RSS Feeds
Coronary Angiography and Angioplasty after Acute Myocardial Infarction
- David W. Bates, MD, MSc;
- Elizabeth Miller, BS;
- Steven J. Bernstein, MD;
- Paul J. Hauptman, MD; and
- Lucian L. Leape, MD
- From Brigham and Women's Hospital and Harvard School of Public Health, Boston, Massachusetts; and University of Michigan Medical Center, Ann Arbor, Michigan. Acknowledgments: The authors thank George Beller, MD, Gottlieb C. Friesinger, MD, Spencer B. King 3d, MD, Thomas J. Ryan, MD, and Eric J. Topol, MD, for their comments on an earlier version of this manuscript. Grant Support: By research grant R01-HS08071-02 from the Agency for Health Care Policy and Research. Requests for Reprints: David W. Bates, MD, MSc, Division of General Medicine and Primary Care, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115. Current Author Addresses: Dr. Bates and Ms. Miller: Division of General Medicine and Primary Care, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115.
Abstract
Purpose: To assess the data that support the use of coronary angiography and angioplasty after acute myocardial infarction, that identify the risks of these procedures, and that analyze their use and costs.
Data Sources: English-language articles published between 1970 and June 1995 identified through a search of the MEDLINE database.
Study Selection: Studies that contained information about benefits, risks, use, and costs of coronary angiography and angioplasty after acute myocardial infarction.
Data Extraction: Descriptive and analytic data from each study were collected.
Data Synthesis: The outcome for patients who have complications of myocardial infarction (such as shock) is poor. Such patients usually undergo angiography, although the evidence that supports this practice is weak. Preliminary data suggest that patients who immediately have angiography and angioplasty after acute myocardial infarction have better outcomes than do patients who receive thrombolytic therapy with angioplasty only for specific indications in experienced centers. After the acute phase of myocardial infarction, patients who have noninvasive evidence of persistent or recurrent ischemia are believed to benefit from angiography. In the remaining patients, however, angiography after myocardial infarction has not been shown to be beneficial. Coronary angiography is done in 30% to 81% of patients after acute myocardial infarction in different settings and regions; for many of these patients, the benefit is questionable. Better outcomes are not always associated with more frequent use of the procedure. In the United States, catheterizations after myocardial infarction cost approximately $1 billion per year.
Conclusions: Although many patients benefit from angiography and angioplasty after myocardial infarction, others probably do not. Substantial resources are at stake.
Coronary angiography may be done during or after myocardial infarction for diagnostic or therapeutic purposes. The procedures are usually done to identify coronary disease that is amenable to coronary revascularization. Therefore, the randomized trials that have evaluated the benefit of coronary angiography after myocardial infarction have focused on percutaneous transluminal coronary angioplasty or coronary artery bypass graft surgery rather than on whether to do angiography [1].
Nevertheless, it may be equally important to evaluate the diagnostic process [2] because interventions closely follow diagnostic tests in what has been called a “clinical cascade” [3]. Verrilli and Welch [2] found strong correlations between rates of diagnostic tests and subsequent use of therapeutic procedures. For example, the correlation between cardiac catheterization and coronary revascularization was 0.91, which suggests that decisions about diagnostic tests may be as important as decisions about coronary therapeutic procedures.
Whether angiography after myocardial infarction is appropriate depends on the efficacy of subsequent interventions. Studies have evaluated the use of angioplasty as an alternative to thrombolytic therapy (primary angioplasty) or in conjunction with thrombolytic therapy at various times: immediately (as soon as possible after onset of symptoms), early (within several hours after myocardial infarction to any time before discharge), or delayed (after discharge but within 12 weeks of myocardial infarction) [4]. In the period immediately after myocardial infarction, angiography can be done routinely (that is, on patients in whom revascularization is not contraindicated) or after the development of such specific complications as recurrent chest pain or shock. After discharge, angiography can also be done routinely, in subsets of symptomatic patients, or when results of noninvasive tests are positive. Strategies in which angiography and, if feasible, angioplasty were done immediately have usually been compared with strategies in which they were done before discharge, within the next several weeks, or not routinely.
