Use of the combination of myoglobin and CK-MB mass for the rapid diagnosis of acute myocardial infarction

Use of the Combination of Myoglobin and CK-MB Mass for the Rapid Diagnosis of Acute Myocardial Infarction MICHAEL C. KONTOS, MD, F. PHILIP ANDERSON, PHD, C. MICHAEL HANBURY, PHD, CHARLOTTE S. ROBERTS, BS, W. GREG MILLER, PHD, ROBERT L. JESSE, MD, PHD Early identification of patients presenting with myocardial infarction (MI) is necessary for rapid initiation of treatment. Currently, MI has been diagnosed using the combination of the history, electrocardiogram (ECG), and biochemical markers of myocardial necrosis. Unfortunately, all lack sufficient sensitivity and specificity to confidently identify most patients with MI in a timely enough fashion to influence early intervention. Development of newer immunochemical assays for CK-MB mass and myoglobin have allowed for earlier, more rapid diagnosis; however, each has important limitations. The diagnostic sensitivity of CK-MB mass, myoglobin, and the combination of both were analyzed at the time of presentation (0 hours) and again 4 hours later in 101 patients admitted from the emergency department (ED) with possible il. Twenty patients were subsequently diagnosed as having al. The sensitivity of the initial ECG was 60%, compared with the sensitivities of the initial myoglobin and CK-MB mass of 70% and 30%, respectively. By 4 hours the sensitivity of myoglobin had increased to 85% and CK-MB mass to 90%. The combination of the initial myoglobin and CK-MB mass had a sensitivity of 85%. Combining these two markers, using both the initial and 4-hour samples, raised the sensitivity to 100%, with a specificity of 100% and negative predictive value of 100%. When patients with diagnostic ECGs were excluded, the sensitivity of the combination at 0 hours was 80% with a specificity of 84%, while the use of the 0- and 4-hour markers had a sensitivity and specificity of 100% and 100% respectively. We conclude that the combination of CK-MB mass and myoglobin can rapidly diagnose or exclude il in as short as 4 hours after ED presentation, and accuracy is not different in patients without diagnostic ECGs. Application of this strategy could potentially lead to more rapid intervention in patients with MI, while also allowing early identification of lower risk patients. (Am J Emerg Med 1997;15:14-19. Copyright © 1997 by W.B. Saunders Company) Each year, more than 5 million patients are evaluated in emergency departments (EDs) for chest pain, 1 the majority of whom are subsequently admitted to coronary care units for possible myocardial infarction (MI). 2 The decision for intensive care admission is based on the presenting clinical symptoms and the initial electrocardiogram (ECG), both of which lack sensitivity and specificity. As a result, only 15% to 20% of these patients ultimately have MI. 3 Application of From the Medical College of Virginia/Virginia Commonwealth University, Richmond. Manuscript received December 27, 1995; accepted January 24, 1996. Presented in part at the 68th Scientific Sessions of the American Heart Association, Anaheim, CA, November, 1995. Address reprint requests to Dr Jesse, Box 281, Medical College of Virginia/VCU, Richmond, VA 23298-0281. Key Words: Myocardial infarction, myoglobin, creatine kinase, coronary artery disease. Copyright © 1997 by W.B. Saunders Company 0735-6757/97/1501-000355.00/0

