Call to Redefine Troponin Threshold for Periprocedural MI

March 05, 2012

March 5, 2012 (Boston, Massachusetts) — Using the currently recommended troponin cutoff to define periprocedural MI is too sensitive and identifies almost a quarter of patients as having had such an MI, new data from almost 5000 individuals undergoing elective coronary stent placement illustrate [1].

Dr Victor Novack (Harvard Clinical Research Institute, Boston, MA) and colleagues used recommended definitions of periprocedural MI--namely, measure of the biomarkers troponin and creatinine kinase-MB (CK-MB) of more than three times the 99th percentile of the upper reference limit--and found that just 7% of patients would have had a procedural MI using the CK-MB criteria, compared with 24.3% using troponin. Their report is published online February 27, 2012 in the Archives of Internal Medicine.

"The major finding from our paper is when CK-MB and troponin are both used to diagnose periprocedural MI, they diagnose very different events if they are taken at the same threshold of three times the [upper limit of normal] ULN, which is the current [ESC/ACCF/AHA/WHF] task force definition, and the task force says use either one," senior author Dr Donald E Cutlip (Beth Israel Deaconess Medical Center, Boston, MA) told heartwire .

The major finding is when CK-MB and troponin are both used to diagnose periprocedural MI, they diagnose very different events if they are taken at the same threshold.

"We really weren't saying one was better than the other, we were just saying they are so different that the thresholds have to be modified," Cutlip notes. "It should either be recommended that we use only troponin and forget about CK-MB or that we have a different threshold for the two."

In an accompanying comment [2], Dr Joseph S Alpert (University of Arizona, Tucson) and Dr Allan S Jaffe (Mayo Clinic, Rochester, MN) point out that a multinational task force is currently revising the universal definition of MI and is "well aware of several of the issues pointed out by Novack et al." The new recommendations are expected to be published in the latter half of 2012, they note.

Troponin "overly sensitive" for periprocedural MI

In their paper, Novack et al examined the effect of a troponin-based definition for periprocedural MI compared with one based on CK-MB fraction for assessing event frequency and predicting one-year mortality in 4930 patients participating in the US Evaluation of Drug Eluting Stents and Ischemic Events (EVENT) registry.

MI occurred in 7.2% of patients by the CK-MB criteria and in 24.3% of patients by the troponin criteria. Both CK-MB (hazard ratio 1.38) and troponin (HR 1.35) as continuous values were associated with one-year mortality.

But the mortality effect of a more than threefold increase was greater for CK-MB (adjusted HR 2.5) than for troponin (adjusted HR 1.7), the authors note.

They say that there "has been concern that troponin may be overly sensitive for periprocedural MI," and their findings seem to bear this out. "The present results have implications for defining end points for clinical trials and for the clinical care of individual patients," they observe.

Novack and colleagues are not the first to highlight this issue; a British group made a similar call last year.

Troponin is best, but cutoff level should be > 20 times ULN

Cutlip told heartwire , "Troponin should be the preferred biomarker, because it's more specific and more sensitive, but I think we need to decide what type of events we are trying to identify after PCI." Using the cutoff of more than three times the ULN for troponin "at the very least is not very discriminatory," he observes.

And he notes that while most of the centers participating in this registry used the older troponin assays, use of the new higher-sensitivity assays commonly employed in Europe now would have "made the diagnosis even more sensitive and the discrepancy larger."

"We are looking for events that we think are going to be chronically important," Cutlip notes, adding that the idea is to home in on those who will be at greatest risk going forward. When his team used an alternative definition of troponin > 20 times the ULN to diagnose periprocedural MI, it provided a very similar frequency (7%) to that obtained with CK-MB more than three times the ULN, he notes.

They therefore suggest that this peak troponin value [ > 20 times ULN] be used as the new cutoff.

In their comment, Alpert and Jaffe say: "Considerable time and energy have been devoted to discussions involving . . . post-PCI MI" by the multinational task force devising the new guidance, and "in its current draft form, higher troponin values along with supplemental clinical information will be required to diagnose an MI after PCI."

Funding for the EVENT registry and analysis was provided by grants from Millennium Pharmaceuticals and Schering-Plough. Novack is a consultant for Harvard Clinical Research Institute. Cutlip has received research support paid to his institution from Medtronic. Disclosures for the coauthors are listed in the paper. The editorialists report no conflicts of interest.