The appropriateness of angiography also depends on whether thrombolytic agents have been or can be given and the length of time that has elapsed since the onset of symptoms (Table 1). Although thrombolytic therapy has been shown to be effective for the treatment of myocardial infarction in several clinical situations, it is contraindicated in some patients and its effectiveness depends on how soon it is given after myocardial infarction [41, 42]. Thrombolytic therapy is most effective if it is administered within 0 to 6 hours after the onset of symptoms [42]. Although the benefit of thrombolytic therapy decreases over time, this therapy remains effective for at least 12 hours [42-44].
We analyze data on the use of coronary angiography after acute myocardial infarction; we also assess the use of angioplasty. The report is structured on the basis of the timing of the procedures after myocardial infarction and whether thrombolytic agents have been or can be given. We also include information on the risks, costs, and use of angiography and angioplasty.
Methods
We included information from articles that were published from 1970 through June 1995. For the period from 1970 to 1989, we relied heavily on a literature review that had been prepared for an expert panel on coronary angiography [45]. We focused on studies that contained information on benefits, risks, use, and costs of coronary angiography and angioplasty after acute myocardial infarction. Studies were categorized by whether thrombolytic agents had been given and by the amount of time that had elapsed after myocardial infarction (0 to 6 hours, 6 to 12 hours, more than 12 hours but before discharge, and after discharge but before 12 weeks). Because the effectiveness of angiography in patients who have had myocardial infarction is largely determined by the effectiveness of reestablishing coronary patency, we also reviewed information on the efficacy of thrombolytic therapy and coronary artery bypass graft surgery.
We identified the articles by searching MEDLINE. The search included only English-language articles that had been published from 1970 through June 1995. To identify other articles, the bibliographies of all articles that were identified in the MEDLINE search were evaluated. The members of an expert panel on coronary angiography evaluated the bibliography and literature review for accuracy and completeness. The panel, which was part of the Acute Myocardial Infarction Patient Outcome Research Team (PORT) study, had convened to update the ratings of appropriateness for angiography in patients who had had acute myocardial infarction. Our search terms included the following Medical Subject Headings: myocardial infarction, indications; angiography; angioplasty; angioplasty, transluminal, percutaneous coronary; thrombolytic therapy; coronary artery bypass; and randomized, controlled.
Data Synthesis
Timing of Angiography and Angioplasty and Relation to Thrombolytic Therapy
Within 6 Hours of Myocardial Infarction, No Thrombolytic Therapy
Through 1995, seven randomized trials had compared immediate angiography and angioplasty without thrombolytic therapy (primary angioplasty) with thrombolytic therapy in acute myocardial infarction [5-12]. In a multicenter study, Grines and colleagues [5] found that in-hospital mortality rates for patients who had primary angioplasty (2.6%) were lower than those for patients who received thrombolytic therapy (6.5%); however, this difference was not statistically significant. The difference in survival was largest in the group that was at highest risk. After 6 months, 8.5% of the group receiving primary angioplasty and 16.8% of the group receiving thrombolytic therapy had had reinfarction or had died. In a smaller, single-center study, Zijlstra and colleagues [6] found that primary angioplasty resulted in a higher infarct-related artery patency rate, better left ventricular function, and less recurrent ischemia and myocardial infarction than did intravenous streptokinase. In preliminary results of the angioplasty substudy of the GUSTO (Global Use of Strategies to Open Occluded Coronary Arteries) study [46], patients who had had angioplasty had a lower rate for the end points of death, disabling stroke, or reinfarction (9.6%) than did patients who had received the most effective thrombolytic therapy regimen (13.1%) (P = 0.06).
