a b s t r a c t

Context: cardiac troponins T and I have replaced creatine kinase-MB (CK-MB) as the criterion standard for diag- nosing myocardial injury. However, many laboratories still routinely perform a high volume of CK-MB testing in conjunction with troponin.

Purpose: The purpose of this study is to study the clinical and financial impact of removing CK-MB from the rou- tine emergency department (ED) test menu at a large academic medical center.

Methods: Creatine kinase-MB was removed from ED ordering templates and laboratory requisitions (ie, interven- tion), although the test could still be manually ordered. Data for creatine kinase (CK), CK-MB, and troponin T (TnT) specimens ordered during a 12-month period (6 months preintervention and 6 months postintervention) (n = 14571) was downloaded from our laboratory information system. All specimens with (1) normal TnT (ie, b 0.01 ng/mL), (2) elevated CK-MB (ie, N 6.6 ng/mL), and (3) elevated CK-MB index (ie, N 5) were considered discrepant and independently reviewed by 2 ED clinicians for the presence of an acute coronary syndrome and for documentation of final diagnosis. Creatine kinase, CK-MB, and TnT ED volumes preintervention and postinter- vention were analyzed to assess laboratory cost savings.

Results: Of the 6444 cases included in the analysis, only 17 were discrepant. Of all 17 cases, no patients were diagnosed with acute coronary syndrome. After removing CK-MB from the templates and requisitions, CK-MB and CK volumes decreased by 80% and 76%, respectively, translating to annual reagent cost savings of approximately $47000.

Conclusions: Creatine kinase-MB can be removed from the routine ED test menu without adversely affecting patient care. In addition, substantial cost savings can be achieved by reducing unnecessary CK-MB testing and associated CK orders.

(C) 2014

1. Introduction

For the past decade, cardiac troponins I and T have replaced Creatine kinase-MB (CK-MB) as the Diagnostic biomarker for acute coronary syn- dromes (ACS) due to their superior sensitivity and specificity [1]. In ad- dition, cardiac troponin levels rise and remain elevated for up to 2 weeks after the onset of chest pain, thus providing a longer window for recent-past diagnosis of ACS than CK-MB levels, which return to baseline levels in 1 to 3 days. A large study conducted on 29357 ACS pa- tients from the CRUSADE registry supported the utilization of troponin as the sole marker of myocardial injury. The study found that only 10% (2988/29357) of cases were discordant (ie, troponin T [TnT] normal, CK-MB elevated) and concluded that an elevated CK-MB alone did not offer additional prognostic value [2].

* Corresponding author. Brigham and Women’s Hospital, 75 Francis St, Amory 2, Boston, MA 02115. Tel.: +617 732 6360; fax: +617 731 5872.

E-mail address: [email protected] (M.J. Tanasijevic).

However, CK-MB is still ordered along with troponin in many hospi- tals, perhaps as a combination of challenges in convincing physicians that troponin alone is sufficient in most ACS settings and CK-MB’s clin- ical utility in certain settings such as diagnosing reinfarction in patients with a recent admission for ACS [1].

In March 2011, Massachusetts General Hospital (MGH, Boston, MA), which cofounded Partners HealthCare alongside our institution, changed their hospital guidelines to restrict CK-MB ordering to only post-percutaneous coronary intervention and post-cardiac surgery pa- tients [3]. In an effort to sustain results, MGH implemented educational alerts in their computerized provider Order entry (OE) system, which reminded providers of the new guidelines at the time of ordering and tracked those that deviated from the guidelines. Their interventions de- creased the number of CK-MB tests ordered by 87% from 1106 to 139 per month [3]. Other institutions including the Mayo Clinic and the Uni- versity of Maryland have also removed CK-MB from their test menus [4]. However, few studies have examined both the clinical and financial impacts of removing CK-MB [5].

http://dx.doi.org/10.1016/j.ajem.2014.10.017

0735-6757/(C) 2014

cost analysis“>R.D. Le et al. / American Journal of Emergency Medicine 33 (2015) 72–75 73

Based on guidelines by several groups [6,7] and a growing number of studies highlighting the superiority of TnT over CK-MB in the diagnosis of ACS [1,8-10], our institution removed the CK-MB option from elec- tronic ordering templates and paper test requisitions in the emergency department (ED). We then studied the clinical and financial effects of this intervention by reviewing 6 months of preintervention and 6 months of postintervention data.

