Cardiac Troponin I in Non-ST-Elevation ACS-- Importance of Gender
Cardiac Troponin I in Non-ST-Elevation ACS-- Importance of Gender
The development of high-sensitivity cardiac troponin assays has taught us 2 major lessons. One is the potential of these assays for the diagnosis of MI within the first hours after admission. The second lesson is their impact on the predictive power of cardiac troponin because concentrations even below the 99th percentile may be predictive of adverse outcomes. This applies both to populations with coronary artery disease, congestive heart failure, or atrial fibrillation but also to community-based cohorts. In the present report, we are taught a third lesson, that of the gender-related differences in NSTE-ACS.
We noted that men with NSTE-ACS had considerably higher cTnI levels compared to women. This is supported by observations from other studies. It has been suggested that the differences in cTnI levels between the genders might mirror differences in the cardiovascular risk pattern between before the occurrence of NSTE-ACS. In fact, women in GUSTO IV were older and exhibited higher prevalence of diabetes, congestive heart failure, and hypertension. Men, in contrast, more often were smokers and had previous manifestations of coronary artery disease. However, in a recent analysis from the PEACE study, only a minor difference in cTnI levels was noted between men and women with stable coronary artery disease (ARCHITECT STAT hsTnI; men 4.6 [3.1–7.4] ng/L, women 4.0 [2.6–6.4] ng/L) (Omland T; personal communication). It, therefore, seems unlikely that the marked difference in cTnI levels between men and women from GUSTO IV might be explained by differences in the cardiovascular risk pattern.
This leads to the question whether our results instead might reflect differences in the pathophysiology of NSTE-ACS. Women usually have a lower prevalence of obstructive coronary artery disease and more often have microvascular and endothelial dysfunction and/or diffuse atherosclerosis. Men, in contrast, more often have complete atherosclerotic plaque ruptures resulting in repetitive distal embolization of thrombotic material. Although not provable within the context of GUSTO IV, we assume that this might contribute to greater myocardial damage in men with higher troponin elevations as seen in women.
However, despite overall lower concentrations, cTnI at the tested 99th percentiles tended to exhibit stronger associations with adverse outcome in women. This was similar for cTnT levels measured with a conventional assay and is on first sight a paradoxical and provocative finding. However, the interaction terms of gender on the association of cTnI with outcome were nonsignificant, and in case of elevated levels, the event rates in men and women were similar. Thus, the difference between men and women was the difference in cTnI levels among those with a favorable outcome and not any differences among those who actually had an event.
Accordingly, the clinical presentation of women with symptoms of unstable coronary artery disease appears to encompass a broader cardiovascular risk panorama compared to men, and this seems to be reflected in the distribution and the prognostic implications of cTnI levels. In this context, it is noteworthy that more women than men were included in GUSTO IV based on ST-segment depression. ST-segment depression in the absence of troponin elevation could be a result of ischemia induced by raised heart work, coronary spasm, or nonischemic causes and thereby mirror less adverse pathophysiologies or no pathology at all. This is illustrated by the low event rate in women without cTnI elevation but with ST-segment depression, which contrasted to men in whom ST-segment depression was strongly prognostic. We, therefore, cannot exclude the possibility that some of the women included in GUSTO IV might have had another condition than NSTE-ACS despite unstable symptoms. This emphasizes the need of additional indicators of coronary instability in troponin-negative women for tailoring of appropriate therapies, especially with respect to the avoidance of potentially harmful procedures. This is illustrated by data from the TACTICS-TIMI 18 study, demonstrating an increased event rate in troponin-negative women treated with early coronary revascularization.
Our results raise the question whether a gender-differentiated approach to cardiac troponins would be more useful than the application of a single cutoff. However, none of the tested gender-specific cTnI 99th percentiles considerably improved risk prediction. This likely depends on the relative low prevalence of cTnI levels below these cut-offs, which resulted in few re-classifications with respect to cTnI status. Our findings are in line with data from another study in NSTE-ACS patients but do not exclude the possibility that gender-specific 99th percentiles might provide greater clinical value in populations with a higher proportion of troponin levels below the 99th percentile, for example, in nonselected chest pain patients.
There are some limitations to our analysis. The patients from GUSTO IV were randomized >10 years ago and conservatively managed and do, therefore, not represent a contemporary NSTE-ACS cohort. For this reason, we were not able to investigate the value of cTnI levels for targeting of contemporary standard therapies in NSTE-ACS such as P2Y12 inhibition or early invasive treatment. Nevertheless, the design of GUSTO IV enabled us to investigate the prognostic importance of cTnI levels without major interference by early mechanical restoration of coronary blood flow. Cardiac troponin I levels were determined at study inclusion and are not reported as peak levels. This might have affected the prognostic estimates. We are unable to further elucidate the associations between cTnI levels, left ventricular function, and coronary status, as echocardiographic and angiographic data were not systematically collected. Patients had to have either a positive cardiac troponin test or ST-segment depression to be included in the study. We, therefore, lack a cohort presenting with ischemic symptoms only, which should be kept in mind when interpreting the interrelation of cTnI elevation and ST-segment depression on outcome. Finally, although being reported as stable analyte, we cannot exclude some deviation in cTnI results due to the long time of storage.