In a meta-analysis that included all seven trials that compared primary angioplasty with thrombolytic therapy (except the recent GUSTO study), Michels and Yusuf [4] reported lower 6-week mortality rates in patients who were treated with angioplasty (odds ratio, 0.56 [95% CI, 0.33 to 0.94]) but no significant difference in 1-year mortality rates (odds ratio, 0.91 [CI, 0.42 to 2.00]). The results of trials in which death or nonfatal reinfarction was the outcome were similar (Figure 1, top left). Data from longer follow-up periods are not yet available. Despite these promising results, access to primary angioplasty is logistically difficult; only about 15% of hospitals have rapidly available, experienced angioplasty teams [47]. Moreover, in a recent observational study that was done in a community setting and compared the outcome of 1050 patients who received primary angioplasty with that of 2095 patients who received thrombolytic therapy, Every and coworkers [48] found no difference in mortality rates during hospitalization or during long-term follow-up. These researchers also noted that costs were lower and that angioplasty was done less frequently in the thrombolytic therapy group at discharge and after 3 years of follow-up.
Thrombolytic therapy is strongly contraindicated in as many as 20% of patients who have had acute myocardial infarction [14]. No randomized trials have compared angioplasty with noninvasive therapy in patients who have strong contraindications to thrombolytic agents; however, several case series report good primary success rates for angioplasty in such patients [13-15]. Two studies also report good 1-year survival rates [14, 15].
Two controlled trials of early coronary artery bypass graft surgery from the pre-thrombolytic era found that patients who were treated surgically had a lower 30-day in-hospital mortality rate (2.2%) than did those who were treated medically (11.1%) [49, 50]. As with angioplasty, however, few hospitals can provide immediate bypass surgery.
Within 6 Hours of Myocardial Infarction, after Thrombolytic Therapy
Most studies of angiography and angioplasty after myocardial infarction have focused on the use of these procedures after thrombolysis [16-2427-29, 37]. Many studies have shown that thrombolytic therapy can preserve myocardium and improve survival rates [42-44, 51-54]. Overall, the evidence suggests that after myocardial infarction has been treated with thrombolytic agents, routine immediate angiography and routine angioplasty are associated with no significant benefit and may be harmful. However, immediate angiography and angioplasty may be useful in specific subgroups that have such complications as angina and shock after myocardial infarction.
The TIMI (Thrombolysis in Myocardial Infarction) II-A study [32, 33] addressed the use of angiography after thrombolytic therapy in patients who had had myocardial infarction. This study compared three treatment strategies: immediate invasive, delayed invasive, and conservative. With the immediate invasive strategy, angiography and angioplasty (if indicated) were done as early as possible. With the delayed invasive strategy, routine angiography was done between 18 and 48 hours after thrombolytic therapy. With the conservative strategy, angiography and angioplasty (if it was indicated) were done only if the patient developed spontaneous or exercise-induced ischemia. The 24-hour complication rate for patients who received the immediate invasive strategy was significantly higher (8.5%) than that for patients who received the delayed invasive strategy (4.6%); no benefit for other major end points was shown. Patients who received the conservative strategy had outcomes that were equal to those for patients who received the delayed invasive strategy for major outcomes: 6-week survival (95.3% in the conservative group compared with 94.8% in the delayed invasive group), absence of reinfarction (93.5% in the conservative group compared with 93.1% in the delayed invasive group), and ejection fraction at discharge (49.9% in the conservative group compared with 50.5% in the delayed invasive group); therapy for patients in the conservative group was also less expensive.
The SWIFT (Should We Intervene Following Thrombolysis) trial [29] compared an invasive strategy with a conservative approach after thrombolytic therapy. The invasive strategy was early elective angioplasty (within 48 hours). Patients in the conservative group had angiography only if they had recurrent ischemic symptoms or positive results on exercise tests. At 3 months, mortality rates were similar between the conservative group (3.2%) and the invasive group (4.8%); ejection fraction was also similar. Thus, the results of the SWIFT trial confirmed those of the TIMI II-A study [32, 33].
In the first TAMI (Thrombolysis and Angioplasty in Myocardial Infarction) trial [20], patients who were randomly assigned to receive immediate angioplasty had a higher mortality rate (4%) than patients who were assigned to the delayed strategy (1%). However, the TAMI-5 study showed that an aggressive strategy of catheterization and selective angioplasty had good results [34, 35]. The aggressive strategy was associated with improved regional wall motion in the infarct region (P = 0.004) and fewer adverse outcomes (P = 0.004).
