Increasing Use of Single and Combination Medical Therapy in Patients Hospitalized for Acute Myocardial Infarction in the 21st Century:  A Multinational Perspective FREE

Physicians involved in the treatment of patients with coronary heart disease have an abundance of proven effective therapies for secondary prevention. The effectiveness of 4 medications—aspirin, β-blockers, angiotensin-converting enzyme (ACE) inhibitors, and lipid-lowering agents (hereinafter, cardiac medications)—in patients discharged from the hospital after acute myocardial infarction (AMI) has been well established in large randomized controlled trials. Accordingly, guidelines for the treatment of patients after AMI encourage the prescribing of these 4 therapies for most hospitalized patients.^1- 2 However, the manner in which data have accumulated for each of these medications (eg, randomized controlled trials that usually have not mandated the use of the other classes of agents) has shed little light on the relative importance of each therapy within a multidrug regimen for patients with acute coronary disease. In addition, implementation and long-term maintenance of a complicated drug regimen can be challenging and expensive.

The primary objective of our multinational observational study was to describe the contemporary, as well as changing, prescribing of 4 effective cardiac medications in patients surviving hospitalization for AMI from 2000 to 2005 without contraindications to the receipt of these medications. Given slight differences in currently published guidelines for the treatment of ST-segment elevation AMI vs non–ST-segment elevation AMI (hereinafter, STEMI and NSTEMI, respectively), a secondary goal of this study was to determine whether use of 4 evidence-based cardiac medications at the time of hospital discharge differed between patients discharged with each type of AMI (STEMI or NSTEMI). Data from the Global Registry of Acute Coronary Events (GRACE) were used for the present investigation.^3- 5

Details of the methods and logistical aspects of this study have been previously described.^3- 5 In brief, the GRACE project is designed to reflect an unselected population of patients with an acute coronary syndrome (ACS), irrespective of geographic region. A total of 113 hospitals located in 14 countries in North America, South America, Europe, Australia, and New Zealand have contributed data to this ongoing observational study.

Patients entered in the registry have to be at least 18 years old and alive at the time of hospital presentation, be admitted for an ACS as a presumptive diagnosis, and have at least 1 of the following present: electrocardiographic changes consistent with an ACS, serial increases in serum biochemical markers of cardiac necrosis, and/or documentation of coronary artery disease.^3- 5 The study aims to enroll an unselected population of patients with ACS, and sites were encouraged to recruit the first 10 to 20 consecutive eligible patients each month. Data are collected by trained study coordinators using standardized case report forms. Patients with STEMI and NSTEMI were included in the present investigation, with standardized working definitions of these groups used.^3- 5 Hospital-specific feedback regarding patient characteristics, treatment, and outcomes are provided to each participating medical center on a quarterly basis in the form of written reports.

Information was collected about each patient's age, sex, medical history, and clinical characteristics through the review of hospital medical records by trained nurse and physician abstractors.^3- 4 Standardized definitions of all patient-related variables, clinical diagnoses, and hospital complications and outcomes were used. Information about the use of the 4 cardiac medications during hospitalization was obtained through the review of hospital medical records and physicians' and nurses' progress notes.³ Based on the review of information contained in medical records by our trained abstractors, we further identified patients who had not previously received these medications and were prescribed them during hospitalization. These patients were classified as new treatment initiators.

Our abstractors were trained to review hospital medical records in a standardized manner and to identify patients in whom contraindications to the receipt of each of the 4 medications under study were present (the criteria for this review are available from the authors). These criteria are based on contemporary published guidelines, are designed to be permissive, and can be noted by the GRACE investigators without the specific contraindication also being noted. Contraindications to the receipt of aspirin, β-blockers, and ACE inhibitors were noted over the entire study period, whereas contraindications to the receipt of lipid-lowering therapy were noted only from 2002 to 2005. The sample of the present study includes patients who were eligible for the receipt of all 4 cardiac medications examined based on our exclusion criteria.

