Point: The Use of Sex-Specific Cutpoints for High-Sensitivity Cardiac Troponin Assays

Sex-specific reference intervals for cardiac troponin exist, with the 99th percentile (the cutpoint for determination of an increased troponin) being substantially lower in females than males (1, 2). Although recent guidelines have recommended their use (3, 4), clinical utilization of sex-specific ranges is limited. Troponin is engrained in our assessment of risk, diagnosis of disease, treatment options, and determination of prognosis (5). Irrespective of sex, the risk of acute myocardial infarction (AMI),3 cardiovascular death, recurrent MI, refractory ischemia, and rehospitalization increases with increasing troponin concentrations. The finding of an increased troponin often leads to targeted medical interventions designed to improve outcomes. The definition of an increased troponin is critical; if the cutpoint is set too high, there is the potential for missed diagnoses, leading to fewer receiving guideline-recommended management, and poorer outcomes. Currently women with acute coronary syndromes (ACS) receive less guideline-recommended treatment and have worse outcomes than men (6). Many assume that men have higher rates of MI compared with women. This is not supported by WHO statistics that report the proportion of men and women dying from coronary heart disease (CHD) in Europe are identical at 20% and 22% respectively (7). It is postulated that in the acute setting fewer women are evaluated for CHD owing to atypical symptoms and fewer electrocardiogram abnormalities than men (8). However, given troponin is integral to the diagnosis of AMI (4), it may also be that sex-specific differences in troponin cutoffs have contributed to differences in diagnostic rates. Women are known to have lower cardiac troponin values than men for any given extent of coronary artery disease (9) and this may result in fewer women being identified as high risk. Indeed, Shah et al. (10) investigated 1126 consecutive patients with suspected ACS. Clinical care was guided by a sensitive assay with a single diagnostic threshold (50 ng/L), but troponin concentrations were determined in parallel using a high-sensitivity (hs) assay. Cardiologists adjudicated the diagnosis of MI separately using the clinical assay and hs-troponin I assay with sex-specific cutpoints. Use of sex-specific thresholds doubled the rate of diagnosis in females (11% to 22%), but had minimal effect on males (19% to 21%). Although overall more men were diagnosed with MI, the proportion of both sexes with MI was comparable when sex-specific thresholds were applied. Women with small increases in troponin concentration, only detectable using the sex-specific threshold, had rates of death or reinfarction that were comparable to or worse than women with much larger infarcts identified using the clinical assay. However, the use of a higher clinical threshold may have magnified the impact of using sex-specific values. The inclusion of consecutive patients in this study avoids selection bias and indicates that a similar proportion of women and men presenting with suspected ACS have MI. These observations raise the possibility that use of sex-specific cutpoints in clinical practice could improve survival in women with MI. A study by Slagman et al. (8) found that the diagnosis of Non ST-segment elevation MI (NSTEMI) in an emergency cohort was substantially lower in women compared with men using a single cutpoint (56.1% vs 70.1%) but the difference reduced with the use of lower cutpoints (63.1% vs 74.8%). The sensitivity and negative predictive value for the diagnosis of AMI in women improved using the lower values. The added benefit in the prediction of inhospital mortality for women presenting via the emergency department has also been described (11). The first report about the true clinical impact of the clinical use of sex-specific ranges in patient care has only recently been published (12). In accordance with guideline-recommendations (3–5), several laboratories have implemented sex-specific ranges for cardiac troponin I (cTnI). Australia and New Zealand have been early adopters and findings from the initial 6 months use from 2 institutions are informative. The cutoff used was 16 ng/L for females at both sites, and 26 ng/L and 34 ng/L for males at the Australian and New Zealand sites respectively. No difference in the proportions of men with increased troponin values (±1.3%; P ≥ 0.20) was reported, however more women were found to have increased values (+5.2–8.4%; P < 0.001). There was a significant increase in the proportion of women admitted to cardiology units (23.2% to 30.8%; P < 0.001) with an increase in cardiovascular diagnoses including heart failure and arrhythmias. There was not an increase in the diagnosis of MI based on administrative databases, and for the New Zealand subset, no increase in angiography rates. Despite the implementation of sex-specific thresholds, women remained 3 times less likely than men to undergo angiography (27.4% vs 72.6%). This may reflect an educational gap with clinicians uncertain of the significance of small increases in troponin concentration. In keeping with others findings, the impact on men was minimal however the optimal male cutoff remains questionable (10, 13). In practical terms, in preparation for the introduction of the sex- specific cutpoints at the New Zealand site a collaborative approach between pathologists, emergency physicians, and cardiologists developed a broad educational program with online support via the laboratory reporting system for interpretation of results. In a clinician survey following the change, only 1% of clinicians “slightly