Two small trials of rescue angioplasty have been reported [30, 31]. In these trials, patients who had persistent occlusion of the infarction-related artery after thrombolytic therapy were randomly assigned to angioplasty or continued medical management. When Michels and Yusuf [4] combined the data from these trials, trends reflecting improved 6-week mortality rates (odds ratio, 0.38 [CI, 0.13 to 1.06]) and 1-year mortality rates (odds ratio, 0.17 [CI, 0.02 to 1.15]) were seen.
Thus, the data suggest that routine, immediate angioplasty done after thrombolytic therapy has little benefit and may be harmful. Rescue angioplasty seems promising, but not many patients have been studied.
Six to 12 Hours after Myocardial Infarction, No Thrombolytic Therapy
No randomized trials of angiography or angioplasty have focused specifically on this period.
Six to 12 Hours after Myocardial Infarction, Thrombolytic Therapy
Two trials [43, 44] reported an advantage for thrombolytic therapy when it is administered 6 to 12 hours after myocardial infarction, but the benefit is smaller than if this therapy is administered 0 to 6 hours after myocardial infarction. No trials examining angiography or angioplasty have focused specifically on this period.
More Than 12 Hours after Myocardial Infarction but before Discharge, No Thrombolytic Therapy
No recent randomized trials on angiography or angioplasty specifically address this group.
More Than 12 Hours after Myocardial Infarction but before Discharge, after Thrombolytic Therapy
Five trials have examined angioplasty [22-2427-29, 37] done more than 12 hours after myocardial infarction in patients who received thrombolytic therapy. As noted earlier, the TIMI II-A and SWIFT trials (in which such patients constituted a majority) found that the clinical outcomes of conservative strategies in which angiography was reserved for patients who had recurrent symptoms or positive results on exercise tests were similar to the outcomes of aggressive strategies but were less expensive [29, 32, 33]. Efforts to identify subgroups that are at high risk have met with only limited success [35, 55]. When Michels and Yusuf [4] combined the data from all five trials, conservative and invasive strategies did not differ in 1) mortality rate or 2) nonfatal reinfarction or mortality rate at either 6 weeks or 1 year.
After Discharge from the Hospital but before 12 Weeks
Three trials [38-40] compared a strategy of routine catheterization (catheterization in all patients) and angioplasty (if feasible) done 2 to 14 days after thrombolytic therapy with a strategy of thrombolytic therapy alone [39, 40] followed by angioplasty that was done only in patients who had ischemia and stenosis of 50% or greater [38]. Some of the patients had not been discharged. Michels and Yusuf [4] combined the results of these trials. Compared with the conservative group, the invasive group showed a trend toward increased mortality rates at 6 weeks (odds ratio, 1.33 [CI, 0.49 to 3.63]) and significantly increased mortality rates at 1 year (odds ratio, 6.79 [CI, 1.32 to 35.03]).
After a patient is discharged from the hospital, reasons for doing angiography change. Factors that predict higher morbidity and mortality rates are the presence of recurrent or persistent chest pain, extent of coronary disease, results of such noninvasive tests as exercise tests and nuclear imaging studies, and left ventricular ejection fraction [56-74].
No trials have addressed the efficacy of routine angiography and angioplasty after discharge from the hospital for patients who did not receive thrombolytic therapy. Although some evidence supports the efficacy of angiography and angioplasty in patients who have angina or myocardial ischemia after myocardial infarction, no data suggest that angiography and angioplasty are effective in patients who do not have these conditions [45].
Results of Trials of Percutaneous Transluminal Angioplasty after Thrombolytic Therapy
To ascertain whether individual trials of angioplasty lack the power to determine the effectiveness of angioplasty, Michels and Yusuf [4] did a metaanalysis of trials of angioplasty that had been done for acute myocardial infarction at various times after thrombolytic therapy [16-2427-29, 37]. When the results for immediate, early, delayed, and rescue angioplasty were compared with those for no angioplasty, 6-week mortality rates, 1-year mortality rates, and 6-week and 1-year mortality or nonfatal reinfarction rates were similar (aggressive angioplasty, Figure 1, top left). Comparisons between routine angioplasty (regardless of the presence of ischemia) or elective angioplasty (angioplasty done if it was medically indicated, such as for ischemia) without angioplasty also showed no difference in 6-week and 1-year mortality rates and mortality or nonfatal reinfarction rates (Figure 1, bottom). The few data that are currently available show no 5-year survival benefit from angioplasty done after thrombolytic therapy [4].