14

tests or procedures that can accurately and rapidly diagnose MI could therefore lead to more efficient use of limited resources and aid in the more accurate triage of these patients. Until recently, serial quantitative analysis of serum creatine kinase (CK) and its predominant cardiac isoenzyme, CK-MB, by agarose gel electrophoresis has been considered the most sensitive, specific, and cost-effective way to diagnose MI. 3 However, this assay is labor-, skill-, and time-intensive, making rapid turnaround time difficult. Samples are often "batched," with results reported only once or twice daily, leading to lengthened intensive care stays as patients await return of the results. Newer assay systems based on specific monoclonal antibodies that measure the actual concentration of CK-MB rather than total CK enzymatic activity or relative electrophoretic distributions of the isoenzymes have been recently developed. This allows the advantages of faster measurement with rapid turnaround time and 24-hour availability in conjunction with higher sensitivity. 4,5 Myoglobin is another biochemical marker that is also used for the detection of myocardial necrosis. This lowmolecular-weight protein is rapidly released from injured myocardium, with concentrations increasing as early as 1 to 3 hours after symptom onset. 6,7 Myoglobin can also be determined via rapid-turnaround monoclonal antibodybased assays. However, its use may be limited by its lower specificity compared to CK-MB mass, as it does not distinguish skeletal muscle injury from myocardial damage. 8 Previous studies have used serial sampling of either myoglobin or CK-MB mass for the early detection of myocardial damage. None, however, have used both in complementary fashion for improved early diagnosis. Because each of these myocardial markers offers certain advantages, we hypothesized that the combination of rapid myoglobin and CK-MB mass assays could decrease the time necessary to accurately determine whether a patient has a MI. The advantage of earlier, more rapid diagnosis of patients with MI may allow for earlier aggressive treatment of patients with actual MI and, in an era of limited resources, more appropriate triage of patients without MI to nonintensive care beds. METHODS This study was approved by the Committee for the Conduct of Human Research of the Medical College of Virginia/Virginia Commonwealth University Investigational Review Board. Patients presenting to the Medical College of Virginia Hospitals ED with

KONTOS ET AL • COMBINATION MARKERS FOR AMI

chest pain subsequently admitted to the coronary care unit for possible MI were eligible for entry. Hospital admission was at the discretion of the ED physician and was based on clinical, historical, and electrocardiographic criteria. Blood samples drawn on admission (referred to as 0 hours) and 4 hours later were sent for measurement of CK-MB mass and myoglobin. In addition, blood samples were sent from admission and 8, 16, and 24 hours later for measurement of total CK activity and percent CK-MB, which was the customary assay at the time of the study. All patients had standard 12-lead ECGs performed on admission and 24 hours later. Diagnosis of MI was based on World Health Organization criteria,9 which included having at least two of the following: evolving diagnostic ECG changes, elevation of both CK and CK-MB above normal values, and prolonged cardiac chest pain. Presenting ECGs were reviewed by two experienced cardiologists blinded to the clinical events and laboratory results and were placed into one of three categories: (l) infarction, if there were significant Q waves or ST segment elevation of ->1 ram; (2) ischemia, if there was ST segment depression of -->l mm or significant (-->3 ram) T wave inversion; (3) no evidence of infarction or ischemia, all other ECG findings.

Assays The Baxter Stratus II (Baxter Diagnostics, Miami, FL) myoglobin procedure is a sandwich enzyme immunoassay incorporating support-bound mouse monoclonal antimyoglobin and an alkaline phosphatase labeled Fah' mouse monoclonal antimyoglobin conjugate. A substrate wash buffer radially diffuses unbound conjugate from the reaction area and the enzymatic reaction rate is measured by front surface florescence. Reagents can be used directly from refrigerated storage and reconstitution is not required. Serum myoglobin and CK-MB assays on the Stratus II require 8 minutes of analyzer time for the first sample result, with subsequent batch results produced at 1-minute intervals. However, these two assays must be processed in separate batches. Between-run imprecision for the myoglobin assay was 2.1% at 43 pg/L and 6.3% at 406 pg/L. Results were evaluated based on the manufacturer's suggested upper reference limit of 90 pg/L. The Behring Opus Plus myoglobin procedure (Behring Diagnostics, Boston, MA) is similar in principle to the Stratus assay. Monoclonal antimyoglobin antibodies are attached to a glass fiber solid phase. A conjugate of polyclonal antimyoglobirdalkaline phosphatase is dispensed after the patient's serum myoglobin has bound to the solid phase antibody. The substrate/wash solution washes away unbound material. The rate of formation of fluorescent product is directly proportional to the concentration of myoglobin in the patient's serum. The Behring's suggested upper reference limits of 92 pg/L for men and 76 pg/L for women was utilized for evaluation of myoglobin results. Opus Plus results are available approximately 20 to 25 minutes after initiating a run, and both CK-MB mass and myoglobin assays can be processed in the same run. Between-run imprecision for the myoglobin assay was 7.0% at 94 gg/L, and 7.6% at 334 lag/L. Total serum immunoreactive CK-MB mass was determined by fluorometric enzyme immunoassay (Stratus II or Opus Plus) and results were evaluated based on the manufacturer's suggested upper reference limit of 4.7 gg/L (Stratus II) or 8.0 ~g/L (Opus Plus). Total serum CK activity (Ectachem; Eastman-Kodak, Rochester, NY) was determined using N-acetylcisteine activation without pretreatment. Serum CK-MB isoenzyme activity was determined in samples with total CK activity > 40 1U/L by agarose gel electrophoresis with fluorescence detection (REP System # 3070; Helena Laboratories, Beaumont, TX). CK-MB activity was expressed as a percentage of total CK activity and results of >6% in

