Correspondence
Subforms of Creatine Kinase MB in the Diagnosis of Myocardial Infarction
N Engl J Med 1995; 332:608-610March 2, 1995
- Article
To the Editor:
Puleo et al. (Sept. 1 issue)1 conclude that the assay of creatine kinase MB (CK-MB) subforms is superior to the conventional glass-bead assay for CK-MB. The routine CK-MB assay used by these authors is not the conventional assay and was never widely adopted. The sensitivity and analytic specificity of a CK-MB assay contribute substantially to the diagnostic cutoff of the assay and the time required for the assay to become positive. The assays of CK-MB mass using monoclonal antibodies, now in wide use, are highly sensitive and analytically specific for CK-MB, and they detect elevated levels of CK-MB in two to six hours in more than 90 percent of patients with acute myocardial infarction.2-4 In a relevant comparison with the mass assay, the subform assay was not superior with regard to the time required to reach a positive result; in addition, it was technically less satisfactory.5 If refinements of the instrumentation have substantially enhanced its analytic sensitivity and ability to interpret the data, the authors should publish a comparison of the subform and the mass assays.
Readers of the paper by Puleo et al. may infer that the use of the assay of CK-MB subforms permitted the rational triage of patients with chest pain. However, this is speculation. In the Methods section, the authors clearly state, “The diagnostic evaluation and care of all patients were conducted by physicians who were unaware of the results of the assays for CK-MB subforms.” As in other studies, the appropriate prospective and randomized use of the test in real time would determine whether admissions would be decreased and savings achieved. For the present, widespread adoption of this form of technology is not warranted on the basis of the data presented.
5 ReferencesJ.H. Keffer, M.D.
University of Texas Southwestern Medical Center, Dallas, TX 75235-90721
Puleo PR ,Meyer D ,Wathen C , et al. Use of a rapid assay of subforms of creatine kinase MB to diagnose or rule out acute myocardial infarction. N Engl J Med 1994;331:561-566
Full Text | Web of Science | Medline2
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 Cardiol 1991;68:1545-1550
CrossRef | Web of Science | Medline3
Collinson PO ,Rosalki SB ,Kuwana T , et al. Early diagnosis of acute myocardial infarction by CK-MB mass measurements. Ann Clin Biochem 1992;29:43-47
Web of Science | Medline4
Marin MM ,Teichman SL . Use of rapid serial sampling of creatine kinase MB for very early detection of myocardial infarction in patients with acute chest pain. Am Heart J 1992;123:354-361
CrossRef | Web of Science | Medline5
Bhayana V ,Cohoe S ,Leung FY ,Jablonsky G ,Henderson AR . Diagnostic evaluation of creatine kinase-2 mass and creatine kinase-3 and -2 isoform ratios in early diagnosis of acute myocardial infarction. Clin Chem 1993;39:488-495
Web of Science | Medline
To the Editor:
The choice by Puleo et al. of 14 U per liter as a fixed diagnostic cutoff for CK-MB activity is questionable for two reasons. First, CK-MB activity has well-recognized sex differences. Second, this cutoff has a high specificity (to rule in myocardial infarction), but the sensitivity is totally inadequate for ruling out infarction effectively. Therefore, the reported low sensitivities are not unexpected, and they could have been improved markedly by moving the diagnostic cutoff to a lower value of CK-MB activity.
V. Bhayana, Ph.D.
A. Ralph Henderson, M.B., Ph.D.
University Hospital, London, ON N6A 5A5, CanadaTo the Editor:
Sensitivity and specificity are used to select the most appropriate test, given the clinician's objectives. A test with a high sensitivity is ideal in screening for disease or ruling it out. One with a high specificity is used to confirm a diagnosis. Certainly, the sensitivity of 95.7 percent and the specificity of 93.9 percent reported by Puleo et al. for subforms of CK-MB in the diagnosis of myocardial infarction are impressive. As is usually the case, however, when the physician does not know whether a patient has the disease in question, the probability of disease before the test (the prevalence) must be incorporated into the data in order to arrive at the correct interpretation of the test results. For the population of patients in the article by Puleo et al., with a 10.9 percent prevalence of myocardial infarction, a positive CK-MB subform test yields a positive predictive value (i.e., the proportion of patients with positive tests who actually have the disease) of 66 percent. Therefore, 34 percent of the patients with positive test results would actually be identified erroneously as having infarctions if this test were performed.