Angiography in Patients Who Have Complications of Myocardial Infarction
Complications of myocardial infarction that may increase the benefit of angiography and revascularization include persistent chest pain, mitral regurgitation, ventricular septal defect, shock, and persistent pulmonary edema.
Persistent Chest Pain
Patients who develop persistent chest pain after myocardial infarction have a greater risk for extension of the infarction and death than other patients [75-77]. In major randomized trials, patients with chest pain have usually had angiography. Although no data from randomized trials have proven its efficacy, angiography is probably effective in this setting [78].
Noninvasive Evidence of Mitral Regurgitation or Ventricular Septal Defect
Cardiac catheterization is useful for assessing the severity of mitral regurgitation or ventricular septal defect if surgery is being considered [78, 79], and angiography is almost always done at the same time to assess the severity of coronary disease. No randomized trials have addressed the efficacy of angiography in patients who have these complications.
Shock or Persistent Pulmonary Edema
Mortality rates are high in patients who have been in shock or have had persistent pulmonary edema after myocardial infarction [26, 32, 33, 80]. Because these patients often have poor outcomes and represent a small proportion of all patients, their rates of complication are often reported together. Patients with shock have been excluded from many trials [34]. Patients with shock or pulmonary edema were included in the TIMI-II trial, but the percentages of patients were low: One percent of patients had had pulmonary edema, and 1.5% had been in shock [26]. Outcomes for such patients did not differ from those for other patients in the TIMI-II trial, regardless of the management strategy. However, the power of these results was limited. The outcomes with conventional and thrombolytic therapy are often poor in patients with shock or pulmonary edema. If direct angioplasty is done, however, survival rates for these patients are better than those for historical controls [81].
Other Factors That Affect the Decision To Do Angiography
Many factors are associated with worse outcomes after myocardial infarction. These factors include worse left ventricular function, increased age, presence of three-vessel coronary disease, previous myocardial infarction, non-Q-wave infarction, female sex, history of smoking, and presence of diabetes [82]. Whether to do angiography partly depends on the factors that are present in a given patient. However, no widely used scale similar to the Parsonnet Index (a validated index that summarizes risk factors to predict the outcome of coronary artery bypass surgery) is available to summarize risk factors and to predict complications of angiography or benefits of revascularization [83]. Kimmel and associates [84] developed and validated an index to predict major complications of angioplasty (emergency coronary bypass surgery, myocardial infarction, or death). Risk factors in the index include multivessel disease, unstable angina, myocardial infarction within 24 hours (odds ratio, 2.2), type C lesion or left main angioplasty, shock, age, and aortic valve disease. The index stratifies patients into groups with low (complication rate, 1.3%), moderate (2.8%), high (12.7%), and very high (29.7%) risk. We present data on how factors associated with poor postinfarction prognosis affect outcomes after myocardial infarction. Patients whose outcomes are predicted to be worse can generally be expected to respond better to aggressive intervention.
Severe Left Ventricular Dysfunction
Ejection fraction is a strong predictor of early and long-term outcome after myocardial infarction [70, 85-87]. Ejection fraction independently predicted in-hospital mortality in the TAMI study [86]. As shown in a study by the European Cooperative Study Group [87], the 1-year mortality rate was 8.2% in patients who had an ejection fraction less than 40% but was only 1.8% in the group with an ejection fraction greater than 40%. Angiography is often done to identify patients who have decreased left ventricular function and three-vessel disease because such patients were shown to have improved survival rates after coronary artery bypass grafting in the Coronary Artery Surgery Study [88]. However, this study did not specifically address patients who had recently had myocardial infarction.