15

association with total CK greater than two times upper limit of normal were considered diagnostic of myocardial infarction.

Statistical Analysis Results are presented as mean _+ SEM. Sensitivity, specificity, and positive and negative predictive values were calculated in the standard fashion. For calculations in which more than one test result was used, sensitivity was calculated as: (patients with a positive test result having myocardial infarction)/(all patients with myocardial infarction). Specificity was calculated as: (patients without MI having all test results negative)/(all patients without myocardial infarction). Positive and negative predictive values were calculated in a similar fashion. Categorical data were compared using Chi-square analysis, and Student's t test was used for comparison of mean levels between groups. A P value -< .05 was considered statistically significant.

RESULTS Samples were collected from a total of 110 patients. Nine were excluded secondary to improper sample collection, leaving 101 patients who formed the study cohort. Of the 101 patients, 20 (13 men, 7 women, mean age 59.7 -+ 3.0 years) sustained a MI, while 81 (43 men, 38 women, mean age 57.9 ___ 1.4 years) did not.

Electrocardiogram Twelve of the 20 (60%) patients who were subsequently diagnosed with M I had an initial E C G consistent with infarction. An additional 3 (15%) had E C G evidence of ischemia. Fourteen of the 81 (17%) patients without M I had an E C G consistent with MI, while evidence of ischemia was seen in 11 (14%). The overall sensitivity of the E C G for diagnosis of myocardial infarction was 60%, with a specificity of 83%, positive predictive value o f 46%, and negative predictive value of 89%. The sensitivity of the E C G was increased to 75% if ischemic ECGs were combined with those of infarction; however, specificity decreased to 69%.

Myoglobin In patients with MI, the mean concentration of the 0-hour myoglobin was 232 + 62 ~g/L, increasing to 448 _+ 123 ~g/L at 4 hours (P = . 13). In comparison, mean myoglobin concentration in patients without M I was significantly lower, 70 _+ 13 ~ag/L at 0 hours and 72 _+ 11 ~g/L at 4 hours (P < .001 for both 0- and 4-hour results compared with similar results in patients without MI) (Figure 1). The first myoglobin level was positive in 14 (70%) and negative in 6 (30%) patients with MI, whereas it was positive in 13 (16%) and negative in 68 (84%) patients without MI. The 4-hour myoglobin level was positive in 17 (85%) and negative in 3 (15%) patients with M I compared with 16 (20%) and 65 (80%) patients without MI, respectively. The diagnostic sensitivity of the initial myoglobin level for M I was 70%, which increased to 85% at 4 hours, with accompanying specificities of 81% and 81%, respectively (Table 1). Eighteen (90%) patients with M I had at least 1 of the 2 myoglobin levels elevated, compared with 20 (25%) of the patients without MI. Sixty-four (79%) patients without M I

16

-g e.O O O1 O >,

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 15, Number 1 • January 1997

1200

160

1000

140

~

120

800 600

""