Before physicians adopt new forms of technology to provide more accurate and cost-effective care, they need to be familiar with Bayesian theory, which will assist them in knowing how to interpret laboratory results correctly.1
1 ReferencesDouglas A. Propp, M.D.
Lutheran General Hospital, Park Ridge, IL 600681
Griner PF ,Mayewski RJ ,Mushlin AI ,Greenland P . Selection and interpretation of diagnostic tests and procedures: principles and applications. Ann Intern Med 1981;94:557-592
Medline
To the Editor:
Puleo and his colleagues conclude that patients with negative assays for CK-MB subforms can be adequately cared for in a “regular hospital unit,” thereby realizing a substantial cost reduction in the triage of patients with acute ischemic symptoms. This conclusion conflicts with the guidelines for unstable angina published by the Agency for Health Care Policy and Research, which recommend that patients with unstable angina and high-risk features, such as ongoing ischemic symptoms or hemodynamic instability, should be admitted to an intensive care unit.1 This recommendation is based on data demonstrating that patients with unstable angina, even without evidence of myocardial injury, have the highest risk of death at the time of admission to the hospital. In fact, data from the study by Puleo and colleagues appear to support this concept, demonstrating that 7 of 41 patients who died or had a serious complication had negative assays for CK-MB subforms. Reliance on this assay as the principal determinant of the need for admission to the intensive care unit is premature.
1 ReferencesGary E. Lane, M.D.
Mayo Clinic Jacksonville, Jacksonville, FL 32224To the Editor:
Hamm mentions twice in his editorial (Sept. 1 issue)1 the new “immunoinhibition” tests for CK-MB, which, as he says, measure the mass concentration of CK-MB rather than the enzyme activity. We would like to draw attention to the fact that the author is confusing two completely different techniques.
The immunoinhibition assays are based on the traditional method of measuring CK-MB activity. An anti–CK-M antibody that inhibits the activity of the CK-M subunit is added, and the remaining CK activity is measured. This remaining activity usually corresponds to the CK-MB activity, since CK-MM and CK-MiMi are not present in normal serum. These immunoinhibition assays are subject to interferences caused by the occasional presence of type 1 (CK-BB and IgG) or type 2 (CK-MiMi and proteins) macro-CK,2 and they have poor analytic sensitivity and specificity. To our knowledge, there are no new automated immunoinhibition assays based on monoclonal anti–CK-MB antibodies.
However, there are indeed newly developed assays for CK-MB, with improved sensitivity and specificity, that measure the mass concentration of CK-MB and that not only allow earlier diagnosis of acute myocardial infarction, as the author says, but may also be used as indicators of the risk of death from coronary causes in patients with unstable angina.3 These new assays are not based on an immunoinhibition principle, however, but rather are immunometric tests using an enzyme-linked immunosorbent assay, an enzyme fluorometric assay, or chemiluminescence techniques with monoclonal anti–CK-MB antibodies.4
4 ReferencesGeorges Gilson, Ph.D.
René L. Humbel, Ph.D.
Centre Hospitalier de Luxembourg, Luxembourg L-1210, Luxembourg1
Hamm CW . New serum markers for acute myocardial infarction. N Engl J Med 1994;331:607-608
Full Text | Web of Science | Medline2
Bayer PM ,Kraus S . Multiple occurrence of macro creatine kinase in one family. Clin Chem 1992;38:1379-1381
Web of Science | Medline3
Pettersson T ,Ohlsson O ,Tryding N . Increased CKMB (mass concentration) in patients without traditional evidence of acute myocardial infarction: a risk indicator of coronary death. Eur Heart J 1992;13:1387-1392
Web of Science | Medline4
Badminton MN ,Dawson CM ,Rainbow SJ ,Francis CM ,Tickner TR . Evaluation of a new rapid immunometric method for the measurement of the MB isoenzyme of creatine kinase in serum. Ann Clin Biochem 1992;29:563-564
Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We appreciate Dr. Keffer's interests but do not agree that total CK-MB assayed by the immunoinhibition (mass) assay detects infarction within two to six hours of the onset of symptoms. The glass-bead assay developed in 1975 was shown to have the same sensitivity as the first radioimmunoassay we developed in 1976.1 It served as the standard for subsequent immunoinhibition mass assays and was used in many national trials sponsored by the National Heart, Lung, and Blood Institute.2 Dr. Keffer is correct in saying that the assay is not used as widely as the most popular assay, electrophoresis, or mass assays. Attaining a diagnostic sensitivity of 90 percent requires 8 to 12 hours from the onset of symptoms with the glass-bead or mass assay and 10 to 12 hours with electrophoresis.