Age
Increased age was associated with higher mortality rates after myocardial infarction in the TIMI-II and TAMI trials as well as several other studies [86, 89, 90]. Udvarhelyi and colleagues [91] found that 30-day mortality rates increased from 19% for the group of patients who were 65 to 74 years of age to 38% for patients who were 85 years of age or older. Elderly patients have often been excluded from trials of thrombolytic therapy because they are presumed to have a greater risk for bleeding [81]. Several age cutoffs have been used, but 70 or 75 years of age is the most frequently used cutoff. However, elderly patients probably benefit from thrombolytic therapy even though their risk is greater. In the primary angioplasty trials, benefits that were seen as a result of angioplasty were more marked in elderly patients [5]. Compared with thrombolytic therapy, direct angioplasty has not been associated with a greater risk for bleeding complications in elderly patients, although death associated with angioplasty is increased [81].
Stress-Induced Myocardial Ischemia
The TIMI-II and SWIFT trials found that when patients were stratified by risk according to the results of noninvasive studies, outcomes were equally good in conservative and invasive treatment groups. However, functional testing in patients who have received thrombolysis may not be as sensitive for identifying a myocardium with reversible injury as it is in patients who have not received thrombolysis [79]. This concern is supported by studies that found that functional testing after thrombolysis did not predict reinfarction or death [18, 79, 92].
Non-Q-wave Myocardial Infarction
Studies from the pre-thrombolytic era showed that patients who had had non-Q-wave infarctions had high rates of recurrent myocardial infarction [93]. In the TIMI-II study [93], 29% of patients had had non-Q-wave infarctions, and no differences were seen in death or reinfarction rates between the aggressive and conservative post-thrombolytic strategies. A main goal of the TIMI-IIIB study [94] was to compare an early invasive strategy with a more conservative strategy in patients who had had non-Q-wave myocardial infarction or unstable angina. At 1 year, no difference in death or nonfatal infarction was seen between groups. However, angioplasty was done more frequently in the early invasive group (39%) than in the conservative group (32%) (P < 0.001), and more patients from the conservative group were subsequently hospitalized (33%) than were those from the early invasive group (26%) (P < 0.05). Outcomes for the patients who had had non-Q-wave infarctions were generally similar in the invasive and conservative groups. However, 73% of patients in the conservative group had angiography within 1 year. Taken together, these data suggest that either a conservative or an aggressive approach may be acceptable, depending on the preference of the patient.
Previous Myocardial Infarction
In the TIMI-II trial [26], patients who had previously had myocardial infarction had a higher 42-day mortality rate (8.8%) than those who had not previously had infarction (4.3%) (P < 0.001). However, this trial was a subgroup analysis; much of this difference may have been the result of a lower ejection fraction in patients who had previously had myocardial infarction (42%) than in those who had not (48%) (P < 0.001) and a higher prevalence of multivessel coronary disease in patients who had previously had myocardial infarction (60%) than in those who had not (28%).
Complications of Coronary Angiography and Angioplasty
Complications of coronary angiography include death, myocardial infarction, stroke, ventricular arrhythmia, vascular complications, and contrast reactions. When Bernstein and colleagues [45] combined the results of reports about complications for patients who have had angiography, they found rates of 0.1% for death, 0.07% for myocardial infarction and stroke, 0.47% for ventricular arrhythmia, and 0.59% for vascular complications.
For angioplasty, the usual definition of major complications includes myocardial infarction, coronary complications leading to bypass surgery, and death [95]. Combining data from reports published in the late 1980s, Hilborne and coworkers [95] reported myocardial infarction rates of 3% to 5% (with transmural infarction rates of 1% to 2%), mortality rates of approximately 1%, and emergency surgery rates of 2% to 4%. These rates have tended to decrease over time [96]; recent rates reported from large series are 0.9% to 1% for death, 1.5% to 5.2% for myocardial infarction, and 1.0% to 2.7% for emergency bypass surgery [97, 98].