1. Methods
   1. Study site

This study was performed at Brigham and Women’s Hospital (BWH), a 777-bed, tertiary care center located in Boston, MA. The ED at the BWH, a primary receiving center for New England, had 60572 visits in 2013. In the ED, tests are ordered through an electronic OE system or di- rectly onto paper requisitions. All electronic orders are ultimately tran- scribed onto paper requisitions and then sent to the laboratory. Creatine kinase (CK), CK-MB, and TnT levels are measured by Roche Cobas c701 (CK) and e601 analyzers (CK-MB and TnT) (Roche Diagnos- tics, Indianapolis, IN). Reference ranges are less than 0.01 ng/mL and less than 6.6 ng/mL for TnT and CK-MB, respectively. The reference range for CK is 39 to 308 U/L for males and 26 to 192 U/L for females. Based on the literature, we considered a CK-MB index (CK-MB [ng/mL]/CK [U/L]

x 100) of greater than 5 elevated [11]. This study was approved by the Partners Human Research Committee.

Study design

Patient’s age; sex; and CK, CK-MB, and TnT results for tests ordered be- tween January 2013 and June 2013 (preintervention) (n = 6573) and be- tween January 2014 and June 2014 (postintervention) (n = 7998) were obtained from our internally developed laboratory information system and included/excluded from the study based on the algorithm outlined in Figure. We considered specimens in which (1) CK, CK-MB, and TnT

were ordered and resulted on the same specimen; (2) TnT was within normal limits; (3) CK-MB was elevated; and (4) CK-MB index was elevat- ed (discrepant). In addition, we reviewed cases in which TnT was normal, CK-MB was elevated, and CK was not ordered or not resulted. Patient charts from the preintervention and postintervention cohorts were inde- pendently reviewed by 2 ED physicians for presence of ACS and final diag- noses in the discrepant cases.

Intervention

Before our intervention, the CK, CK-MB, and TnT ordering options were nested together under the chemistry/cardiac marker section of the ED test requisition. In November 2013, the CK-MB option was re- moved from the paper test requisitions, the main ED electronic OE screen, and the “Chest Pain” OE template. Creatine kinase was kept on the requisition, although it was clearly separated from TnT. Old requisi- tions were removed from circulation by January 2014. It should be noted that physicians were not prevented from ordering CK-MB and were able to search for the test in ED OE and write it on the requisitions.

Cost analysis

Test volumes preintervention (January 2013 through June 2013) and postintervention (January 2014 through June 2014) were collected to cal- culate cost savings in reagents, quality control, and calibration material. Creatine kinase, CK-MB, and TnT levels are performed on a high-volume automated system processing several thousands of specimens each day; therefore, no labor savings were associated with the intervention.

1. Results

Table 1 displays the patient characteristics and the test results of all discrepant cases during the preintervention period. The percentage of discrepant cases was small (13/5854, 0.22%). No patients with ACS were missed, and almost all final diagnoses were connected with severe

All TnT and CK-MB ordered during study period (n = 6,573)

1. Only TnT (n = 569) ordered

excluded

1. Only CK-MB (n = 38) ordered
2. Test not resulted (e.g., hemolyzed, contaminated, cancelled by provider, etc.; n=112)

TnT and CK-MB included in analysis (n = 5,854)

1. Both TnT and CK-MB normal1 (n = 4,287)

excluded

1. Both TnT and CK-MB elevated (n = 266)

TnT normal, CK-MB elevated (n = 276)

1. TnT elevated, CK-MB normal

(n = 1,025)

CK-MB index normal (n = 253)

excluded

CK not ordered (n = 10)*

TnT normal, CK-MB elevated, CK-MB index elevated (n = 13)

Figure. Flowchart of exclusion and inclusion criteria for study. *Cases in which TnT was normal, CK-MB was elevated, and CK was not ordered were also reviewed. ¹Reference ranges are less than 0.01 ng/mL and less than 6.6 ng/mL for TnT and CK-MB, respectively. The reference range for CK is 39 to 308 U/L for males and 26 to 192 U/L for females. Based on the literature we considered, a CK-MB index greater than 5 is elevated.

74 R.D. Le et al. / American Journal of Emergency Medicine 33 (2015) 72–75

Table 1

Characteristics of discrepant cases preintervention (ie, TnT normal, CK-MB elevated, CK-MB index elevated; n = 13)

Patient	Age	Sex	CK-MB Indexa	CK-MB (ng/mL)a	CK (U/L)a	Subsequent TnT (ng/mL)a,b	Disposition	Final diagnosis
1	83	M	5.00	8.5	170	NA	ED observation	Vertigo
2	65	M	5.07	7.1	140	NA	Medical ICU	Chronic obstructive pulmonary disease; exacerbation
3	80	F	5.11	6.9	135	b 0.01	Trauma ICU	Fall; sternal fracture
4	77	M	5.30	9.6	181	NA	Surgical ICU	Fall; subdural hematoma
5	57	M	5.58	7.2	129	b 0.01	Cardiology	Congestive heart failure; exacerbation
6	79	M	5.63	12.6	224	b 0.01	Oncology	Pneumonia; cancer
7	55	M	5.69	9.5	167	NA	Medical ICU	Diabetic ketoacidosis
8	58	F	5.76	8	139	NA	Medical ICU	Hyponatremia; cancer
9	83	F	5.78	26.3	455	b 0.01	Medical ICU	Pneumonia; respiratory distress; sepsis
10	79	F	5.98	6.7	112	NA	External transfer	Pneumonia; new onset atrial fibrillation
11	77	F	7.09	9	127	b 0.01	Medical ICU	Hypothermia
12	75	F	7.88	6.7	85	NA	Medicine	Vertigo; ataxia
13	75	F	18.46	9.6	52	NA	Surgical ICU	Mesenteric ischemia