Differences in various characteristics between patients prescribed the 4 cardiac medications vs those not prescribed those medications were examined by using χ² tests of statistical significance for discrete variables and t tests or analysis of variance for continuous variables. Changes over time in the percentages of patients treated with the 4 cardiac medications were examined through the use of χ² tests for trends. A logistic regression analysis was used to examine changing trends in the prescribing of all 4 cardiac therapies in our sample of patients while controlling for the patient's age, sex, medical history of cardiovascular disease (eg, angina, diabetes mellitus, heart failure, hypertension, myocardial infarction, and stroke), geographic region, and development of atrial fibrillation, cardiogenic shock, or heart failure during hospitalization. A polytomous logit regression analysis was performed to examine sociodemographic, medical history, and clinical factors associated with the receipt of 2, 1, or no cardiac medications during hospitalization (dependent variable) compared with the receipt of 3 or 4, or all 4, medical therapies; the time period of hospitalization was also controlled for in these regression analyses. We did not perform an ordered logit regression analysis because the assumptions used for ordinal regression were not met. Multivariable-adjusted odds ratios and accompanying 95% confidence intervals are presented herein, in the text and tables.

A total of 26 413 patients with AMI were hospitalized from January 2000 to December 2005 without contraindications to the 4 medications examined and comprised the present study population. This included 13 969 patients with STEMI and 12 444 patients with NSTEMI. A total of 4833 patients were included in 2000, 5213 in 2001, 5464 in 2002, 6027 in 2003, 5379 in 2004, and 3746 in 2005. The percentage of study patients with STEMI declined slightly over time from 56.4% during the first 6 months of 2000 to 51.8% during the second 6 months of 2005. The percentages of patients with contraindications noted to the 4 study drugs who were excluded from this study population were as follows: 3.2% had contraindications to aspirin; 7.4%, to β-blockers; 4.8%, to ACE inhibitors; and 0.4%, to statin therapy. Overall, 12.7% of the original GRACE sample were excluded from this study because they had contraindications to 1 or more of the 4 study medications. The percentage of patients who had contraindications to the receipt of any of these therapies was 12.9% in 2000, 12.6% in 2003, and 7.3% in 2005.

Increases were observed in the singular use of 3 of the 4 cardiac medications examined during the study period (2000-2005) (Figure 1). Particularly marked increases were observed with regard to the use of statin therapy (from 45% to 85%) and ACE inhibitors (from 63% to 77%), whereas less dramatic increases were noted in the use of β-blockers (from 83% to 91%) over time. There were few to no increases in the use of aspirin over time because most (approximately 95%) of hospital survivors of AMI were treated with this therapy.

There were marked increases in the use of multiple effective cardiac medications, particularly in the use of all 4 cardiac medications over time (Figure 2). Approximately 23% of patients discharged from participating hospitals during the first 6 months of 2000 were prescribed each of the 4 cardiac medications during hospitalization; in the second 6 months of 2005, approximately 58% of hospital survivors were prescribed all 4 of these therapies.

Virtually identical trends in the use of single and combination medication therapy in patients with STEMI and NSTEMI were observed from 2000 to 2005. Among patients with STEMI, increases in the use of statins (from 47% to 86%), ACE inhibitors (from 68% to 81%), and β-blockers (from 86% to 92%) were observed. Aspirin use remained steady (used in approximately 95% of patients with STEMI) (Figure 3). In patients with NSTEMI, increases in the use of statins (from 42% to 83%), ACE inhibitors (from 57% to 73%), and β-blockers (from 80% to 90%) were noted. Aspirin use (in approximately 95% of patients with NSTEMI) remained consistently high over all periods examined (Figure 4). Marked increases in the use of all 4 cardiac medications were observed in hospital survivors of STEMI (from 27% to 65%) and NSTEMI (from 18% to 56%) over time.

We examined changing trends in the prescribing of all 4 cardiac medications vs 2, 1, or no cardiac medications while controlling for several potentially confounding demographic and clinical factors (Table 1). There was a progressive increase over time in the prescribing of all 4 cardiac medications in all study patients as well as in those with STEMI and NSTEMI.

Increases in the use of new multiple cardiac therapies at the time of hospital discharge in patients who were not previously receiving these medications were observed, as were corresponding declines in the use of a few effective cardiac medications. In 2000, 38% of the potentially eligible 1974 hospital survivors of AMI were discharged while receiving 1, or no, new cardiac medication, whereas 10% were discharged while receiving all 4 therapies. In 2005, only 28% of the 1208 patients who were not receiving these therapies previously were discharged while receiving 1, or no, new cardiac medication, whereas 29% of such patients were treated with all 4 cardiac medications. Among patients with STEMI, increases were observed in the percentage of patients newly discharged while receiving all 4 medications from 2000 to 2005 (from 12% to 41%). Among patients with NSTEMI, an increase (from 6% to 17%) was observed in the percentage of patients who were newly prescribed all 4 of these medications.