Patients who have angioplasty or angiography soon after myocardial infarction have a higher risk for complications than do other patients. For angioplasty, Kimmel and associates [84] found the multivariate odds ratio for myocardial infarction within 24 hours to be 2.2 (CI, 1.25 to 3.86). However, for myocardial infarction within the previous 1 to 14 days, the odds ratio was only 0.73 (CI, 0.46 to 1.16). For angiography (Table 2), all available estimates have wide CIs because few studies have reported each outcome of angiography [45]. Five studies have examined angiography that was done soon after myocardial infarction (generally 24 to 72 hours); most of these studies were done in the early 1980s and reported complications after angiography [18, 99-102]. The mortality rate related to angiography was 0.97%, 10 times that for all patients; the rate of ventricular arrhythmia was 9.3%, 20 times that for all patients. For angiography that was done 1 to 8 weeks after myocardial infarction, the mortality rate was 0.22%, the myocardial infarction rate was 0.83%, the stroke rate was 0.53%, and the ventricular arrhythmia rate was 1.06%. However, these data were obtained from fewer than 1000 patients. The rates are higher than those reported for catheterizations in general; the mortality rate was twice as high, the myocardial infarction rate was eight times higher, and the stroke rate was five times higher. However, most of these data are now more than 10 years old. Because catheterization techniques have improved, absolute rates of complications may currently be lower. However, differences between risk after recent myocardial infarction and risk after routine catheterization probably persist.
Use of Coronary Angiography and Angioplasty
In 1992, 1 084 000 inpatient cardiac catheterizations were done in the United States, making this the second most frequent surgical procedure done in patients who were discharged after short hospital stays [103]. In 1992, 369 000 angioplasties were done in this group [103]. In 1994, about 10% of catheterizations were done in patients who had a Diagnosis-Related Group code for acute myocardial infarction [104]. Among patients who had angioplasty in New York in 1990, 2.5% had the procedure done within 6 hours of myocardial infarction and 24% had it done 6 hours to 21 days after myocardial infarction [105].
Rates of angiography and revascularization after myocardial infarction seem to be increasing in the elderly. Between 1987 and 1990, the percentage of patients older than 65 years of age who had angiography after myocardial infarction increased from 24% to 33%, the bypass surgery rate increased from 8% to 11%, and the angioplasty rate doubled from 5% to 10% [106].
Geographic variation in the use of angiography and revascularization has been seen in the United States and abroad. In the SAVE (Survival and Ventricular Enlargement) trial [107], 68% of patients in the United States who had had myocardial infarction subsequently had angiography compared with 35% of patients in Canada. Revascularization procedures were also used more often in the United States. These differences were not associated with any difference in mortality or reinfarction rates, but the incidence of activity-limiting angina was higher in Canada (33%) than in the United States (27%). In the GUSTO trial [108] (which also compared use of angiography in the United States and Canada), Canadian patients had a lower rate of angiography (25%) than did U.S. patients (72%), were less likely to have angioplasty (11%) than were U.S. patients (29%), and were less likely to have coronary artery bypass graft surgery (3%) than were U.S. patients (14%). In this study, Canadian patients also had worse functional status and more symptoms at 1 year than did U.S. patients: The prevalence of chest pain was 34% in Canadian patients and 21% in U.S. patients, and the prevalence of dyspnea was 45% in Canadian patients and 29% in U.S. patients.
Rates of cardiac catheterization and revascularization also vary among different areas of the United States. In 1987, the catheterization rate was 323 per 100 000 patients in the Midwest compared with 173 per 100 000 patients in the West [45]. Pilote and associates [109] used the GUSTO-1 data to evaluate variation in the management of patients who had had acute myocardial infarction and found that the proportion of patients who had various cardiac procedures substantially differed among regions. Angiography was done in 52% to 81% of patients, angioplasty was done in 22% to 35% of patients, and coronary artery bypass graft surgery was done in 9% to 17% of patients. Among eight regions, the use of these procedures was lowest in New England. In every region but New England, the use of procedures was closely related to their availability. Procedure rates did not seem to be related to rates of recurrent myocardial infarction or mortality at 30 days or 1 year. The study did not evaluate whether increased use of angiography and revascularization resulted in a better quality of life.