80

400

l

100

i °° 40

200

20

Rule In

Rule Out

Rule In

0'hr

Rule Out

0

Rule In

Rule Out

0 hr

4'hr

FIGURE 1. Myoglobin results at 0 and 4 hours after admission for patients who ruled in and ruled out for MI. Lines indicate the 10%, 25%, 50%, 75%, and 90% levels. Values below the 10th or above the 90th percentiles are indicated by closed circles. had two negative myoglobin results, while only 2 (10%) patients with MI had two negative myoglobin results. The overall negative predictive value for MI of two negative myoglobin results was 97%. The sensitivity of having either the 0- or 4-hour myoglobin test result positive was 90%, with a specificity of 75%, positive predictive value of 47%, and negative predictive of 97%. CK-MB Mass

Rule In

Rule Out

4 hr

FIGURE 2. CK-MB mass results at 0 hour and 4 hours after admission for patients who ruled in and ruled out for MI. Lines indicate the 10%, 25%, 50%, 75%, and 90% levels. Values below the 10th or above the 90th percentiles are indicated by closed circles. both samples were 100% (Table 2). The overall negative predictive value of having both the 0-hour and 4-hour CK-MB mass assays negative was 99%. The one patient with a low 0- and 4-hour CK-MB mass had a peak total CK by electrophoresis of 162 U/L with 9% MB and an ECG consistent with anterior MI, with 0- and 4-hour myoglobin measurements of 200 and 160 ~g/L, respectively. A subsequent coronary angiogram showed significant left main and two vessel disease. Although the total CK concentration was not above the normal range, criteria for MI were met based on serial ECG changes associated with typical symptoms.

The mean CK-MB mass of the 0-hour sample was significantly higher in patients with MI (21.7 _+ 12.4 og/L) compared with patients without MI (1.9 + 0.2 ~ag/L) (P < .001) (Figure 2). At 4 hours the mean CK-MB mass significantly increased to 53.9 + 12.9 jag/L in patients with MI (P < .001), versus an insignificant increase in patients without MI (2.0 + 0.2 ~g/L). Similarly, the mean 0- and 4-hour total CK concentrations in patients without MI were significantly lower (P < .01 for 0- and 4-hour results) (0 hours, 147 + 22 U/L; 4 hours, 136 +- 20 U/L) than in patients with MI (0 hours, 508 + 267 U/L; 4 hours, 939 +- 417 U/L). Only 6 of the 20 (30%) patients with MI had an initial CK-MB mass that was positive, while at 4 hours the CK-MB mass was positive in 18 (90%) patients. None of the 81 patients without MI had an elevated 0-hour or 4-hour CK-MB mass. The sensitivity of the first CK-MB mass for MI was 30%, which was significantly lower (P < .001) compared with the 4-hour CK-MB mass sensitivity of 90%. The specificities of

The combination of CK-MB mass and myoglobin was used to improve the sensitivity for the diagnosis of MI. Seventeen of the 20 (85%) patients with MI had 1 of these elevated at 0 hours, whereas only 3 (15%) had both an initial CK-MB mass and myoglobin negative. Two of these 3 patients presented very early with the initial blood samples taken less than 2 hours after the onset of symptoms. The third patient had 3 hours of chest pain before the initial blood sample. At 4 hours, all of the patients with MI had at least one of the 0-hour or 4-hour CK-MB mass or myoglobin results elevated. Sixty-one of the 81 (75%) patients without MI had all 4 measurements negative. The remaining 20 patients (25%) had a least 1 false positive marker; in all cases this was secondary to an elevated myoglobin level, not CK-MB mass. The sensitivity and specificity of combining

TABLE 1. Myoglobin Predictive Indices

TABLE 2. CK-MB Mass Predictive Indices

Sensitivity Specificity Positive predictive value Negative predictive value

First (0 hour)

Second (4 hour)

Either Positive

70 81 47 92

85 81 52 96

90 75 47 97

NOTE: Predictive indices of 0 and 4 hours, and combination of the 0-hour and 4-hour myoglobin results,

Combination

Sensitivity Specificity Positive predictive value Negative predictive value

First (0 hour)

Second (4 hour)

Either Positive

30 100 100 85

90 100 100 98

95 100 100 99

NOTE: Predictive indices of 0 and 4 hours, and combination of 0-hour and 4-hour CK-MB mass.

