Concerns about Using Sex-Specific Cutoffs for Cardiac Troponin Testing

Cardiac troponins are clinically sensitive and specific biomarkers of myocardial injury, and high-sensitivity cardiac troponin (hs-cTn) assays have changed the workup of patients with chest pain suspected to be acute coronary syndrome (ACS) in the emergency department. There is a newfound ability to rule out suspected ongoing ischemic heart disease without obvious electrocardiographic signs of acute myocardial infarction (AMI). The 99th percentile of hs-cTn is the concentration endorsed by guidelines to support the diagnosis of acute myocardial injury and AMI. The Fourth Universal Definition of Myocardial Infarction already recommends the use of sex-specific cutoffs for hs-cTn, given that there are circulating concentrations among women compared with men (1). Meanwhile, the most recent European Society of Cardiology guidelines for the management of ACS suggest that age, renal dysfunction, time from chest pain onset, and sex all affect cTn concentrations and recommend the use of a uniform cutoff until these factors can all be included in a model (2). The impact of sex-specific cutoffs for the clinical diagnosis of AMI is still being investigated and discussed; a wrong diagnostic strategy can be harmful and lead to underdiagnosis and poor outcomes or overdiagnosis and unnecessary treatments, hospital admissions, and increased costs. Given these conflicting recommendations, should clinicians use sex-specific cutpoints? There is variability in determining a reference population in which the 99th percentile upper reference limit (URL) is based on Sandoval and Apple (3). The AACC and IFCC provide recommendations on the definition of healthy reference participants (4). Some reference studies, however, utilized only a health questionnaire screening form, and others used vital signs, spirometry, electrocardiography, or other cardiac biomarkers. Consequently, there are differences in the URL in each cohort. On the whole, in a healthy population, the values of the 99th percentile URL for hs-cTn are between 1.2 and 2.4 times higher in men than in women. This difference is hypothesized to result from differences in left ventricular mass (5). However, this is just one of the factors that affect differences in baseline cTn concentrations (6). Age and renal function affect cTn even more than sex and are not currently accounted for in the definition of the 99th percentile URL. In addition, the absolute difference between a general 99th percentile URL and sex-specific URLs is determined by the difference between the values. The strongest impact of sex-specific URLs is observed with larger differences between URL values. Giannitsis (7) showed that for the hs-cTnT assay, the 99th percentile URL determined in numerous reference populations is narrow (14 ng/L for women vs 19 ng/L single cutoff), but larger differences exist using hs-cTnI assays (cutoff 16 ng/L for women vs 26 ng/L single cutoff). Therefore, reclassification would be more significant with larger ranges and would not necessarily be a product of difference in diagnosis but rather a difference in statistics. There is mixed evidence as to whether sex-specific URLs are beneficial. In the clinical setting, one must consider making the diagnosis and then the actions taken by clinicians once the diagnosis has been made. For the diagnosis of AMI, by using sex-specific URLs, the tradeoff is loss of clinical specificity for female patients and potentially loss of clinical sensitivity for male patients. A subanalysis of the TRAPID-AMI (Roche cTnT assay) study using sex-specific URLs showed that the AMI rate increased from 16.6% to 22.6% (entire cohort, n = 477) and from 62.6% to 71.7% (ACS subcohort, n = 99) in women, whereas in men, rates decreased from 23.1% to 21.1% (entire cohort, n = 805) and from 48.8% to 45.9% (ACS, n = 281), respectively (8). Trambas et al. (9) evaluated the change from contemporary cTnI to hs-cTnI (Abbott cTnI assay) using sex-specific URLs at 2 sites (n = 23,576), and although an increased number of female patients had increased cTnI concentrations, there was no increase in AMI diagnosis in women or men. Shah et al. (10) reclassified 1126 patients using sex-specific URLs (Abbott hs-cTnI assay), and this approach increased the number of women diagnosed as having type 1 myocardial infarction (MI) from 111 from 80. In contrast, in men, the diagnosis of type 1 MI using the sex-specific URLs for the high-sensitivity assay decreased from 142 (23%) to 131 (21%) with the sex-specific URLs (men, 34 ng/L). The diagnosis of MI, however, is only one part of the clinical algorithm. Although use of sex-specific URLs identifies more men and women with myocardial injury, women in these studies were still less likely than men with a similar diagnosis to undergo coronary angiography, revascularization, and prescribed guideline‐directed medical therapy, despite having a similar diagnosis (10, 11). These findings highlight the possibility that clinicians did not consider modest cTn increases, identified by the lower sex‐specific URL in women, significant enough to warrant procedures or a change in clinical decisions. A diagnostic test alone cannot lead to improved outcomes if it is not accompanied by a change in provider behavior. An ongoing trial, CODE-MI, will evaluate the impact of sex-specific URLs on the diagnosis, treatment, and outcomes of women with chest pain across Canada (12). Last, the potential value in incorporating sex‐specific URLs into clinical practice may be partially counterbalanced by the simplicity of a single URL. Eggers et al. (13) evaluated whether using a sex-specific cutoff would improve risk prediction in a Swedish population and found that there was no difference. Instead, there was increased risk of death and MI starting at hs-TnT concentrations around 10–12 ng/L in both men and women. The cutoff values to allow for a single hs-cTn rule-out value are all on average <5 ng/L (2), but all other values warrant a repeated assay at 1, 2, or 3 hours and evaluation of the absolute Δ measurement. No sex-specific Δ values are provided. In addition, the benefit of a lower sex-specific cutpoint is attenuated if serial measurements are used. Peacock et al. (14) studied 2 patient cohorts totaling >1400 patients and found that the diagnostic performance of the hs-TnT assay was not substantially affected when sex-specific URLs were applied to serial results: clinical sensitivity was not altered, and clinical specificity was slightly lower in women but higher in men. In the future, incorporation of decision rules that include age, sex, and renal function into cTn values in the electronic health record may help clinicians utilize different cutoffs based on patient characteristics. Although there are established sex-based differences in cTn measurements, it is unclear whether using sex-specific URLs has a meaningful impact on outcomes. Diagnosis is only one piece of the patient care pathway. With the use of accelerated diagnostic protocols that incorporate lower cutoffs and Δ cTn values, a sex-specific URL may not be as useful in the evaluation of patients with chest pain. hs-cTn, high-sensitivity cardiac troponin; ACS, acute coronary syndrome; AMI, acute myocardial infarction; URL, upper reference limit; MI, myocardial infarction. All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest: None declared. None declared. None declared. A.M. Chang, Roche. A.M. Chang, Roche, Abbott, Ortho Diagnostics, Siemens, Beckman Coulter. None declared. None declared.