To address quality-of-life issues and rates of reinfarction and mortality, Guadagnoli and colleagues [110] compared the use of cardiac procedures after acute myocardial infarction in the Medicare populations in New York and Texas. Coronary angiography was done 50% more often in Texas (45% of patients) than in New York (30% of patients). However, the adjusted mortality rate for a period of 2 years was significantly lower in New York (hazard ratio, 0.87 [CI, 0.78 to 0.98]), and patients in Texas were 41% more likely to report angina and 62% more likely to say they that could not do activities that required 5 or more metabolic equivalents at 2 years than were patients in New York. The lower rate of angiography in New York probably reflects tighter regulatory control exercised by the state's Department of Health certificate of need process and supervision by its Cardiac Advisory Committee. Guadagnoli and colleagues [110] suggest that differences in long-term results may be an effect of different patterns of care after myocardial infarction, such as the greater use of β-blockers in New York.
Cost of Coronary Angiography and Angioplasty
The cost of coronary angiography includes the cardiac catheterization laboratory fee, professional fees, room charges, cost of repeating inadequate studies, and cost of complications [45]. Metropolitan Life reported that in 1992, the mean charge for inpatient cardiac catheterization for their beneficiaries was $10 880 [111]; total charges varied from $6400 in Maryland to $17 600 in California. Assuming a figure of $10 000, total annual charges for cardiac catheterization in the United States would be $10 billion, and charges after myocardial infarction would be approximately $1 billion. These figures reflect the assumption that coronary angiography would be done as an inpatient procedure. Although it can be done safely as an outpatient procedure at lower cost [112-116], angiography is associated with greater risk when it is done after myocardial infarction. Thus, the procedure is usually done on an inpatient basis.
A 1993 study estimated that charges of angioplasty were approximately $15 000 [117]. Assuming that about 400 000 such procedures are done per year, the charges in the United States would be $6 billion; about 25%, or $1.5 billion of the total amount, would be for care of patients after myocardial infarction. Using data from 1992 to 1993, one study [97] estimated that the median in-hospital cost of angioplasty was $9176.
Different strategies for the use of angiography after myocardial infarction may have a substantial effect on costs. The TIMI-II study compared the costs of the conservative strategy with those of the invasive strategy. The conservative strategy avoided 50% of coronary angiograms, and the investigators estimated the potential savings to be more than $700 million per year [79]. The actual cost data showed that the average total costs per hospital admission were $10 496 for the invasive strategy and $8826 for the conservative strategy [118].
Conclusions
Decisions about how to use coronary angiography after myocardial infarction have substantial implications for clinical factors and use of resources because they are closely linked with coronary revascularization procedures. Expert panels support the use of angiography after myocardial infarction in patients who have complications and in patients who have recurrent symptoms or noninvasive evidence of ischemia. The data that support this approach vary by indication and are often equivocal; however, outcomes are poor without revascularization, and angiography is probably beneficial. The potential benefits of primary angioplasty over thrombolytic therapy in patients who have had acute myocardial infarction are being vigorously debated [119, 120]. Small randomized trials suggest that short-term outcomes after angioplasty are better in centers that have substantial experience with the procedure, but a recent large observational study did not find a similar benefit [48]. This may partly be because rapid thrombolytic therapy is easier to administer than immediate angioplasty. Recent data suggest that outcomes vary substantially, particularly within the first 2 hours, depending on the amount of time that has elapsed before the artery opens [121].
Use of angiography and angioplasty varies substantially across regions. The findings from the TIMI-II study (that the conservative strategy reduced the number of angiographies by 50% without an increase in mortality rates) suggest that coronary angiography is currently done in many patients whose outcome is unlikely to be improved as a result of the procedure. Substantial resources are at stake: The total charges for catheterization after myocardial infarction are approximately $1 billion per year; this figure does not include the cost of the coronary revascularization procedures that follow.
Dr. Bernstein: University of Michigan Medical Center, Division of General Medicine, 3116 Taubman Center, Ann Arbor, MI 48109.
Dr. Hauptman: Division of Cardiology, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115.
Dr. Leape: Department of Health Policy and Management, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115.
- Copyright ©2004 by the American College of Physicians
References
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.↵
- 114.↵
- 115.↵
- 116.↵
- 117.↵
- 118.↵
- 119.↵
- 120.↵
- 121.↵
Related Clinical Content
