| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
1 Department of Emergency Medicine
2 Department of Medicine, Gerontology Division, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts.
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. To assess the value of serial cardiac enzymes in elderly patients who present to the ED with syncope, a retrospective chart review was performed on consecutive patients aged 65 and older with syncope in an urban teaching hospital ED between July 1, 1998 and June 30, 1999. Charts were screened for presenting history, cardiac risk factors, testing, and outcomes including acute coronary syndromes, MI, death, and patients returning to the ED or admitted within 72 hours of discharge.
Results. 319 patients met the study criteria of syncope with confirmed loss of consciousness in the absence of seizure or stroke. 141 of 228 admitted patients (62%) had creatine phosphokinase (CPK) drawn and 5% of these had Troponin I (TnI). 3 of 141 patients, or 2.1% (95% CI [confidence interval]: 0.04%–6.09%), had positive cardiac enzymes during their hospitalization. CPK was positive in all 3, and TnI, drawn in 1 patient, was also positive. Two of these patients had chest discomfort and ST segment and T-wave abnormalities on electrocardiogram (ECG) in addition to a syncopal event. The third patient had dementia and could not recall the details surrounding her syncopal event. In addition, her baseline ECG demonstrated a left bundle branch block, limiting ECG interpretation.
Conclusions. Cardiac enzymes may be of little additional value if drawn routinely on elderly patients with syncope who are admitted to the hospital from the ED, unless they have other signs or symptoms suggestive of myocardial ischemia.
Currently, most elderly patients with syncope admitted to the hospital from the ED department do not have a myocardial infarction (MI), yet common practice is to draw serial cardiac enzymes.
Previously published data may help to understand the rationale behind routine cardiac enzyme evaluations in patients with syncope. Syncope is difficult to diagnose, with as many as 34% of patients remaining without an etiology on hospital discharge (5). Syncope in elderly patients is even more difficult to diagnose and is associated with a higher level of morbidity and mortality (6). Sixty percent of patients do not have a readily diagnosed etiology based on initial history, physical exam, and electrocardiogram (ECG) (7). Kapoor followed 443 patients for 40 months following a syncopal event. Forty-three patients (10%) died suddenly. Twenty-eight of those 43 (32%) died from cardiac causes (8).
In the evaluation of syncope, the presence of a structural heart disease (coronary artery disease, congestive heart failure, valvular heart disease, or congenital heart disease) has emerged as the most important factor for predicting death and cardiac arrhythmias (9). Patients, who are over the age of 45, have a history of ventricular arrhythmias, congestive heart failure, or an abnormal ECG, have an increased risk of death of up to 28% within 1 year following syncope (9).
Early studies suggested an incidence of MI and syncope to be between 6% and 21%, with increased incidence associated with advancing age (8,10,11). However, more recent data describing patients of all ages suggest an incidence of approximately 1% (12,13). A newly published study of younger patients with syncope (average age 66 years) noted 1 out of 80 patients, in whom cardiac enzymes were drawn, to have had an MI (13). Previous studies have also shown no prognostic significance in elderly patients with minimal elevation of cardiac enzymes (14).
Several factors have forced a reassessment of this practice of drawing serial cardiac enzymes on syncope patients, including increasing emphasis on evidence-based medicine, efforts to reduce unnecessary and expensive medical testing, and concerns over the rapidly rising costs of medical care.
Therefore, we sought to evaluate the utility of routine cardiac enzymes in geriatric patients presenting to the ED with syncope.
|
METHODS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
An institutional database was searched for all records during a 1-year period (July 1, 1998 to June 30, 1999) of patients aged 65 and older to identify those patients who presented to the ED with either a documented chief complaint of syncope, noted by the triage nurse, or who were given an ICD-9-CM code 780.2 (syncope and collapse) from the ED as either an admitting or discharge diagnosis. Syncope was investigated regardless as to whether it was reported as the primary or the secondary diagnosis. The method we employed to identify patients with syncope based on coded diagnoses is similar to those utilized in other similar retrospective studies (12,16,17). Inclusion criteria included age 65 or older, documented loss of consciousness, and no obvious other etiology of that loss of consciousness such as stroke or seizure. Each chart was reviewed by a study author to confirm that the patient met inclusion criteria. In order to ensure uniform abstraction, if discrepancies were found between the ICD9 diagnosis of syncope and there was no documented loss of consciousness or stroke, the chart was then reviewed by the primary investigator. The chart of each patient with abnormal cardiac enzymes was also reviewed by the primary investigator in addition to another study investigator. If a chart was unavailable, data were abstracted from the patients' computerized discharge summary, laboratory results, and ECG.
