Gender-Specific Decision Limits for Cardiac Troponin for Risk Stratification?

Cardiac troponin is the most important marker for the diagnosis of acute coronary syndromes (ACS) and risk stratification for major short-term (30 days) adverse events. Despite its widespread use in clinical laboratories, testing for cardiac troponin has only been available for the past 15 yr. As of the year 2000, the diagnosis of acute myocardial infarction (AMI) is predicated on the finding of an abnormal level of troponin in blood, in the context of ischemia.1 The troponins are more sensitive for detecting myocardial damage than the creatine kinase-MB isoenzyme and myoglobin due to its higher myocardial tissue content, and is more specific because the cardiac troponin isoform is not found in any other organs or tissues of the body besides the heart.2 As with any new technology, assays for troponin have undergone and continue to undergo their share of “growing pains.” Some initial confusion arose with the existence of two different troponin assays that are in routine clinical use: cardiac troponin T (cTnT) is available through a single manufacturer (due to restrictions in intellectual property rights), while there are dozens of assays available for cardiac troponin I (cTnI). Both cTnT and cTnI are part of a complex of three proteins (including troponin C) that regulates muscle contraction and are released from the damaged myocardium at the same time. cTnT is a slightly larger protein and stays in the circulation slightly longer than cTnI, which may explain why there is a higher incidence of increased cTnT than cTnI in patients with chronic renal failure. Otherwise, the two biomarkers are theoretically equivalent for detection of cardiovascular disease. As such, the clinical laboratories will use either cTnT versus cTnI depending on the availability of the clinical instrumentation available in the laboratory and not on the clinical or analytical performance of one marker over the other. A major source of confusion for the clinical use of troponin has been the evolution of cutoff concentrations. When originally designed as a marker of AMI, receiver-operating characteristic (ROC) curve analysis were used to derive relatively high cutoff concentrations to differentiate between unstable angina and AMI (cutoff “A,” Fig. 1). When the importance of detecting “minor myocardial damage” was evident in the context of risk stratification, troponin cutoff concentrations were decreased. This prompted a need to improve the precision of the troponin assay, as use of low cutoff concentrations and imprecise assays can produce falsely positive results due to “analytic noise.” With the release of next-generation commercial assays, the recommended cutoff concentration for cardiac troponin is now set at the 99th percentile of a healthy population (cutoff “B,” Fig. 1). Using this low limit, troponin is not just a marker of AMI but for any etiology that can cause cardiac damage, including heart failure, sepsis, myocarditis and renal failure.3 For diagnosis of AMI, marginal increases in troponin must be demonstrated within the clinical context of myocardial ischemia, as determined by clinical history, electrocardiographic changes, angiography and anatomic pathology (where available).Figure 1.: The assignment of cutoff concentrations for cardiac troponin for acute myocardial infarction (AMI) and coronary artery bypass graft (CABG) surgery. ROC, receiver operating characteristic curve analysis. MACE: major adverse cardiac events, i.e., cardiac death and AMI.Until now, there has been no attempt to identify gender-specific cutoff concentrations for cardiac troponin. There was a presumption that the release of troponin from the damaged heart occurs to the same extent for men and women. However, recent reports have shown that the pathophysiology of ACS are different between the sexes. Women have smaller coronary arteries, more diastolic dysfunction, more vague symptoms and higher morbidity and mortality after revascularization.4 This has prompted some to suggest that separate male and female cutoff concentrations be established for cardiac troponin to optimally detect myocardial injury. In this issue of Anesthesia & Analgesia, Stearns et al.5 examined the prognostic value of cTnI levels for predicting adverse cardiovascular outcomes in 175 postmenopausal women undergoing coronary artery bypass graft (CABG) surgery who were pretreated with estrogens or placebo. Using ROC curve analysis, they determined an optimum cTnI cutoff of 7.6 ng/mL for predicting risk for adverse cardiac events (cutoff “C,” Fig. 1). As this cutoff is lower than the 9–15 ng/mL reported in previous risk stratification studies conducted largely in males using similar testing technologies (cutoff “D,” Fig. 1), they suggest that separate troponin decision limits be considered. Should this concept be given serious consideration? For AMI diagnosis, the cutoff concentration is established from healthy subjects not from ROC curve analysis. Using highly sensitive troponin assays, Apple et al.6,7 were not able to consistently find gender differences among commercial troponin assay results in serum or plasma. For the few assays in which there was a difference, the minor differences in the cutoff concentration have minimal clinical significance as after AMI, troponin can increase some 100–1000 fold from the baseline, at a fairly high rate of change. Moreover, there is no pathophysiologic basis for gender-specific differences. In a small study of 34 patients who died of various noncardiac causes, Swaanenburg et al.8 showed no significant gender differences in the tissue content of troponin T, myoglobin, and α-hydroxybutyric acid dehydrogenase. Cutoffs for risk stratification for patients who present with chest pain and are excluded for ACS are also set at the 99th percentile of healthy subjects. There are data suggesting that use of age-specific troponin cutoff concentrations may provide additional risk stratification value,9 but such limits have not yet reached routine clinical practice. It may be possible that gender-specific cutoffs have prognostic value for the specific case of death and AMI after cardiac surgery. This situation is different than the myocardial damage that occurs in patients with naturally occurring disease. After CABG, there is some baseline release of cardiac troponin due to the heart injury caused by the surgical procedure itself. Any additional release into the circulation may be due to a perioperative AMI, or a higher degree of injury that is indicative of future adverse events. For example, in the Stearns et al. report,5 the troponin cutoff of 7.6 ng/mL is 100-fold higher than the 99th percentile value of 0.07 ng/mL reported for the Dade (now Siemens Medical Solutions) cTnI assay,10 which was used in their study. Therefore in this situation, the cutoff troponin cutoff concentrations must be determined by ROC curve analysis and not on the 99th percentile of a healthy population (cutoff “B,” Fig. 1). While these data are interesting, they must be considered only a concept. As acknowledged by the authors, there are many surgical confounding variables that make it difficult to compare cutoff concentrations from this study conducted in women versus others conducted primarily in men. Others have opined that CABG is too complex to suggest simple cutoff values, let alone gender-specific limits.10 There are also confounding analytical issues regarding the use of different troponin assays. Although there is a troponin reference standard that is now available through the National Institute of Standards and Technology, not all manufacturers have adopted this standard and there is no standardization or complete harmonization of results across troponin testing platforms.11 Even within a manufacturer, there are differences between instruments and between generations of assays.12 Therefore gender-specific troponin cutoffs must be established with each surgical practice and troponin assay used, which is not realistically possible. Nevertheless, the Stearns et al. work suggests that there may be differences in postsurgical risks for women versus men with regards to the extent of myocardial damage. Therefore, I applaud the efforts of these physicians in their attempt to find cardioprotective drugs specifically designed for women. Although 17β-estradiol did not demonstrate benefits, this should not discourage trials of other experimental drugs, particularly if there is a link to the pathophysiology of the disease for women.