size of the infarction and mortality [6].

In the past decade, collaborative research has developed a large and compelling database supporting the survival benefit of antithrombotic (thrombolytic [fibrinolytic] and antiplatelet) therapies in acute myocardial infarction. Thrombolytic therapy was evaluated in nine randomized, controlled trials, each of which consisted of more than 1000 patients suspected of having had acute myocardial infarction (58 600 patients total). An overview of these trials [7] conclusively showed the beneficial effect of therapy in patients presenting with ST-segment elevation or bundle-branch block (respective relative mortality reductions at 5 weeks, 21% [P < 0.001] and 25% [P < 0.01]). Benefit was seen regardless of age, sex, blood pressure, heart rate, or history of myocardial infarction or diabetes and was greater the earlier treatment was begun. An international collaborative study involving more than 17 000 cases of suspected acute myocardial infarction (Second International Study of Infarct Survival [ISIS-2]) [8] also showed the important role of antiplatelet therapy with aspirin. When aspirin was added to streptokinase, the reduction in odds of vascular death increased from 25% to 42%. Moreover, when both drugs were given within 4 hours of symptom onset, the odds of death were reduced by 53% (P < 0.001), a dramatic result. Aspirin also reduced (by 49%) the incidence of early recurrent nonfatal reinfarction.

On the basis of this experience, a 1990 task force of the American College of Cardiology and American Heart Association designated thrombolytic therapy as a class I indication (therapy usually indicated and considered effective) in patients younger than 70 years of age who had no contraindications, presented with chest pain consistent with acute myocardial infarction, had ST-segment elevation, and could receive therapy within 6 hours of pain onset [9]. (The therapeutic window has since been extended to 12 hours [7].) The task force recommended aspirin as a class I indication in patients of all ages who were suspected of having had acute myocardial infarction and had no contraindications. According to the guidelines, aspirin therapy should be started immediately and continued indefinitely (at a dose of 162 to 325 mg/d).

Long-term therapy after myocardial infarction has also been shown to be of benefit. In an overview of 10 randomized, placebo-controlled trials of antiplatelet agents (primarily aspirin), a 25% reduction in vascular events was noted (P < 0.001) (mortality reduction, 13%; rate of nonfatal reinfarction, 31%; rate of nonfatal stroke, 42%) [10]. These data provide a strong rationale for the long-term (indefinite) use of aspirin after myocardial infarction in doses as small as 80 to 325 mg/d.

The use of thrombolytic and antiplatelet therapy in clinical practice has been increasing. According to one recent U.S. registry report [11], as many as 70% of patients who have had ST-segment elevation myocardial infarction received reperfusion therapy, and, overall, only 12% of patients with no contraindications did not receive thrombolysis or primary angioplasty. Recent in-hospital use of aspirin has exceeded 90% among thrombolysis candidates but has been lower among other patients and long-term users (70% to 75%).

Although these therapies are broadly used in younger patients, acceptance of them has been more gradual in older patients who have had myocardial infarction. Treatment decisions must always balance benefit and risk [12]. The hemorrhagic consequences of thrombolytic therapy may be severe (death or disability), and the risk for intracerebral hemorrhage increases substantially (more than two-fold) with age [13], from 0.5% overall to as much as 1.5% to 4% or more with some therapeutic regimens in persons older than 70 to 75 years of age [7, 13-17]. When considering thrombolytic therapy for elderly patients, physicians are well aware of the hemorrhage-related death and disability caused by treatment but are uncertain about which patients have specifically benefited: Primum non nocere. Unfortunately, evidence for clear benefit in elderly patients, despite their high baseline risk, has been slow in coming. Beginning with the first major mortality trial [18], a clear mortality benefit in the elderly has been difficult to show. Indeed, many early trials excluded patients older than 70 to 75 years of age on the basis of safety concerns, and initial Food and Drug Administration-approved labeling for use of thrombolytic agents in myocardial infarction contained only indications for persons 75 years of age and younger. In the 1990 guidelines [9], the task force recommended thrombolytic therapy as a class IIA indication (of uncertain efficacy, with weight of evidence in favor of usefulness) for persons aged 70 to 75 years and as a class IIB indication (acceptable, but not well established by evidence) for those older than 75 years of age. For younger patients, the task force considered thrombolytic therapy to be a class I indication.

In their study, the results of which are published in this issue, Gurwitz and colleagues [3] examined recent trends in the use of thrombolytic therapy in elderly patients who have had acute myocardial infarction and determined the extent to which late presentation and failure to meet electrocardiographic criteria (that is, ST-segment elevation) account for differences in use. Their experience derives from more than 1200 hospitals and 350 000 patients. Thrombolytic therapy was indeed used less frequently in the elderly: 19% of those aged 75 to 84 years and 7% of those 85 years of age and older (compared with 33% to 51% for younger patients). However, temporal increases in use were seen in these oldest two age groups, averaging 34% and 73%, respectively, from 1990 to 1994. As expected, increasing age at myocardial infarction was associated with female sex, late presentation (> 6 hours after myocardial infarction), and entry electrocardiogram without ST-segment elevation. After adjustment for baseline factors, the odds ratio for treatment was 0.3 for patients aged 75 to 84 years and 0.1 for those 85 years of age and older, suggesting that age itself did affect treatment decisions.