^a Reference ranges are less than 0.01 ng/mL and less than 6.6 ng/mL for TnT and CK-MB, respectively. The reference range for CK is 39 to 308 U/L for males and 26 to 192 U/L for females. Based on the literature we considered, a CK-MB index greater than 5 is elevated.

^b Subsequent TnT tests are those that resulted less than 24 hours since the previous test. All previous TnT results were normal.

noncardiac illness (eg, cancer, pneumonia) or trauma (eg, fall). All patients except 1 were admitted to BWH or a surrounding hospital. It should also be noted that only mild elevations in CK-MB index were observed and there were no instances in which subsequent TnT levels (n = 5) were elevated (Table 1.).

In addition, we reviewed 10 preintervention cases in which TnT was normal, CK-MB was elevated, and CK was not ordered. Similarly, we found no instances of missed ACS, only mild elevations in CK-MB index, and no subsequent elevations in TnT. All 10 final diagnoses in this group were severe noncardiac illness or trauma (data not shown). Table 2 displays the patient characteristics and test results in the postintervention period. The percentage of discrepant cases was 0.68% (4/590). Similar to the preintervention period, no ACS cases were

missed, and most of the final diagnoses were related to trauma.

The volume of CK, CK-MB, and TnT tests ordered in the ED during a 6-month period preintervention was 6242, 5995, and 6535, respective- ly, with an associated reagent cost of $42874 (Table 3). As expected, the number of CK-MB tests ordered postintervention decreased to 810 (80% decrease). The number of CK tests ordered postintervention also de- creased to 1526 (76% decrease). The volume of TnT increased by 10% postintervention. Our interventions were associated with a cost savings of $23586 during the 6 months postintervention or approximately

$47000 annually (Table 3).

1. Discussion

We found only a small number of cases (0.22%) in which TnT was normal and both CK-MB and CK-MB index were elevated in our study population. None of these discrepant cases carried the diagnosis of ACS. Coupled with the much larger percentage (17.5%) of cases in which TnT was elevated and CK-MB was normal, our findings support previous studies, which demonstrate that TnT alone is sufficient for ac- curate diagnosis of ACS.

In this study, we choose not to review cases in which CK-MB was el- evated but the TnT and CK-MB fraction were both normal (n = 253), as- suming that the elevated CK-MB in these patients was due to noncardiac injury.

Most discrepant cases in the preintervention and postinterven- tion groups had only mild elevations of the CK-MB index (15 of the 17 cases had indices <= 7.88). These mild elevations occurred primar- ily in elderly patients and may be potentially due to lower skeletal mass or nonspecific elevation in the setting of severe noncardiac illness or trauma.

The increase in TnT volumes postintervention was most likely due to providers substituting TnT for CK-MB and/or correctly ordering TnT alone. In fact, the volume of CK-MB-only orders decreased from 38 to 2 postintervention. The increase in TnT volumes had only a slight impact on the overall cost savings (Table 3).

Based on our findings, we have expanded our initiative beyond the ED. In May 2014, we removed CK-MB from our inpatient test requisi- tions and added the following statement (similar to MGH) to our inpa- tient electronic test ordering screen: “Routine measurement of CK isoenzymes (ie, CK-MB and CK) is no longer recommended. The use of CK-MB should be restricted to the following exceptions: post-percuta- neous coronary intervention and post-cardiac surgery” [3]. We have also introduced similar changes at our off-site urgent care center.

Many institutions use troponin I instead of TnT for diagnosis of ACS. Studies have shown that the analytical and Clinical performances of TnI and TnT are similar [8,12,13]. At our institution, we recently switched from TnI to TnT, and our validations corroborated published studies and demonstrated comparable accuracy of TnI and TnT for the diagnosis of ACS. Therefore, removing CK-MB in the ED setting at insti- tutions using TnI should have similar outcomes to those outlined in this study.

In conclusion, we have shown that removal of CK-MB from the rou- tine test menu in the ED reduces cost and does not negatively impact patient care.

