Methods

Top Methods Results Discussion Author & Article Info References

We conducted a retrospective analysis of admissions to the emergency departments of four hospitals affiliated with Brown University. Each hospital represented a different patient population and a different level of cardiovascular service. One was a Veterans Administration hospital without a cardiac catheterization laboratory, two were community hospitals with diagnostic cardiac catheterization laboratories, and one was a community hospital with an active referral center where percutaneous cardiovascular revascularization procedures and cardiac surgery are performed.

After we received approval from the institutional review board of each hospital, we evaluated the records of all patients who were admitted between 1 January 1994 and 31 December 1994. We identified patients who had any of the following discharge diagnoses: acute myocardial infarction, other acute and subacute ischemic heart diseases, angina pectoris, other ischemic heart diseases, heart failure, other heart diseases, and chest pain (diagnosis-related group codes 410, 411, 413, 414, 428, 429, and 786.4). Records were excluded if documentation was incomplete, if the patient was directly admitted or transferred [that is, if an emergency department record did not exist], if the hospital record could not be located or was incomplete, or if an acute coronary syndrome had not been suspected by the emergency department physician. We tabulated 1) demographic variables; 2) history of cardiovascular disease and its risk factors [whether patients were older than 55 years of age; had a history of diabetes, hypertension, or hypercholesterolemia; had a family medical history positive for cardiovascular disease; and had a previous coronary event]; 3) current medical therapy; 4) initial clinical impression of the emergency department physician [acute myocardial infarction, unstable angina, chest pain with high or low suspicion of myocardial ischemia, and congestive heart failure]; 5) interpretation of the initial electrocardiogram; 6) allergy or contraindication to aspirin [history of a hemorrhagic stroke, gastrointestinal bleeding, or a bleeding disorder]; 7) amount of time between arrival in the emergency department and initiation of aspirin therapy; and 8) documentation of acute myocardial infarction during the subsequent hospital stay (the diagnosis was made if creatine kinase-MB levels increased to above normal as documented by serial measurement or if two of the following three criteria were met: chest pain, twofold elevation of the creatine kinase level, or new Q waves on the electrocardiogram).

The mean age ±SD was determined, and the Student t-test was used to evaluate significant differences (P < 0.05) in aspirin administration (yes or no) as the grouping variable. Frequency tables were generated for remaining variables. The Pearson chi-square test was used for these data.

Results

Top Methods Results Discussion Author & Article Info References

Demographic and Clinical Characteristics

We found 2383 medical records that met our study criteria (Table 1). Hypertension was the most prevalent cardiac risk factor (58% of patients); 43% of patients had previously documented coronary artery disease, and 35% reported receiving long-term aspirin therapy. Allergy or contraindication to aspirin was confirmed in 101 patients (4%). The initial electrocardiogram taken by emergency department staff showed an acute injury pattern in 190 patients (8%). Aspirin use was higher in younger persons, particularly those with ST-segment elevation on electrocardiography (P < 0.001). Aspirin was also used more frequently in male patients, current smokers, and patients with a family history of heart disease.

Many factors were associated with less frequent use of aspirin. Among the patients already receiving long-term aspirin therapy, 23% received aspirin in the emergency department compared with 33% of patients who were not previously receiving an aspirin regimen (P < 0.001). Similarly, 17% of patients receiving long-term warfarin therapy were given aspirin compared with 31% of patients who were not receiving warfarin therapy (P < 0.001).

Diagnosis on Admission to the Emergency Department

The initial diagnosis made by the emergency department physician (Table 2) was acute myocardial infarction in 220 patients (9%), unstable angina in 911 patients (38%), chest pain with high suspicion of ischemia in 561 patients (24%), chest pain with low suspicion of ischemia in 262 patients (11%), and congestive heart failure in 382 patients (16%). In 47 patients (2%), the diagnosis on admission to the emergency department was not documented.

Timing and Relation of Aspirin Therapy to Diagnosis

The overall frequency of the use of aspirin by emergency department physicians was 30% (712 of 2383 patients). The difference in the use of aspirin on the basis of initial clinical impression was significant: More patients received aspirin if the physician initially believed that they had acute myocardial infarction (P < 0.001) (Table 2).

Of the patients who received aspirin, 22% received it within 30 minutes of arrival, 24% received it between 30 and 60 minutes of arrival, and 54% received it 1 hour after arrival.

