Sex-Specific Thresholds of High-Sensitivity Troponin in Patients With Suspected Acute Coronary Syndrome

Background Major disparities between women and men in the diagnosis, management, and outcomes of acute coronary syndrome are well recognized. Objectives The aim of this study was to evaluate the impact of implementing a high-sensitivity cardiac troponin I assay with sex-specific diagnostic thresholds for myocardial infarction in women and men with suspected acute coronary syndrome. Methods Consecutive patients with suspected acute coronary syndrome were enrolled in a stepped-wedge, cluster-randomized controlled trial across 10 hospitals. Myocardial injury was defined as high-sensitivity cardiac troponin I concentration >99th centile of 16 ng/l in women and 34 ng/l in men. The primary outcome was recurrent myocardial infarction or cardiovascular death at 1 year. Results A total of 48,282 patients (47% women) were included. Use of the high-sensitivity cardiac troponin I assay with sex-specific thresholds increased myocardial injury in women by 42% and in men by 6%. Following implementation, women with myocardial injury remained less likely than men to undergo coronary revascularization (15% vs. 34%) and to receive dual antiplatelet (26% vs. 43%), statin (16% vs. 26%), or other preventive therapies (p < 0.001 for all). The primary outcome occurred in 18% (369 of 2,072) and 17% (488 of 2,919) of women with myocardial injury before and after implementation, respectively (adjusted hazard ratio: 1.11; 95% confidence interval: 0.92 to 1.33), compared with 18% (370 of 2,044) and 15% (513 of 3,325) of men (adjusted hazard ratio: 0.85; 95% confidence interval: 0.71 to 1.01). Conclusions Use of sex-specific thresholds identified 5 times more additional women than men with myocardial injury. Despite this increase, women received approximately one-half the number of treatments for coronary artery disease as men, and outcomes were not improved. (High-Sensitivity Troponin in the Evaluation of Patients With Acute Coronary Syndrome [High-STEACS]; NCT01852123).

I mportant disparities exist in the diagnosis, management, and outcomes of acute coronary syndrome between women and men (1)(2)(3). Women with suspected acute coronary syndrome are less likely to undergo evidence-based investigations and treatment, and outcomes following myocardial infarction are consistently poorer compared with men (1).
The development of high-sensitivity cardiac troponin assays has resulted in the identification of important sex differences in the reference range of cardiac troponin, with the 99th centile in men being twice that in women (4)(5)(6). We have previously demonstrated that the use of a high-sensitivity assay with sex-specific thresholds may double the diagnosis of myocardial infarction in women and identify women at risk for future cardiac events (7). This raises the question as to whether the use of single diagnostic thresholds has contributed to inequalities in the diagnosis, management, and outcomes of women with acute coronary syndrome.
The recently published fourth universal definition of myocardial infarction recommends the use of sexspecific thresholds for the diagnosis of myocardial infarction (2).

High-STEACS (High-Sensitivity
Troponin in the Evaluation of Patients With Suspected Acute Coronary Syndrome) was the first randomized controlled trial to evaluate the introduction of a high-sensitivity cardiac troponin I (hs-cTnI) assay with sex-specific thresholds into clinical practice (8).
The hs-cTnI assay reclassified 1 in 6 patients with myocardial injury, but this was not associated with a reduction in recurrent myocardial infarction or cardiovascular death at 1 year. In this pre-specified secondary analysis, we evaluated the impact of implementing sex-specific diagnostic thresholds on the use of investigation and treatments for coronary heart disease and on clinical outcomes in women and men separately.

METHODS
STUDY DESIGN. High-STEACS was a steppedwedge, cluster-randomized controlled trial that evaluated the implementation of an hs-cTnI assay in consecutive patients presenting with suspected acute coronary syndrome across 10 secondary and tertiary hospitals in Scotland (NCT01852123) (Online Table 1). The study design has been described in detail previously (8), and the study was conducted with the approval of  Sex-Specific Thresholds of hs-cTnI international guidelines in use during enrollment (9,10 OUTCOMES. We used regional and national registries to ensure complete follow-up of the trial population (7,13,14). The primary outcome was a composite of type 1 or type 4b myocardial infarction following the initial presentation to hospital or cardiovascular death within 1 year. The primary outcome was independently adjudicated by 2 physicians blinded to study phase, and any disagreements were resolved by a third physician.
STATISTICAL ANALYSIS. All patients with peak hs-cTnI above the sex-specific 99th centile were classi-  Table 2).     (21) Values are mean AE SD, n (%), or median (interquartile range). *Presenting symptom was missing in 5,615 patients (12%). †Two medications from aspirin, clopidogrel, prasugrel, or ticagrelor. ‡Includes warfarin and direct oral anticoagulant agents. §Electrocardiographic and physiological data were available in 1,377 of reclassified patients (78%) and 6,470 of identified patients (75%). kDefined as 2 or more tests within 24 h of presentation. ¶The adjudication panel was able to achieve consensus diagnoses in 9,098 of 10,360 patients (88%) with hs-cTnI concentrations above the sex-specific 99th centile.