KONTOS ET AL • COMBINATION MARKERS FOR AMI

17

both 0-hour biochemical markers was 85% and 80%, respectively. The positive predictive value was 52%, and the negative predictive value (both initial assays negative) was 96% (Table 3). Combining the results of both the 0-hour and 4-hour CK-MB mass and myoglobin results improved the sensitivity to 100%, with an increase in the negative predictive value to 100%, as none of the patients with all 4 markers negative had a MI. When patients having ECGs consistent with MI were excluded and patients with ischemic changes were considered diagnostic of MI, the ECG sensitivity fell to 38%. In comparison, the combination of the 0-hour CK-MB mass and myoglobin was 75% with a specificity of 82%. When patients having ECGs consistent with MI or ischemia were excluded, the 0-hour sensitivity of the combination of myoglobin and CK-MB mass was 80%, with a specificity of 84%, while the combination of both the 0-hour and 4-hour myoglobin and CK-MB mass results had a sensitivity and specificity of 100%. By definition, the sensitivity of the ECG was 0%.

DISCUSSION This study found that the combination of serum CK-MB mass and myoglobin drawn at the time of initial presentation to the ED and again 4 hours later in patients with presumed cardiac symptoms has a higher sensitivity and negative predictive value for diagnosing acute MI than either marker alone. No patient with a normal myoglobin and CK-MB mass at 0 and 4 hours was subsequently diagnosed with MI. The advantage of using the combination of markers was also evident when just the 0-hour samples were analyzed; despite the initial myoglobin result having a higher sensitivity than CK-MB mass, almost a third of patients with MI remained undiagnosed. By using the combination of either an initial abnormal myoglobin level or abnormal CK-MB mass, sensitivity of the initial markers was increased from 70% to 85% without loss of specificity or positive or negative predictive value. Importantly, diagnostic accuracy of the markers was not affected when patients with diagnostic ECGs were excluded, as the sensitivity and specificities of the combination of CK-MB mass and myoglobin remained high when the ECG was nondiagnostic. Myoglobin appears in the serum as early as 1 to 2 hours after the onset of myocardial cell damage. 6,7 Earlier diagnosis of MI is possible because elevated serum levels occur 2 to 4 hours before CK-MB mass elevations. ~°,H Gibler et al~0 compared myoglobin to CK-MB mass in 59 patients presenting to the ED with presumed cardiac chest pain. They TABLE 3. Predictive Indices Using the Combination of CK-MB Mass and Myoglobin

Sensitivity Specificity Positive predictive value Negative predictive value

Either First 2 Positive

Any Positive

85 80 52 96

100 75 50 100

NOTE: Predictive indices with either the 0-hour myogiobin or CK-MB mass positive, and if the 0-hour or 4-hour CK-MB mass or myoglobin measurements were elevated.