We examined whether physicians were more (or less) likely to prescribe these 4 cardiac medications to potentially eligible patients further classified according to various characteristics (Table 2). Increases in the use of all 4 cardiac medications were observed in patients of all ages, of both sexes, and from all geographic regions studied; in those with vs without each of the comorbidities being studied; and in those who did or did not develop various hospital clinical complications over time. Relatively similar trends were observed in patients with STEMI and NSTEMI.

To more systematically examine the characteristics of patients who were not treated with these effective cardiac medications, we performed a polytomous logit regression analysis. The characteristics of patients who were prescribed 2, 1, or no cardiac medications at the time of hospital discharge were compared with the characteristics of patients who received all 4 cardiac therapies for the first analysis (Table 3), whereas the characteristics of those who were prescribed 2, 1, or no cardiac medications were compared with the demographic and clinical characteristics of those who were treated with any 3 therapies in a second regression analysis (Table 4). Demographic, medical history, and clinical characteristics of patients (Table 2) were considered as potential predictors of our treatment-related outcome, and we also controlled for the time period of hospitalization in our regression models. Among potentially eligible patients of advancing age (≥ 65 years), female sex, those with a history of heart failure or stroke, those hospitalized in participating centers from Argentina or Brazil, and those who developed atrial fibrillation during hospitalization were more likely to be discharged receiving none, 1, or 2 vs all 4 of the cardiac medications (Table 3). In patients with STEMI or NSTEMI, relatively similar factors were associated with the underuse of these 4 cardiac therapies (Table 3). Relatively similar factors were associated with the underuse of effective cardiac medications when we compared the characteristics of patients who were treated with relatively few effective cardiac therapies with those who received any 3 medications, although some factors had a more attenuated association (Table 4).

The results of our study suggest encouraging increases over time in the prescribing of single medications, and combination medical therapy, in a multinational sample of patients hospitalized with AMI who were without contraindications to the use of all 4 of the cardiac medications being studied. We identified several demographic and clinical factors associated with suboptimal pharmacologic treatment.

Prior and current American College of Cardiology and American Heart Association^1- 2,6 guidelines strongly support the use of aspirin and β-blockers in most patients who survive NSTEMI or STEMI. Members of these consensus committees have previously limited their recommendations for the use of ACE inhibitors to patients with left ventricular dysfunction and for the use of lipid-lowering agents in patients with elevated low-density lipoprotein levels. However, several recent clinical trials^7- 8 have demonstrated the beneficial effects associated with the use of ACE inhibitors and lipid-lowering agents in patients with ACS, irrespective of the extent of left ventricular dysfunction or serum cholesterol levels. Furthermore, recently updated American College of Cardiology and American Heart Association² guidelines for patients with STEMI now recommend the use of all 4 medications for hospital survivors of AMI. One multidisciplinary group⁹ has recently published a simplified risk-oriented algorithm to develop a more optimal early treatment strategy for patients with NSTEMI. As guidelines for the treatment of patients with AMI become more and more complex, it will be increasingly important to monitor adherence to these recommendations in more generalizable samples of patients hospitalized with AMI.

We observed marked increases in the prescription of each of the 4 cardiac medications examined, and combinations thereof, in this large coronary disease registry. Despite these encouraging trends, current gaps remain in the effective use of these agents. The data described herein highlight a unique challenge to physicians currently caring for patients with AMI. Institution of several new medications in a previously healthy young patient with an uncomplicated AMI can be difficult; successful use of an enhanced treatment regimen in older patients with multiple comorbidities is even more daunting, as are issues related to long-term adherence to potentially costly prescribed medications.

Limited data are available about the contemporary, as well as changing, prescribing of effective cardiac medications to patients hospitalized with AMI from a more generalizable community-wide perspective. In the Worcester Heart Attack Study,^10- 11 increases in the use of ACE inhibitors, aspirin, β-blockers, and lipid-lowering therapy have been noted in this community sample of patients hospitalized with AMI in central Massachusetts. This study¹¹ has also described increases in the use of combination medical therapy in hospital survivors of AMI over time. In the Olmsted County, Minnesota, study,¹² the use of ACE inhibitors, aspirin, and β-blockers increased from 1989 to 1998 in hospital survivors of AMI. A report from the National Registry of Myocardial Infarction¹³ documented marked increases in the use of aspirin (approximately 10%), β-blockers (approximately 37%), and ACE inhibitors (approximately 40%) from 1994 to 1999. In an analysis of Medicare patients hospitalized with myocardial infarction during the mid to late 1990s,^14- 15 discharge prescriptions for aspirin and β-blockers increased for potentially eligible patients. Additional investigations^14- 15 have documented the underuse of cardiac medications of proven benefit in patients with acute coronary disease, albeit with improvements over time. A recent overview¹⁶ has documented the underuse of medications of proven efficacy in the treatment of patients with AMI in developing countries.

