American Journal of Emergency Medicine (2012) 30, 188-190

Brief Report

Creatine kinase-MB does not add additional benefit to a negative troponin in the evaluation of chest pain^?

Kathryn A. Volz MD^a, Daniel C. McGillicuddy MD^a,

Gary L. Horowitz MD^b, Leon D. Sanchez MD, MPH^a,?

^aDepartment of Emergency Medicine, Beth Israel Deaconess Medical Center, Boston, MA 02215, USA

^bDepartment of Pathology, Beth Israel Deaconess Medical Center, Boston, MA 02215, USA

Received 12 July 2010; revised 8 September 2010; accepted 15 October 2010

Abstract

Objective: The aim of this study was to determine whether current troponin assay alone can be used for initial screening for acute myocardial infarction (AMI) and whether creatine kinase-MB (CK-MB) can safely be eliminated from this evaluation in the emergency department (ED).

Methods: A retrospective cohort study of patients who had Cardiac troponin T (Roche, Basel, Switzerland) and CK-MB ordered at an urban academic level 1 trauma center with more than 55,000 annual visits. Patients with Troponin testing in the ED were identified over a period of 12 months, and corresponding CK-MB indexes were examined identifying patients with negative troponins (b0.01) and positive CK-MB indexes (N6.0). In these patients, further cardiac markers, hospital course, and 30-day mortality were then evaluated. A 99% confidence interval around point estimate was used in data analysis. Results: During the study period, there were 11,092 separate ED patient encounters where a patient had at least one troponin resulted. Most (97.9%) of the samples had an associated CK-MB ordered. There were 7545 initial negative troponins representing 68% of all initial samples. Seven of these had an associated positive MB index. When subsequent troponins were evaluated, an additional 4910 negative troponins were identified, with 4 patients having a positive MB. None of these 11 patients were judged to have ruled in for AMI by the treating physicians. The rate of true-positive CK-MB index with negative troponin was 0% (99% confidence interval, 0-0.04%).

Conclusion: Our results suggest that CK-MB is not necessary in the initial screening for AMI and may safely be omitted in patients with negative troponins.

(C) 2012

Introduction

Historically, creatine kinase-MB (CK-MB) has been used as an early identifier of non-ST-elevation acute myocardial

^? Poster presentation at Society for Academic Emergency Medicine (SAEM) 2010 Annual Meeting and New England Regional SAEM 2010.

* Corresponding author. Fax: +1 617 754 2350.

E-mail address: [email protected] (L.D. Sanchez).

infarction (AMI). Troponin has been shown to be signifi- cantly more sensitive and specific in identifying myocardial ischemia [1,2]. Consensus recommendations from the American College of Cardiology, the European Society of Cardiology, the World Heart Federation, and the American Heart Association state that the preferred biomarker for Myocardial necrosis is cardiac troponin [3].

Studies have suggested that given these findings, CK-MB is no longer necessary in the evaluation of AMI [4]. Despite these recommendations, many physicians continue to order

0735-6757/$ - see front matter (C) 2012 doi:10.1016/j.ajem.2010.10.016

CK-MB’s role in the evaluation of chest pain 189

both troponin and CK-MB in the workup of AMI in the emergency department (ED). This study attempted to determine whether it is safe to remove CK-MB from the initial screening for AMI in the ED.

Methods

As part of our ongoing quality improvement program, we conducted a retrospective cohort study. The study was conducted at an urban academic level 1 trauma center with more than 55,000 annual visits. All patients with troponin testing in the ED were identified over a period of 12 months. Measurements of cardiac troponin T, CK-MB, and total CK were performed with Roche Diagnostics equipment and reagents. The CK-MB index was calculated as CK-MB divided by total CK, and the threshold for positive results in our laboratory was established as 6.0%. Patients with an initial nonnegative troponin were excluded from further analysis. Serial cardiac biomarkers of all patients with initial negative troponins that were performed in the ED, in ED observation, and as inpatients were examined. Again, patients with negative troponins but positive CK-MBs were identified. Data were entered into a Microsoft Excel 2003 (Redmond, WA) database. Data were analyzed using SPSS14 (Chicago, IL). The project was reviewed by our hospital institutional review board.

In patients with negative troponin but positive CK-MB index, hospital course, and 30-day mortality were evaluated. Descriptive statistics, confidence intervals, chi square are reported when appropriate. To assure more conservative sensitivity estimates, a more restrictive 99% confidence interval around point estimate was used in data analysis.

Results

Patient encounters were retrospectively identified based on troponin results in the laboratory database. There were 11,092 patient visits with a troponin ordered from the ED during the study period; 54.5% of the patients were female; with a mean age of 68 years for female and 64 years for male. There were 11,718 troponins sent from the ED over 12 months, with 97.9% associated CK-MBs sent. This repre- sented 11,092 separate patient encounters. There were 7545 initial negative troponins identified, or 68% of all initial samples for the study period; 67.6% of these patients were admitted to the hospital. The leading ED diagnosis for these patients was chest pain/acute coronary syndrome (32.2% of patients), followed by dizziness/syncope (8.6%), abdominal

pain (4%), and atrial fibrillation or flutter (2.5%).