MANAGEMENT OF WOMEN AND MEN DURING INDEX
ADMISSION. Women with myocardial injury presenting in the implementation phase were more likely than those presenting during the validation phase to undergo coronary angiography (18% vs. 26%) and coronary revascularization (10% vs. 15%) (p < 0.001

Non-ischemic myocardial injury
Adjudicated diagnoses are presented for patients with troponin concentration above the contemporary cardiac troponin I assay threshold of 50 ng/l and those with troponin concentration above the sex-specific 99th centile threshold of 16 ng/l in women and 34 ng/l in men. *Where there was consensus among the adjudication panel that there was insufficient clinical information to make a definitive diagnosis, because of missing admission or discharge letters, we did not attempt to adjudicate the diagnosis (1,245 of 10,360 [12%]). As we had access to all other information, including medical history, clinical investigations, management, and outcomes, these patients were not excluded from our primary or secondary analyses.
Lee et al. for both) ( Table 2, Online Tables 3 and 4). Similarly, in men with myocardial injury, coronary angiography and revascularization increased from the validation to implementation phase (38% vs. 46% and 26% vs. 34%, respectively; p < 0.001 for both). However, across both phases, rates of coronary angiography and revascularization were lower in women compared with men (p < 0.001 for both). Likewise, prescriptions for preventive therapies increased following implementation in both women and men but across both phases remained lower in women ( Table 2). These differences in management between women and men were also observed in patients older and younger than the median age of 73 years (Online Figure 2). Across both phases, women with type 1 myocardial infarction were less likely than men to undergo coronary angiography (43% vs. 66% during the validation phase and 53% vs. 73% during the implementation phase; p < 0.001 for both) and coronary revascularization (26% vs. 48% and 35% vs. 57%, respectively; p < 0.001 for both) (Online Table 5, Online Figure 3). They were also less likely than men to receive prescriptions for secondary prevention such as dual-antiplatelet therapy (48% vs. 61% during the validation phase and 54% vs. 67% during the implementation phase; p < 0.001 for both), statins (24% vs. 37% and 31% vs. 41%, respectively; p < 0.001 for both), and beta-blockers (26% vs. 42% and 33% vs.   Table 11). Primary and secondary outcomes in those women and men with Values are median (interquartile range) or n (%). *Chi-square and Mann-Whitney U-tests comparing the validation and implementation phases. †p < 0.05 comparing women and men.

DISCUSSION
The implementation of sex-specific thresholds for    Paired log-rank test results are p ¼ 0.01 for men with myocardial injury and p ¼ 0.06 for men without myocardial injury. (Right) Cumulative incidence time-to-event curves for the primary outcome of myocardial infarction or cardiovascular death at 1 year for women admitted during the validation phase (dashed line) and implementation phase (solid line). Patients are grouped according to whether myocardial injury was present (red) or absent (gray). Paired log-rank test results are p ¼ 0.40 for women with myocardial injury and p ¼ 0.08 for women without myocardial injury. Sex-Specific Thresholds of hs-cTnI infarction in women and men (15). The impact of sex-specific thresholds on the diagnosis of myocardial infarction has been evaluated in a number of observational studies with divergent findings (7,(16)(17)(18)(19)(20).
Most of these studies enrolled selected patients with acute coronary syndrome, of whom the majority were men. Furthermore, sex-specific thresholds were not used to guide clinical care or subsequent investigation for coronary artery disease. Here, we imple-

CENTRAL ILLUSTRATION Implementation of High-Sensitivity Troponin and Sex-Specific Thresholds
Sex-specific thresholds identified 5-times more additional women than men with myocardial injury Despite this, women received fewer treatments for coronary artery disease than men And their outcomes were not improved. Sex-specific thresholds identified 5 times more additional women than men with myocardial injury. Despite this, women received fewer treatments for coronary artery disease than men, and their outcomes were not improved. CI ¼ confidence interval; MI ¼ myocardial infarction.
Lee et al. were women, but we did not observe any improvement in their outcomes. There was also no difference in the primary outcome in men, but as very few men were reclassified, we were underpowered for this subgroup.
A number of observations could explain our find-  Sex-Specific Thresholds of hs-cTnI (29,30). The impact of implementing high-sensitivity cardiac troponin testing is likely to differ when used in a less selected patient population and when introduced at hospitals that have used either higher or lower cardiac troponin thresholds than used here.
Finally, we did not directly compare whether the use of sex-specific thresholds was superior to a uniform threshold using a hs-cTnI assay. This question is being evaluated in a prospective cluster randomized controlled trial (CODE-MI [Hs-cTn-Optimizing the Diagnosis of Acute Myocardial Infarction/Injury in Women] trial; NCT03819894) that will include a health economic analysis to determine whether adopting sex-specific thresholds and identifying more women at risk is cost effective.

CONCLUSIONS
We report that use of a hs-cTnI assay with sexspecific thresholds identified 5 times as many additional women with myocardial injury than men, such KEY WORDS acute coronary syndrome, high-sensitivity cardiac troponin, myocardial infarction, sex-specific threshold APPENDIX For supplemental methods, tables, and figures, and a list of High-STEACS investigators, please see the online version of this paper.