Table 2

Characteristics of discrepant cases postintervention (ie, TnT normal, CK-MB elevated, CK-MB index elevated; n = 4)

Patient	Age	Sex	CK-MB indexa	CK-MB (ng/mL)a	CK (U/L)a	Subsequent TnT (ng/mL)a,b	Disposition	Final diagnosis
1	47	M	5.33	8.0	150	b0.01	Home	Fall; cervical spine fracture
2	85	F	6.50	8.0	123	b0.01	Cardiology	Unstable angina
3	92	F	7.76	8.3	107	b0.01	Medical ICU	Altered mental status; hyponatremia; congestive heart failure
4	85	F	10.64	8.3	78	b0.01	Trauma service	Fall; head laceration

Abbreviations: M, male; F, female; ICU, intensive care unit.

^a Reference ranges are less than 0.01 ng/mL and less than 6.6 ng/mL for TnT and CK-MB, respectively. The reference range for CK is 39 to 308 U/L for males and 26 to 192 U/L for females. Based on the literature we considered, a CK-MB index greater than 5 is elevated.

^b Subsequent TnT tests are those resulted less than 24 hours since the previous test. All previous TnT results were normal.

R.D. Le et al. / American Journal of Emergency Medicine 33 (2015) 72–75 75

Table 3

Volume and cost of ED cardiac marker testing preintervention and postintervention

1. Singh G, Baweja PS. Creatine kinase-MB: the journey to obsolescence. Am J Clin Pathol 2014;141(3):415-9.
2. Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD, Hochman JS, et al.

Preintervention

(January 2013-June 2013)

Test	Volume	Cost		Volume	Cost
CK	6242	$1748		1526	$427	$2642
CK-MB	5995	$27337		810	$3,694	$47286
TnT	6535	$13789		7188	$15167	-$2756
Total	18772	$42874		9524	$19288	$47172

References

Postintervention

(January 2014-June 2014)

Estimated annual cost savings

ACC/AHA guidelines for the management of patients with unstable angina and Non-ST-segment elevation myocardial infarction: executive summary and recom- mendations. A report of the American College of Cardiology/American Heart Associ- ation Task Force on Practice Guidelines (committee on the management of patients with unstable angina). Circulation 2000;102(10):1193-209.

1. Body R, Carley S, McDowell G, Pemberton P, Burrows G, Cook G, et al. The Manchester Acute Coronary Syndromes (MACS) decision rule for suspected cardiac chest pain: derivation and external validation. Heart 2014;100(18):1462-8.
2. Hawkins RC, Tan HL. Comparison of the diagnostic utility of CK, CK-MB (activity and mass), troponin T and troponin I in patients with suspected acute myocardial infarc- tion. Singapore Med J 1999;40(11):680-4.
3. Collinson PO, Gaze DC, Bainbridge K, Morris F, Morris B, Price A, et al. Utility of ad- mission cardiac troponin and “ischemia modified albumin” measurements for rapid evaluation and rule out of suspected acute myocardial infarction in the emer- gency department. Emerg Med J 2006;23(4):256-61.
   1. Saenger AK, Jaffe AS. Requiem for a heavyweight: the demise of creatine kinase-MB. Circulation 2008;118(21):2200-6.
   2. Newby LK, Roe MT, Chen AY, Ohman EM, Christenson RH, Pollack Jr CV, et al. Frequency and clinical implications of discordant creatine kinase-MB and troponin measurements in acute coronary syndromes. J Am Coll Cardiol 2006;47(2):312-8.
   3. Baron JM, Lewandrowski KB, Kamis IK, Singh B, Belkziz SM, Dighe AS. A novel strategy for evaluating the effects of an electronic test ordering alert message: optimizing cardiac marker use. J Pathol Inform 2012;3:3. http://dx.doi.org/10. 4103/2153-3539.93400 [Epub 2012 Feb 29].
   4. Retiring CK-MB Troponin‘s diagnostic advantages may make CK-MB redundant [Internet]. Washington, DC: AACC; 2008 [cited 07/09/2014]. Available from: http://www.aacc.org/publications/strategies/archives/2008/Pages/121108.aspx.
4. Goodman SG, Steg PG, Eagle KA, Fox KA, Lopez-Sendon J, Montalescot G, et al. The diagnostic and prognostic impact of the redefinition of acute myocardial infarction: lessons from the Global Registry of Acute coronary events . Am Heart J 2006; 151(3):654-60.
5. Lee M, editor. Basic skills in interpreting laboratory data. 5th ed. Bethesda, MD: ASHP; 2013.
6. Bhagat CI, Langton P, Lewer M, Ching S, Beilby JP. Cardiac troponin I should replace CKMB for the diagnosis of acute myocardial infarction. Ann Clin Biochem 1997;34(Pt 5):511-6.
7. D’Costa M, Fleming E, Patterson MC. Cardiac troponin I for the diagnosis of acute myocardial infraction in the emergency department. Am J Clin Pathol 1997;108 (5):550-5.