A team of cardiologists from the United Kingdom, Australia, and Italy found that the cTnI threshold for MI type 4a is oversensitive and fails to accurately discriminate between periprocedural myocardial necrosis (PMN) and MI type 4a. They found that CK-MB more readily differentiates necrosis and infarction than cTnI.
“When applying the current universal definition of MI to periprocedural injury, CK-MB should currently be the preferred biomarker,” Dr. Adrian P. Banning, consultant cardiologist at the Oxford Heart Centre, John Radcliffe Hospital, Oxford, UK, and colleagues conclude in the February 8 issue of the Journal of the American College of Cardiology, available online now.
The researchers assessed the differential implications for CK-MB and cTnI measurement with the universal definition of periprocedural injury after PCI in 32 patients enrolled in the prospective study known as MICASA (Myocardial Injury following Coronary Artery Surgery vs Angioplasty) trial.
All of the patients underwent multivessel PCI and late gadolinium enhanced cardiac magnetic resonance imaging (LGE-CMR) and had cTnI, CK-MB and inflammatory markers measured at baseline and 1, 6, 12 and 24 hours after PCI.
Patients were stratified into 3 periprocedural injury groups using the universal biomarker definition: no injury (normal CK-MB, no cTnI elevation); PMN (CK-MB elevation between 1 and 3 times 99th percentile URL, the same for cTnI); MI type 4a (CK-MB > 3 times 99th percentile URL, the same for cTnI).
Using a highly-sensitive cTnI assay, 3 of 32 patients (9%) had no evidence of injury, 3 (9%) had PMN, and 26 (82%) met the definition of MI type 4a, having a cTnI elevation > 3 times the URL. Yet only 5 of the 26 patients had evidence of myocardial necrosis on LGE-CMR, yielding a positive predictive value of only 19% and a poor sensitivity of 22%.
Using CK-MB radically altered this distribution and better predicted LGE-CMR-defined infarction. With CK-MB, 17 of 32 (53%) had no evidence of injury, 10 (32%) had PMN and only 5 (15%) were classified as MI type 4a. Moreover, with CK-MB there was a “high specificity of 93%” for LGE-CMR, and 3 of 5 patients meeting MI type 4a had new LGE-CMR evidence of injury, giving a positive predictive value of 60%.
Changes in inflammatory markers were closely correlated to levels of CK-MB but not cTnI.
“These data clarify the significance of the abnormalities you might expect to find in routine PCI practice,” Dr. Banning commented in an email to Reuters Health. The current troponin threshold of the universal definition of procedural MI is “over sensitive,” the investigator added.
In further analysis, the researchers found evidence that redefining the cTnI thresholds to 12 times the 99th percentile URL for PMN and 40 times the 99th percentile ULR for MI type 4a “allows a comparable categorization” to those identified by post-PCI CK-MB.
Although this “seems reasonable,” it will require validation in a larger cohort of patients with long-term outcome data, Dr. Cindy L. Grines and Simon Dixon, of the department of cardiovascular medicine, William Beaumont Hospital in Royal Oak Michigan, write in a linked commentary.
“At this point in time, we discourage use of high-sensitivity troponin assays for determination of PCI-related MI. Rather, we recommend continued use of CK-MB measurements after PCI, on the basis of well-accepted standards and a very large body of clinical trial data,” they conclude.
J Am Coll Cardiol 2011;57:653-661.