The studies quoted by Dr. Keffer support our results rather than his interpretation. Collinson et al.3 measured diagnostic sensitivity 4 and 8 hours after admission to the cardiac care unit, and after correcting for the time to the onset of symptoms, the authors found a sensitivity of 90 percent 8 and 12 hours after that time. Similarly, Marin and Teichman4 required four hours after arrival at the hospital to achieve 90 percent sensitivity, and, as they indicate, eight hours after the onset of symptoms. However, 22 percent of patients were lost to follow-up, which may have spuriously increased sensitivity to 90 percent. The study by Mair et al.5 appears promising, but it included only 36 patients with acute myocardial infarction, and the number discharged from the emergency room without follow-up is not indicated, an omission that may have spuriously affected the sensitivity of the assay. In the study by Bhayana et al.,6 the activity of CK-MB subforms was not quantified individually, and therefore their assay is not comparable to the quantitative assay of CK-MB subforms used in our study. Keffer's comment about speculation is unwarranted, since this point was clarified in the discussion section of our article.
In reply to Bhayana and Henderson, we have not consistently detected significant differences in the activity of CK-MB subforms that warrants separate criteria according to sex. With respect to cutoff criteria, as the receiver operating characteristics in Figure 1 show, no cutoff in the conventional assay for total CK-MB would have produced diagnostic accuracy four to six hours from the onset of pain.
We agree with Dr. Lane, and so stated in our article, that patients whose condition is unstable, as indicated by hemodynamic measurements or recurrent chest pain, should be admitted to the coronary care unit. Nevertheless, several studies have shown that patients with unstable angina can be taken care of in a regular hospital bed with electrocardiographic monitoring. None of the seven patients with serious complications (including two who died) among the patients without elevations of CK-MB subforms would have been treated less effectively had they been admitted to an intermediate unit rather than to intensive care. If the assay is used for the triage of patients other than those with ST-segment elevations, unstable hemodynamics, or high-risk unstable angina, it would still represent a savings of $2 billion to $3 billion. We agree fully with Dr. Lane that a prospective study of triage using CK-MB subforms is warranted before the widespread adoption of this technique.
6 ReferencesRobert Roberts, M.D.
Baylor College of Medicine, Houston, TX 77030Peter Puleo, M.D.
Christ Hospital, Cincinnati, OH 45219-29891
Roberts R ,Sobel BE ,Parker CW . Radioimmunoassay for creatine kinase isoenzymes. Science 1976;194:855-857
CrossRef | Web of Science | Medline2
Roberts R ,Croft C ,Gold HK , et al. Effect of propranolol on myocardial-infarct size in a randomized blinded multicenter trial. N Engl J Med 1984;311:218-225
Full Text | Web of Science | Medline3
Collinson PO ,Rosalki SB ,Kowana T , et al. Early diagnosis of acute myocardial infarction by CK-MB mass measurements. Ann Clin Biochem 1992;29:43-47
Web of Science | Medline4
Marin MM ,Teichman SL . Use of rapid serial sampling of creatine kinase MB for very early detection of myocardial infarction in patients with acute chest pain. Am Heart J 1992;123:354-361
CrossRef | Web of Science | Medline5
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 Cardiol 1991;68:1545-1550
CrossRef | Web of Science | Medline6
Bhayana V ,Cohoe S ,Leung FY ,Jablonsky G ,Henderson AR . Diagnostic evaluation of creatine kinase-2 mass and creatine kinase-3 and -2 isoform ratios in early diagnosis of acute myocardial infarction. Clin Chem 1993;39:488-495
Web of Science | Medline
Author/Editor Response
The clarifications of Gilson and Humbel are appreciated. The new immunoassays for CK-MB with improved diagnostic potential are correctly described as immunometric tests based on monoclonal anti–CK-MB antibodies. The immunoinhibition assay is the traditional technique.
Christian W. Hamm, M.D
University Hospital Hamburg, D-20246 Hamburg, Germany- Citing Articles (1)
Citing Articles
1
Sophie Poirey, Anne Polge, Jean-Pierre Bertinchant, Etiennette Bancel, Jean-Christophe Boyer, Pascale Fabbro-Peray, Bernard Magnan de Bornier, Bertrand Ledermann, Michel Bonnier, Jean-Pierre Bali. (2000) CK-MB mass test in ischemic myocardial injury. Comparison of two tests: Biomerieux Vidas and Sanofi Access immunoassays. Journal of Clinical Laboratory Analysis 14:2, 43-47
CrossRef