reported that for diagnosis of MI, the initial myoglobin level had a significantly higher sensitivity, 61%, versus 14% for CK-MB, though 8 of 21 (38%) patients with MI had a normal initial myoglobin level. By 3 hours the sensitivities of both tests had increased, with no significant difference between the two. Other studies have also demonstrated the earlier diagnostic ability of myoglobin compared to CK-MB mass.ll. 12 Although its rapid appearance in the serum after onset of necrosis would appear to make it an ideal marker for early identification, a significant number of patients with MI do not have elevated initial myoglobin values. I°-I2It is clear that serial samples are necessary to accurately exclude MI. Another limitation of myoglobin is that it lacks specificity in some patient populations; levels increase with either myocardial or skeletal muscle damage, so even minor amounts of skeletal muscle injury can cause interpretative problems. 8,13 Finally, myoglobin's rapid clearance may result in normal serum values in patients presenting late after onset of MI.~3 Newer immunochemical methods have been developed for measuring CK-MB mass that are capable of detecting smaller amounts of CK-MB than older electrophoretic methods, resulting in a higher sensitivity. This increased sensitivity leads to earlier diagnosis of MI. Gibler et aP performed serial CK-MB mass measurements in 183 patients presenting to the ED with chest pain using 4 different immunochemical assays and compared them with an electrophoretic determination. The initial CK-MB mass had only moderate sensitivity, with a mean value of 54% among the 4 assays. By 3 hours, the sensitivity had increased dramatically, to 95%, with a concomitant specificity of 94%. This was significantly higher than the electrophoretic assay completed at the same time which had only a 35% sensitivity and 88% specificity. In another study, Mair et al 5 compared an immunoassay for CK-MB mass to a less direct immunoinhibitory method for measurement of CK-MB activity for the diagnosis of MI. They found that direct measurement of CK-MB mass was more sensitive, with the first serum elevation occurring more than 1 hour before the appearance of CK-MB activity. The sensitivity of CK-MB mass was 71% in samples drawn between 2 and 4 hours, and increased to 100% in samples drawn 4 to 6 hours after presentation. This was significantly higher than the sensitivities of 23% and 33%, respectively, for CK-MB activity at those same time periods. Our results are consistent with these studies, as the initial CK-MB mass had a sensitivity of 30%, while the 4-hour CK-MB mass was elevated in almost all the patients with MI. Decisions for triage or treatment based on only 1 sample have the potential to result in significant diagnostic errors. By combining serial results, diagnostic capability can be improved. Brogan et al 1I compared the sensitivity and specificity of CK-MB mass to myoglobin measured on admission and 1 hour later in 189 patients presenting to the ED with chest pain. There was a significantly higher sensitivity of myoglobin than CK-MB mass for diagnosis of M1 using either the 0-hour (55% v 23%) or 1-hour results (73% v 41%). However, myoglobin sensitivity was significantly increased to 91% if diagnosis was based on the presence of an elevation of either the 0- or 1-hour sample. Other studies have found similar results. I:,H We combined

18

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 15, Number 1 • January 1997

measurements of CK-MB mass and myoglobin at two separate time points because that offered important advantages over measurement of a single marker alone. By taking advantage of the different release kinetics of these two proteins, with the earlier appearance of myoglobin and the slightly later but more sustained release of CK-MB, early and consistent identification of patients with MI is possible. The combination of myoglobin and CK-MB mass leads to a reduction in the time necessary to confidently make this diagnosis. There are a number of reasons why markers of myocardial necrosis are important in the initial evaluation of patients presenting to the ED with chest pain. Serial sampling with myoglobin and CK-MB mass can help avoid inadvertent discharge of patients with MI. Although the number of patients mistakenly discharged with MI is small, ranging from 2% to 8%, a significant number suffer serious complications. 15-18In addition, a substantial amount of malpractice dollars awarded are associated with misdiagnosis of MI by ED physicians. ~6 Therefore, small increases in correct diagnoses can become clinically and financially very significant. Because of the large number of patients seen in EDs for chest pain, maintaining specificity is also important in order to not significantly increase the number of patients incorrectly diagnosed with myocardial infarction. The combination of initial myoglobin and CK-MB mass was successful for this purpose because the negative predictive value increased from 92% to 96%, in concert with an increase in sensitivity. The sensitivity and specificity remained high even after exclusion of patients with diagnostic ECGs. Early reperfusion has been shown to lead to significant myocardial salvage 19 and decreased mortality.2° Until recently, the ECG has been the only method for early accurate diagnosis of myocardial necrosis; however, the initial ECG is diagnostic in only 50% to 60% of patients. While patients having nondiagnostic ECGs can still have significant myocardial necrosis, 2~treatment with thrombolyfics has not been shown to be efficacious, presumably because of a "dilutional effect" of a mixture of patients with transmural or subendocardial infarction, unstable angina, and noncardiac chest pain. 22 Identification of rapidly released myocardial markers, such as myoglobin and CK-MB mass, allow early diagnosis of MI, changing it from the previous retrospective diagnosis to a prospective one. Although imaging techniques, such as echocardiography and nuclear imaging, have been applied for early diagnosis of MI, they are unable to distinguish ischemia from new or old infarction. When combined with an elevated myoglobin and/or CK-MB mass, accurate identification of patients with a high probability of ongoing myocardial necrosis may facilitate early intervention, increasing myocardial salvage. A recent small study23 found a reduction in infarct size in patients with nondiagnostic ECGs having evidence of ongoing ischemia by technetium-99m-sestamibi imaging who were treated with early reperfusion. Although promising, confirmation in larger studies will be needed. Because release of biochemical markers occurs only after myocardial necrosis, most patients with unstable angina will have markers within the normal rangeY ,25 Because these patients may still be at risk for later MI and death, use of markers should not be the sole basis for deciding on