Discussion
The development of high-sensitivity cardiac troponin assays has taught us 2 major lessons. One is the potential of these assays for the diagnosis of MI within the first hours after admission. The second lesson is their impact on the predictive power of cardiac troponin because concentrations even below the 99th percentile may be predictive of adverse outcomes. This applies both to populations with coronary artery disease, congestive heart failure, or atrial fibrillation but also to community-based cohorts. In the present report, we are taught a third lesson, that of the gender-related differences in NSTE-ACS.
We noted that men with NSTE-ACS had considerably higher cTnI levels compared to women. This is supported by observations from other studies. It has been suggested that the differences in cTnI levels between the genders might mirror differences in the cardiovascular risk pattern between before the occurrence of NSTE-ACS. In fact, women in GUSTO IV were older and exhibited higher prevalence of diabetes, congestive heart failure, and hypertension. Men, in contrast, more often were smokers and had previous manifestations of coronary artery disease. However, in a recent analysis from the PEACE study, only a minor difference in cTnI levels was noted between men and women with stable coronary artery disease (ARCHITECT STAT hsTnI; men 4.6 [3.1–7.4] ng/L, women 4.0 [2.6–6.4] ng/L) (Omland T; personal communication). It, therefore, seems unlikely that the marked difference in cTnI levels between men and women from GUSTO IV might be explained by differences in the cardiovascular risk pattern.
This leads to the question whether our results instead might reflect differences in the pathophysiology of NSTE-ACS. Women usually have a lower prevalence of obstructive coronary artery disease and more often have microvascular and endothelial dysfunction and/or diffuse atherosclerosis. Men, in contrast, more often have complete atherosclerotic plaque ruptures resulting in repetitive distal embolization of thrombotic material. Although not provable within the context of GUSTO IV, we assume that this might contribute to greater myocardial damage in men with higher troponin elevations as seen in women.
However, despite overall lower concentrations, cTnI at the tested 99th percentiles tended to exhibit stronger associations with adverse outcome in women. This was similar for cTnT levels measured with a conventional assay and is on first sight a paradoxical and provocative finding. However, the interaction terms of gender on the association of cTnI with outcome were nonsignificant, and in case of elevated levels, the event rates in men and women were similar. Thus, the difference between men and women was the difference in cTnI levels among those with a favorable outcome and not any differences among those who actually had an event.
Accordingly, the clinical presentation of women with symptoms of unstable coronary artery disease appears to encompass a broader cardiovascular risk panorama compared to men, and this seems to be reflected in the distribution and the prognostic implications of cTnI levels. In this context, it is noteworthy that more women than men were included in GUSTO IV based on ST-segment depression. ST-segment depression in the absence of troponin elevation could be a result of ischemia induced by raised heart work, coronary spasm, or nonischemic causes and thereby mirror less adverse pathophysiologies or no pathology at all. This is illustrated by the low event rate in women without cTnI elevation but with ST-segment depression, which contrasted to men in whom ST-segment depression was strongly prognostic. We, therefore, cannot exclude the possibility that some of the women included in GUSTO IV might have had another condition than NSTE-ACS despite unstable symptoms. This emphasizes the need of additional indicators of coronary instability in troponin-negative women for tailoring of appropriate therapies, especially with respect to the avoidance of potentially harmful procedures. This is illustrated by data from the TACTICS-TIMI 18 study, demonstrating an increased event rate in troponin-negative women treated with early coronary revascularization.
Our results raise the question whether a gender-differentiated approach to cardiac troponins would be more useful than the application of a single cutoff. However, none of the tested gender-specific cTnI 99th percentiles considerably improved risk prediction. This likely depends on the relative low prevalence of cTnI levels below these cut-offs, which resulted in few re-classifications with respect to cTnI status. Our findings are in line with data from another study in NSTE-ACS patients but do not exclude the possibility that gender-specific 99th percentiles might provide greater clinical value in populations with a higher proportion of troponin levels below the 99th percentile, for example, in nonselected chest pain patients.
Limitations
There are some limitations to our analysis. The patients from GUSTO IV were randomized >10 years ago and conservatively managed and do, therefore, not represent a contemporary NSTE-ACS cohort. For this reason, we were not able to investigate the value of cTnI levels for targeting of contemporary standard therapies in NSTE-ACS such as P2Y12 inhibition or early invasive treatment. Nevertheless, the design of GUSTO IV enabled us to investigate the prognostic importance of cTnI levels without major interference by early mechanical restoration of coronary blood flow. Cardiac troponin I levels were determined at study inclusion and are not reported as peak levels. This might have affected the prognostic estimates. We are unable to further elucidate the associations between cTnI levels, left ventricular function, and coronary status, as echocardiographic and angiographic data were not systematically collected. Patients had to have either a positive cardiac troponin test or ST-segment depression to be included in the study. We, therefore, lack a cohort presenting with ischemic symptoms only, which should be kept in mind when interpreting the interrelation of cTnI elevation and ST-segment depression on outcome. Finally, although being reported as stable analyte, we cannot exclude some deviation in cTnI results due to the long time of storage.