Limitations of Gurwitz and colleagues' study [3] include the fact that no information was provided on either the percentage of patients with contraindications to therapy (such as previous stroke, hemorrhage, or hypertension) or the benefit-risk ratio of therapy in the very elderly. Therefore, the "optimal" percentage that should be treated remains unclear. The study does, however, document an increasing trend among physicians toward giving thrombolytic agents to elderly patients who have had myocardial infarction.

A recent overview [7] has enriched our understanding of the benefit of thrombolytic therapy in the elderly. Among 6000 patients 75 years of age or older, mortality after 5 weeks was reduced modestly, from 25.3% to 24.3% (a saving of 10 lives per 1000 persons treated). These data can be used either to support (smaller proportional benefit but an absolute benefit similar to that for persons younger than 55 years of age) or to argue against the use of therapy in the elderly (P equals not significant compared with control and substantially less than the 27 lives saved per 1000 persons for persons aged 65 to 74 years). Smaller benefits coupled with higher complication rates do suggest that a more cautious approach be used in the very elderly. Careful assessment of the potential for both risk (such as for cerebral hemorrhage) and benefit (for example, if a patient is examined within 4 to 6 hours of having ST-segment elevation myocardial infarction) should be part of the decision making. Consideration of body weight and risk for bleeding may lead to modification of the type and dose of the thrombolytic agent and heparin. If the anticipated benefit is small (small, late, or evolved myocardial infarction) and the risk (such as risk for cerebral hemorrhage) is great, thrombolytic therapy may indeed be inappropriate [19], whereas a favorable risk–benefit profile should weigh in favor of therapy regardless of age [19]. Thus, it appears reasonable to modify the 1990 guidelines for treating myocardial infarction to recommend thrombolytic therapy as a class I indication for persons 75 years of age and younger and as a class IIA indication for persons older than 75 years of age. Given the importance of death from myocardial infarction in the elderly, further progress is needed, both through education and additional clinical trials.

Also in this issue, Krumholz and colleagues [4] explore a second aspect of therapy for myocardial infarction in the elderly: the prescription of aspirin for secondary prevention. Their study sample consisted of 5490 consecutive Medicare beneficiaries who survived myocardial infarction, were hospitalized from June 1992 to February 1993, and had no contraindications to aspirin. Krumholz and colleagues report that 76% of patients were discharged with a prescription for long-term aspirin therapy. According to multivariable analyses, aspirin was prescribed more often for persons with indicators of better overall health. Persons prescribed aspirin were found to be at lower risk for subsequent death (odds ratio, 0.79), even after adjustment for their better baseline indicators of health. Krumholz and associates conclude that prescription of aspirin to an additional 25% of elderly survivors of acute myocardial infarction appears feasible and may be an excellent opportunity to provide secondary prevention at low cost and morbidity. Given the relatively low adverse potential of aspirin compared with its potential benefits, it is hard to argue with these conclusions. As Maggioni and colleagues note [20], mortality after discharge (as well as before) is substantially greater in the elderly than in younger patients (a threefold increase in mortality at 6 months for persons older than 80 years of age compared with those 66 to 70 years of age). If the proportional benefit of aspirin is similar in the elderly, then the opportunity for absolute benefit would be great. Indeed, in the ISIS-2 trial [8], risk within 35 days of myocardial infarction was reduced 21% in persons older than 70 years of age, a percentage similar to the 18% reduction for persons younger than 60 years of age. The absolute benefit was 47 lives saved per 100 000 patients treated compared with 10 lives saved per 100 000 patients, a four- to fivefold increase.

Some limitations of Krumholz and colleagues' study [4] should be pointed out. Prescription data are provided only for persons older than 65 years of age, and these suggest only a modest prescription bias with advancing age (80% of those aged 65 to 74 years compared with 67% of those 85 years of age and older). In addition, unidentified and identified patient characteristics may have contributed to differences in survival outcomes by aspirin prescription and would not be adjusted for in multivariable analyses; because sicker patients may be less likely to be prescribed aspirin, estimates of mortality benefit are probably an overestimates. Randomized clinical trials provide a less biased estimate of mortality benefit. However, even assuming that the treatment-related reduction in mortality from vascular causes is smaller (that is, 13%, as achieved in randomized trials [10]), the opportunity for benefit is still substantial. Finally, the study leaves open the question of why aspirin was not prescribed in the 25% of "eligible" patients. A prospective study to examine these undefined factors is needed. Given the suboptimal use of aspirin in other age groups, further efforts at understanding aspirin underuse and educating physicians to its potential benefits in all age groups appear appropriate.

Directing attention to health problems and treatment in the elderly is timely, given the aging of the U.S. population and the fact that the effect of cardiovascular diseases increases with age. The studies of Gurwitz [3] and Krumholz [4] and their colleagues represent important benchmarks in defining our current treatment approach and suggest opportunities for future information gathering, education, and treatment. Wider, appropriate application of therapies such as thrombolysis and aspirin in the elderly have substantial potential to further the beneficial cardiovascular mortality trends already underway [1, 2].