admission or discharge, but must be interpreted in context with the clinical and ECG findings. However, rapid exclusion of MI does allow for selection of a population of patients at lower risk for short-term adverse outcomes,26 and could lead to more appropriate triage to less costly stepdown units rather than admission to intensive care. While rapid identification of patients with MI is critical, the increased pressure to reduce costs may make rapid exclusion just as important. The majority of patients admitted with possible cardiac ischemia have neither MI nor unstable angina. 2 Despite this, hospital stays are frequently extended awaiting further diagnostic tests. 27,28 By rapidly excluding MI, expedient diagnostic testing can be performed with low risk for complications, 29 leading to early discharge in patients who prove to have noncardiac chest pain. Significant cost reductions may result from the decrease in both intensive care and total hospital admission time. 3° This study was limited in that it may not be extrapolated to all patients presenting to the ED with cardiac symptoms, because only patients that were admitted were studied. However, the proportion of patients with infarction in this study, 20%, is similar to the incidence of MI in other studies, 2 so should be applicable to patients with symptoms severe enough to warrant admission. Any study evaluating new markers for diagnosing MI is limited in that diagnostic accuracy is compared to MB fraction measured electrophoretically, which may be less sensitive for recognizing low amounts of myocardial necrosis. This may result in an apparent increase in the false positive rate of the newer marker, despite its greater sensitivity. In conclusion, we found that the use of a combination of both CK-MB mass and myoglobin led to early diagnosis of MI with a high sensitivity and negative predictive value. Earlier diagnosis can lead to earlier, more appropriate treatment in patients with MI and more appropriate triage of patients with a low likelihood of myocardial ischemia. Whether the earlier confirmation or exclusion of MI can result in alteration in management of these patients will need to be evaluated in future studies.

REFERENCES 1. Selker HP: Coronary care unit triage decision aids: How do we know when they work? Am J Med 1989;87:491-493 2. Lee TH, Goldsman L: The CCU turns 25: Historical trends and future directions. Ann Intern Med 1988; 108:887-894 3. Lee TH, Rouan GW, Weisberg MC, et al: Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 hours of hospitalization. Ann Intern Meal 1987;106:181-186 4. Gibler WB, Lewis LM, Erb RE, et al: Early detection of acute myocardial infarction in patients presenting with chest pain and nondiagnostic ECGs: Serial CK-MB sampling in the emergency department. Ann Emerg Med 1990; 19:1359-1366 5. Mair J, Artner-Dworzak E, Dienstl A, et al: Early detection of acute myocardial infarction by measurement of mass concentration of creatine kinase-MB. Am J Cardio11991 ;68:1545-1550 6. Drexel H, Dworzak E, Kirchmain W, et al: Myoglobinemia in the early phase of acute myocardial infarction. Am Heart J 1983; 105:642651 7. Chapelle JP: Serum myoglobin determinations in the assessment of acute myocardial infarction. Eur Heart J 1982;3:122-129 8. Puleo P, Roberts R: Early biochemical markers. Cardiovasc Clin 1989;20:143-154 9. Gillum RF, Fortmann SP, Prineas RJ, et al: International diagnostic criteria for acute myocardial infarction and acute stroke. Am Heart J 1984; 108:150-157