Of the initial 7545 negative troponins sent from the ED, 7 had associated positive CK-MB indexes. Of the patients with negative troponin and positive CK-MB indexes, none ruled in for myocardial infarction and were judged to be false

positives by treating clinicians. From the review of the medical records, none of these patients underwent cardiac catheterization during their admission, and 30-day mortality in these 7 patients was zero.

Expanding analysis to subsequent samples drawn on these patients within 12 hours even outside the ED (as inpatients or in ED observation) yielded an additional 4910 negative troponins, increasing the number of negative samples to 12,455 of a total of 17,817 samples (69.9%). An additional 4 patients with negative troponins but positive CK-MB index were identified. Based on a chart review, none of these additional patients were judged to have had a true- positive CK-MB or a myocardial infarction by the treating clinicians. Four of the 11 patients had Stress testing during their admission. None of the 11 patients were discharged with the diagnosis of myocardial infarction. One patient was lost to follow-up after discharge. That patient was searched for in the social security death index and is not listed in the death records. All other patients had at least one more note on the medical record at 30 days after their visit. The rate of true-positive CK-MB index with a negative troponin was 0% (99% confidence interval, 0-0.04%). The patients’ inpatient course and management were reviewed by a board-certified cardiologist for any quality assurance issues. This person was blinded to the purpose of the study. No problems with management were identified, and the reviewer agreed with the treating clinicians conclusions.

Discussion

Research has demonstrated that CK-MB is less sensitive and specific than troponin assays [1,2]. Our retrospective review of a year worth of data with 11,092 separate patient encounters showed no cases where a negative troponin was associated with a clinically relevant positive CK-MB index. Only 0.08% of the negative troponin samples or 0.13% of patient encounters had an associated positive CK-MB index, and all of these were judged to be false positives. Routine co- ordering of CK-MB index added nothing to the evaluation of AMI. Converting the concurrent ordering of CK-MB to a reflex test only ordered if the troponin is not negative would eliminate close to 70% of all CK-MBs currently ordered at our institution. Neither of these tests should be used as the sole method of ruling out the presence of cardiac disease or acute coronary syndromes.

An additional benefit suggested by some authors is the smaller variation based on race and sex of some troponin assays when compared with CK-MB [4,5]. Most laboratories do not adjust their reference ranges for these tests based on patient characteristics.

Although the elimination of CK-MB has been advocated by others, the 97.9% rate of co-ordering at our institution reflects that this has not crossed over into our clinical practice [4-6]. We believe that this may be a common occurrence in other EDs [7].

190 K.A. Volz et al.

Limitations

This was a retrospective study, and as such, it is subject to the standard bias associated with retrospective studies including incomplete records, enrollment bias, and patients lost to follow-up. Another limitation is that we did not evaluate the usefulness of CK-MB in patients with intermediate and positive troponins. We did not follow up patients with both negative CK-MB and troponin assays. This follow-up was not within the scope of this project.

Given that different hospitals use different troponin assays, the particular characteristics of the test may yield differences in the results. The particular assay used will also have an impact on both relative cost and Time Savings. As a single-center study, the results are subject to local practice pattern and patient population biases.

Conclusions

Our study shows that it is safe to eliminate CK-MB in the evaluation of AMI if the troponin is negative. We conclude that CK-MB should be removed from the initial evaluation of AMI in the ED, as it appears to be of no additional value to the use of troponin alone. Troponin assay alone is appropriate in screening for AMI.

When using troponin assays as an initial screening test in the evaluation of AMI, the concurrent use of CK-MB assays does

not identify any AMI that would have been missed by using troponin alone. If troponin assays are being used, we conclude that CK-MB should be removed from the initial evaluation of AMI in the ED. Troponin assay alone is appropriate in screening for AMI. Neither test can rule out the presence of coronary artery disease or acute coronary syndrome without the assistance of other diagnostic modalities.

References

1. Katus HA, Remppis A, Neumann FJ, et al. diagnostic efficiency of troponin T measurements in acute myocardial infarction. Circulation 1991;83:902-12.
2. Adams JE, Bodor GS, Davila-Roman VG, et al. Cardiac troponin I: a marker with high specificity for cardiac injury. Circulation 1993;88: 101-6.
3. Jaffe AS, World Health Organization, European Society of Cardiology, American College of Cardiology. New standard for the diagnosis of acute myocardial infarction. Cardiol Rev 2011;9(6):318-22.
4. Saenger AK, Jaffe AS. Requiem for a heavyweight: the demise of creatine kinase-MB. Circulation 2008;118:2200-6.
5. Apple FS, Quist HE, Doyle PJ, et al. Plasma 99th percentile reference limits for cardiac troponin and Creatinine kinase MB mass for use with European Society of Cardiology/American College of Cardiology consensus recommendations. Clin Chem 2003;49:1331-6.
6. Thygesen K, Alpert JS, White HD. Joint ESC/ACCF/AHA/WHF task force for the redefinition of myocardial infarction. Universal Definition of Myocardial Infarction. Circulation 2007;116:2634-53.
7. Parker RB, Suter RE. Survey examines EDs’ use of cardiac markers. ACEP News 2009;28(9):8-9.