agreed” that sex specific cutoffs had been difficult to use and 0% “strongly agreed.” There were minimal concerns expressed by clinicians relating to any changes in admission practices and access to inpatient beds. This study provides important insights about the clinical impact of sex-specific ranges (12). However, owing to absence of long-term outcome data, it remains unclear if changes in medical therapy occurred and if improved risk assessment had prognostic benefits for women. Although some women may not benefit from urgent angiography, it is clear the introduction of sex-specific ranges into clinical practice highlighted women at increased risk of cardiovascular events and death, leading to changes in disposition and specialist management. The challenge now lies in our ability to modify this risk with improvements in therapeutic interventions. Understanding the true value of sex-specific cutpoints is considerably hampered by much of the evidence available, owing to methodological limitations. All studies have limitations; however, a good understanding of these is needed to allow correct interpretation of studies reporting the utility of sex-specific ranges. Studies conducted in cohorts comprised of patients with diagnosed non-ST segment elevation ACS, or patients admitted to coronary care units do not allow a fair assessment of sex-specific cutpoints (13, 14). In emergency patients, single or even serial testing of troponin values may not be included leading to many women being excluded as their lower troponin values made them ineligible. In some observational studies, in which the gold standard diagnosis of AMI was determined using older assays and overall cutpoints, stored samples have been retested with hs-assays (13). Such studies incorporate data on patients managed according to clinical information that was gathered at the time, influenced by older assays using the overall range. Women without clinically apparent increases in troponin were unlikely to be managed as high risk and receive further imaging for ischemia, a key criterion to support a readjudicated diagnosis of ACS. Women who presented with initial troponin concentrations below the overall 99th percentile cutpoint but higher than the sex-specific value were not necessarily recognized as high-risk for serious cardiac events (10, 11). Such studies lack the ability to reveal the benefit of sex-specific cutpoints. A subanalysis from the TRAPID-AMI (High-Sensitivity Cardiac Troponin T Assay for Rapid Rule Out of Acute Myocardial Infarction) study attempted to overcome this limitation by readjudicating clinical outcomes based on results of testing of stored samples with hs assays (15). The use of sex-specific cutpoints for hs-cardiac troponin T (cTnT) resulted in biomarker reclassification of 2.7% of the entire cohort (n = 1282). The rate of meeting the AMI-biomarker criteria doubled for females (4.5% to 9.8%) when applying sex-specific cutpoints, and decreased slightly for males (9.4% to 7.8%). The doubling effect on the event rate for women (similar to that reported by Shah (9) was deemed only ‘modest’, however lower rates of diagnosis may also reflect study selection bias (38% females). The limitation described above is applicable also to this study. Separate reference limits for women and men may be inconvenient (13, 16), however different reference limits are currently in use based on local troponin assay(s) available. Sex-specific intervals are used for other biomarkers (e.g., hemoglobin), and in the era of automated pathology reporting, most clinicians remain unaware about gender issues. Additionally, we are hampered by a lack of standard to define the 99th percentile, affecting values reported for both the overall and sex-specific ranges of assays (2). When using serial troponin testing, there is only a small change in the number of women reclassified with NSTEMI using sex-specific cutpoints, raising the possibility that differing cutpoints are unnecessary (13). This argument ignores the fact that management of emergency patients presenting with symptoms of possible ACS is often dependant on the interpretation of the first troponin result. Despite recommendations for serial testing for all patients with a possible AMI, many patients have troponin tested only once, reinforcing the importance of use of correct reference limits, particularly in the emergency setting. The advantages of sex-specific cutpoints are suggested to be unique to the Abbott hs-cTnI assay, the only manufacturer currently recommending their use (16), despite differences in troponin ranges existing for both hs-cTnT and hs-cTnI (1). A report from 3 large studies [the Dallas Heart Study, ARIC (Atherosclerosis Rick in Communities), and the Cardiovascular Health Study] suggests that the use of the overall hs-cTnT cutpoint of 14 ng/L may lead to overdiagnosis of AMI in men over 50 years of age, and the cutpoint should be lower in women (17). More data are needed to clarify this issue. Women with ACS are less frequently diagnosed with AMI, more likely to be managed conservatively, and often have worse outcomes than men. Although the reasons are multifactorial, myocardial injury in women is currently underrecognized. The use of sex-specific reference limits improves our ability to detect women with coronary disease and provides better prognostic information, while having little impact on men. The current use of a single cutpoint may contribute to the underdiagnosis of MI in women, and negate our ability to optimize treatment to reduce harm. acute myocardial infarction acute coronary syndrome coronary heart disease high-sensitivity non–ST-segment myocardial infarction cardiac troponin 1 cardiac troponin T.