KONTOS ET AL • COMBINATION MARKERS FOR AMI

10. Gibler WB, Gibler CD, Weinshenker E, et al: Myoglobin as an early indicator of acute myocardial infarction. Ann Emerg Med 1987; 16:851-856 11. Brogan GX, Friedman S, McCuskey C, et al: Evaluation of a new rapid quantitative immunoassay for serum myoglobin versus CK-MB for ruling out acute myocardial infarction in the emergency department. Ann Emerg Med 1994;24:665-671 12. Tucker JF, Collins RA, Anderson AJ, et al: Value of serial myoglobin levels in the early diagnosis of patients admitted for acute myocardial infarction. Ann Emerg Med 1994;24:704-708 13. Vaidya HC: Myoglobin. Lab Med 1992;23:306-310 14. Grenadier E, Keidar S, Kahana L, et al: The roles of serum myoglebin, total CPK, and CK-MB isoenzyme in the acute phase of myocardial infarction. Am Heart J 1983;105:408-416 15. Lee TH, Rouan GW, Weisberg MC, et al: Clinical characteristics and natural history of patients with acute myocardial infarction sent home from the emergency room. Am J Cardiol 1987;60:219-224 16~ Rusnak RA, Stair TO, Hansen K, et al: Litigation against the emergency physician: Common features in cases of missed myocardial infarction. Ann Emerg Med 1989;18:1029-1034 17. Hedges JR, Rouan GW, Toltzis R, et al: Use of cardiac enzymes identifies patients with acute myocardial infarction otherwise unrecognized in the emergency department. Ann Emerg Med 1987; 16:248-252 18. Schor S, Behar S, Modan B, et al: Dispositions of presumed coronary patients from an emergency room: A follow-up study. JAMA 1976;236:941-943 19. Gibbons RJ, Holmes DR, Reeder GS, et al: Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993;328:685-691 20. The GUSTO Angiographic Investigators: Effect of tissue plasminogen activator, streptokinase, or both on coronary-artery patency, ventricular function, and survival after acute myocardial infarction. N Engl J Med 1993;329:1615-1622

19

21. Christian TF, Clements IP, Gibbons RJ: Noninvasive identification of myocardium at risk in patients with acute myocardial infarction and nondiagnostic electrocardiograms with technetium-99m-sestamibi. Circulation 1991;83:1615-1620 22. Rouan GW, Lee TH, Cook El=, et al: Clinical characteristics and outcome of acute myocardial infarction in patients with initially normal or nonspecific electrocardiograms. Am J Cardiol 1989;64:10871092 23. O'Keefe JH, Sayed-Taha K, Gibson W, et al: Do patients with left circumflex coronary artery-related acute myocardial infarction without ST-segment elevation benefit from reperfusion therapy? Am J Cardiol 1995;75:718-720 24. Botker HE, Ravkilde J, Sogaard P, et al: Gradation of unstable angina based on a sensitive immunoassay for serum creatine kinase MB. Br Heart J 1991;65:72-76 25. Ravkilde J, Nossen H, Herder M, et al: Independentprognostic value of serum creatine kinase isoenzyme MB mass, cardiac troponin T and myosin light chain levels in suspected acute myocardial infarction. J Am Coil Cardio11995;25:574-581 26. Gaspoz JM, Lee TH, Cook EF, et al: Outcome of patients who were admitted to a new short-stay unit to "rule out" myocardial infarction. Am J Cardio11991 ;68:145-149 27. UdvarhelyJ IS, Goldman L, KomaroffAL, et al: Determinants of resource utilization for patients admitted for evaluation of acute chest pain. J Gen Intern Med 1992;7:1-10 28. Selker HE Beshansky JR, Pauker SG, et al: The epidemiology of delays in a teaching hospital: The development and use of a tool that detects unnecessary hospital days. Med Care 1989;27:112-129 29. Gibler WB, Runyon JP, Levy RC, et al: A rapid diagnostic and treatment center for patients with chest pain in the emergency department. Ann Emerg Med 1995;25:1-8 30. Gaspoz JM, Lee TH, Weinstein MC, et al: Cost-effectiveness of a new short-stay unit to "rule out" acute myocardial infarction in low risk patients. J Am Coil Cardiol 1994;24:1249-1259