If after chart review, no documentation of loss of consciousness could be found, these patients were excluded from the study. Similarly, patients were excluded if they were noted to have had a seizure or stroke.
Charts were screened for presenting history, cardiac risk factors, and outcomes including acute coronary syndromes, MI, and death. Information was gathered about the performance and outcome of diagnostic studies done in the ED and during hospitalization including ECGs and cardiac enzymes. Cardiac enzymes included both serial creatinine kinase of muscle band (CK-MB) and Troponin I (TnI). MB index of 
5% or TnI of 1.5 was considered indicative of an MI. Cardiac enzymes were not sent unless the patient was to be admitted to the hospital. If cardiac enzymes were sent, each patient would have at least 3 sets of cardiac enzymes timed 8 hours apart.
Patients returning to the ED or admitted as an inpatient within 72 hours of discharge were reviewed as well.
The principal investigator had full access to all the data in the study and took responsibility for the integrity of the data and the accuracy of data analysis.
|
RESULTS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
ECG was performed on 312 of 319 patients (98%). Two hundred nine ECGs (67%) were read as abnormal. ST segment and T-wave abnormalities were seen in 27 of 209 ECGs (13%). Other ECG abnormalities that were commonly seen included atrial fibrillation/atrial flutter (10%) and atrioventricular block (8.3%) (see Table 2). Patients with abnormal ECGs were far more likely to have coronary artery disease as a baseline comorbidity ( p <.003).
|
|
DISCUSSION |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
One patient with an MI had diffuse ST segment and T-wave abnormalities of unknown age on ECG, another had nonspecific ST and T-wave abnormalities unchanged from a prior ECG. The third patient's baseline ECG demonstrated an LBBB, limiting the interpretation of the ECG. It would appear that patients with an ECG demonstrating new ST and T-wave abnormalities or an LBBB might trigger serial enzyme draws. However, only half of the patients with new ST and T-wave abnormalities or LBBB had enzymes drawn.
The average charges for cardiac enzymes are significant. Total CK is $23, CK-MB is $34, and cardiac TnI is $71 in our clinical laboratory. For 3 sets of enzymes, the charge per patient would total $384. Hospital costs are approximately $51 per run (includes CK, CK-MB, and Troponin), but this only includes incremental reagent costs and does not include long-term staffing and machinery costs. Thus, based on costs or charges, if routine drawing of cardiac enzymes in patients with syncope ceases, the result would be notable health care savings.
Recently published guidelines on management of syncope suggest that patients with syncope do not require basic laboratory tests (18,19). Laboratory testing is only indicated if syncope may be due to reduced circulating volume or from a metabolic cause. Routine cardiac enzyme determinations are not recommended (18,19). Our data similarly suggest that cardiac enzymes are frequently drawn on patients with syncope, despite not having chest discomfort and having normal enzymes.
Based on our study and previously published data, it may be plausible to suggest guidelines for drawing cardiac enzymes in elderly people with syncope. Specifically, patients with associated pertinent history findings consistent with cardiac ischemia, such as chest discomfort, or patients from whom an adequate medical history cannot be elicited, such as those with dementia, should have cardiac enzymes tested. All other patients presenting with syncope have a low likelihood of acute MI and may not require cardiac enzyme determinations.
Limitations of this study include the restrictions of retrospective analysis, use of a single testing site, and small sample size. Since 38% of patients included in this study did not have enzymes examined, it is possible that MI could have been missed in those patients. In addition, for unclear reasons, only 7 of 11 patients with chest discomfort and syncope had enzyme determinations. This study did not look at dementia and syncope, but further analysis of laboratory testing in this population is probably warranted. A larger prospective study in which all patients with syncope have serial cardiac enzymes would help clarify the value of cardiac enzymes in syncope. Nevertheless, as current published guidelines do not recommend routine serial enzyme evaluations, we believe that most patients should not have routine cardiac enzyme determinations.
Conclusion
Cardiac enzymes may be of little additional value if drawn routinely on elderly patients with syncope who are admitted to the hospital from the ED, unless they have other signs or symptoms suggestive of myocardial ischemia by history, such as chest discomfort, or are patients from whom an adequate medical history cannot be elicited, such as those with dementia. The savings from limiting the number of unnecessary blood tests drawn on patients with syncope may be substantial as well.
|
Acknowledgments |
|---|
Address correspondence to Shamai A. Grossman, MD, MS, Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, WCC2, One Deaconess Road, Boston, MA 02115. E-mail: sgrossma{at}caregroup.harvard.edu
Received December 27, 2002
Accepted January 3, 2002
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