According to hospital records, acute myocardial infarction was subsequently diagnosed in 463 patients (19%); 253 of these patients (55%) did not receive aspirin in the emergency department. The discharge diagnosis of acute myocardial infarction was established in 17% of patients whose initial diagnosis had been unstable angina or congestive heart failure. Less than 5% of patients who had chest pain associated with a low likelihood of ischemia had a discharge diagnosis of acute myocardial infarction.

Discussion

Top Methods Results Discussion Author & Article Info References

Our study demonstrates that aspirin is underutilized in patients who are admitted with suspected myocardial infarction and related syndromes that prove to be acute myocardial infarctions at discharge. Furthermore, of the patients who did receive aspirin in the emergency department, the delay in administration of aspirin after arrival was substantial. Overall, 45% of patients who had been documented at discharge as having had acute myocardial infarction had received aspirin in the emergency department. Aspirin was most likely to be used when acute myocardial infarction was initially suspected. That this diagnosis also proved to be accurate in 17% of patients who were admitted with unstable angina or congestive heart failure is noteworthy. A rigid approach that restricts aspirin use to patients who are admitted with the strongest evidence of acute myocardial infarction deprives many others of the benefits of early aspirin therapy.

Guidelines published in 1994 recommended the immediate administration of 160 mg of aspirin to patients in whom myocardial infarction is suspected [4]; this recommendation was recently reaffirmed [7]. Strong clinical evidence supports the efficacy of aspirin. The ISIS-2 (Second International Study of Infarct Survival) trial [3] that involved more than 17 000 patients with suspected myocardial infarction revealed a 23% reduction in the mortality rate of patients who received aspirin alone and an additional 19% reduction when aspirin was given with streptokinase. The incidence of nonfatal myocardial infarction was also reduced by 49% [3]. In patients who had unstable angina, several studies have shown strong evidence that supports the role of aspirin [8-10].

The use of aspirin has apparently been influenced by clinical trials. Between 1987 and 1990, aspirin use before myocardial infarction increased from 16% to 24% and aspirin use after myocardial infarction increased from 39% to 72% [11]. Krumholz and colleagues [6] reported that aspirin therapy was initiated in 56% of elderly Medicare patients within the first 24 hours of hospitalization. However, they did not evaluate the use of aspirin in emergency departments. Other evaluations of aspirin therapy have also shown that practitioners who refer patients for hospital admission underutilize aspirin therapy [12, 13].

Our data show that 17% of patients who were admitted with a diagnosis of unstable angina proved to have acute myocardial infarction at discharge. An identical rate was noted in patients who presented with congestive heart failure. Many elderly patients who have acute myocardial infarction may initially show signs of pulmonary congestion in the absence of chest pain. On the basis of our data, we propose that emergency department physicians give aspirin to all patients who are to be admitted to the hospital for possible myocardial infarction, with the possible exception of patients in whom suspicion of an acute coronary event is low. Aspirin may be particularly useful in elderly patients who present with unexplained heart failure.

Our study, which addresses the underutilization of aspirin in patients with acute coronary syndromes in the emergency department, has some limitations. First, it is an analysis of 1994 data; a study of more current data might show improved practice, particularly because interim reports have documented and publicized the underutilization of aspirin [5, 6]. Furthermore, our study is retrospective and has the weaknesses inherent to such an approach. In particular, except for contraindications to aspirin use, we did not explore other reasons why aspirin might not be given, such as whether the patient had taken aspirin at home before admission to the emergency department and heterogeneity of physician training. Despite these limitations, we believe that because aspirin has an excellent safety profile, is easy to administer, is cost-effective, and can be immediately discontinued if acute coronary disease is excluded, physicians in the emergency department should give aspirin more frequently and earlier to patients with suspected or potential myocardial infarction. Promulgating our findings and establishing an emergency department critical path that emphasizes early administration of aspirin would facilitate achievement of this goal.

Dr. Conte: 6600 Boulevard East, West New York, NJ 07093.

Dr. Noris: Truesdale Cardiology Associates, 1030 President Avenue, Fall River, MA 02720.

Drs. Tilkemeier, Forman, and Parisi: Division of Cardiology Department of Medicine, The Miriam Hospital, 164 Summit Avenue, Providence, RI 02906.

Dr. Miller: Grand River Cardiology, 1000 East Paris SE, Suite 215, Grand Rapids, MI 49546.

Drs. Cannistra, Leavitt, and Garber: Division of Cardiology, Department of Medicine, Memorial Hospital of Rhode Island, 111 Brewster Street, Pawtucket, RI 02860.

Dr. Sharma: Division of Cardiology, Department of Medicine, Veterans Affairs Medical Center, 830 Chalkstone Avenue, Providence, RI 02908.