We identified a number of factors strongly associated with the failure to receive multiple effective cardiac therapies after hospital discharge for AMI. Older age was the strongest predictor of failure to employ combination therapy in our study. This finding is not surprising because advanced age has been associated with the underuse of beneficial cardiac therapies in multiple studies.^17- 19 Women were considerably less likely to be prescribed combination therapy. Women have been previously identified as being less likely to be prescribed effective cardiac therapies in other studies.^19- 20 Patients hospitalized at participating centers in Argentina and Brazil were less likely to be treated with optimal cardiac therapy, which suggests that issues related to medication costs or health care delivery systems need to be considered in examining trends in global medication use patterns.

Patients with a medical history of either heart failure or stroke were less likely to be prescribed multiple combination therapy than patients without these important comorbidities. Although some comorbidities may represent a legitimate impediment to initiation of multiple medication regimens, they are rarely an absolute contraindication to treatment. These high-risk patient subsets often have the most to gain from aggressive therapy of underlying vascular and platelet dysfunction. Patients who developed atrial fibrillation during hospitalization were less likely to be discharged while receiving multiple cardiac therapies. This may be due to differences in the manner by which these patients can be treated and their need for anticoagulant therapy. In the Worcester Heart Attack Study,¹¹ advanced age, female sex, medical history of selected comorbidities, presence of an initial non–Q-wave myocardial infarction, and development of atrial fibrillation during hospitalization were associated with the underuse of combination medical therapies in patients hospitalized with AMI.

We were unable to record the reasons for nonprescription of certain therapies in the present investigation. Although we included only patients who were deemed eligible for the receipt of all 4 cardiac therapies, these criteria were reasonably permissive and may have included some patients in whom these therapies were not indicated for use. Details about medication dosages or use of specific medications within a class of agents were not available. We were also unable to examine long-term medication use patterns after hospital discharge and relate these patterns to therapies prescribed at the time of hospital discharge. Feedback is provided to participating study sites through quarterly reports, which may have affected prescribing practices over time.

In conclusion, the results of our large observational study suggest increases from 2000 to 2005 in the use of combination medical therapy in eligible patients discharged from the hospital after AMI. Despite these encouraging trends, gaps in the use of combination medical therapies continue to exist. Closing this gap will require novel and concerted efforts.^21- 22 Increased understanding and minimization of drug interactions, increased use of “combination pills,” and education about the differences between “polypharmacy” and effective combination therapy are still needed to enhance the treatment of patients with acute coronary disease and their long-term outcomes.

Correspondence: Robert J. Goldberg, PhD, Division of Cardiovascular Medicine, Department of Medicine, University of Massachusetts Medical School, 55 Lake Ave, Worcester, MA 01655 (goldberr@ummhc.org).

Accepted for Publication: April 21, 2007.

Author Contributions:Study concept and design: Goldberg, Steg, Montalescot, and Gore. Acquisition of data: Steg, Montalescot, and Goodman. Analysis and interpretation of data: Goldberg, Spencer, Steg, Flather, Montalescot, Gurfinkel, Kennelly, Goodman, Dedrick, and Gore. Drafting of the manuscript: Goldberg, Spencer, and Dedrick. Critical revision of the manuscript for important intellectual content: Goldberg, Spencer, Steg, Flather, Montalescot, Gurfinkel, Kennelly, Goodman, and Gore. Statistical analysis: Goldberg and Dedrick. Administrative, technical, and material support: Steg and Goodman. Study supervision: Steg, Flather, Montalescot, and Gore.

Financial Disclosure: None reported.

Funding/Support: The GRACE project is supported by an unrestricted educational grant from Sanofi-Aventis to the Center for Outcomes Research, University of Massachusetts Medical School.

Group Information: A list of the Global Registry of Acute Coronary Events Investigators is available at http://www.outcomes-umassmed.org/GRACE/study.cfm.

Role of the Sponsor: Sanofi-Aventis had no involvement in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the manuscript for publication.

Additional Information: Further information about the GRACE project is available at http://www.outcomes.org/grace.