Although there has been significant improvement in healthcare1,2 and a decrease in mortality,1,3–5 cardiovascular diseases, with the significant weight of myocardial infarction, continue to be the most common cause of death in Europe1,6–8; and their burden is increasing in women.5,9 Improved healthcare is mainly based on three factors: the implementation and dissemination of coronary reperfusion,3 better myocardial revascularization techniques10 and the widespread application of guideline-directed medical therapy.3

Around 30–40% of acute myocardial infarctions (AMI) occur in women; ST-segment myocardial infarction (STEMI) is less likely than non-ST-segment myocardial infarction (NSTEMI).4 STEMI is described as increasing exponentially with age in women compared to a linear rate in men.3

Sex differences in AMI are an issue that is continually under debate,11,12 with conflicting results in the literature,1,7,8,10 partly due to the heterogeneity of studied populations, including both STEMI and NSTEMI.1,13 Although comorbidities, complications, and mortality remain higher in women,1 long-term differences appear to attenuate after adjusted analysis.1,6 These differences may be related to worse baseline characteristics,8,14 pathophysiology and treatment disparities4 in women, but they are not fully understood.1,15,16

This matter has raised awareness not only inside the cardiology community with several publications and campaign advertisements,17 but also among the general population.7 Nevertheless, differences still persist.4,17,18

In Portugal, cardiovascular diseases are the most common cause of death. They are the most common etiology in women, in contrast to oncologic diseases in men.19 A previous Portuguese study on acute coronary syndromes (ACS) identified differences in treatment and prognosis between sexes.5 Nevertheless, as of the present date and to the best of the authors’ knowledge, the specific situation in Portugal, particularly regarding STEMI, is not known. The authors consider that including only STEMI could help reduce the heterogeneity of the population and eliminate potential confounding factors associated with the spectrum of ACS, particularly in diagnosis and treatment.

The purpose of the current analysis is to investigate sex-related differences among patients who have experienced STEMI in Portugal focusing on baseline characteristics, clinical findings, treatment modalities, prognosis and mortality. Based on the literature, the authors expected to find sex-related differences in these patients.

The Portuguese Registry on Acute Coronary Syndromes (ProACS) was established in 2002, founded by the Portuguese Society of Cardiology and is coordinated by the National Center for Data Collection in Cardiology. This multicenter, continuous, prospective observational national registry is registered at ClinicalTrials.gov (NCT01642329).20 All Portuguese cardiology departments were invited to participate and consecutively include adult patients who were hospitalized. The inclusion and exclusion criteria were established and published elsewhere.20

By 2016, a total of 49 Portuguese centers had been included (11 university hospitals, 12 hospitals with cardiac surgery and 25 with catheterization laboratories), distributed across Portugal: North (11 centers), Central region (12 centers), South (22 centers) and in the Autonomous Regions of Madeira and the Azores (four centers).20

The authors conducted a retrospective analysis of STEMI patients included in the ProACS between October 2010 and 2022.

Between October 2010 and 2022 a total of 14470 STEMI patients were registered on ProACS.

Survival analysis and adjusted analysis

Survival analysis is shown in Figure 1. Multivariate logistic regression for in-hospital mortality and Cox proportional hazard regression for one-year mortality are shown in Tables 6 and 7, respectively.

Figure 1.

Kaplan–Meier curves for one-year mortality.

(0.18MB).

Table 6.

Multivariate analysis for in-hospital mortality (other variables included not shown).

Variables	OR	95% CI	p-Value
Sex (F. vs. M.)	1.672	1.078–2.593	0.022
GRACE risk score >140	3.802	1.543–9.366	0.004
BB (inpatient)	0.272	0.166–0.445	<0.001
AECi (inpatient)	0.492	0.298–0.814	0.006
Statin (inpatient)	0.213	0.118–0.384	<0.001
Inotropes (inpatient)	9.082	5.598–14.757	<0.001
Amiodarone (inpatient)	2.711	1.644–4.468	<0.001
Diuretics (inpatient)	1.710	1.040–2.812	0.034
LVEF <30%	20.231	10.648–38.439	<0.001

CI: confidence interval; OR: odds ratio.

Table 7.

Cox's hazard regression (one-year mortality) (other variables included not shown).

Variables	HR	95% CI	p-Value
Sex (F. vs. M.)	1.158	0.851–1.590	0.359
GRACE risk score >140	2.100	1.196–3.687	0.010
Age >75 vs. <60 years	2.683	1.560–4.616	<0.001
Killip class >1	1.524	1.062–2.187	0.022
Dyslipidemia	0.616	0.455–0.833	0.002
Heart failure	1.78	1.039–3.049	0.036
Ischemic stroke/TIA	1.163	1.043–1.297	0.007
ACEi (inpatient)	0.543	0.382–0.773	<0.001
CCB (inpatient)	2.231	1.233–4.038	0.008
Diuretics (inpatient)	1.627	1.123–2.358	0.010
LVEF <50% vs. >50%	1.930	1.362–2.735	<0.001
LVEF <30% vs. >30%	1.817	1.124–2.937	0.015
HF (complication)	1.541	1.074–2.210	0.019
Aspirin (outpatient)	0.454	0.279–0.738	0.001

CI: confidence interval; HR: hazard ratio.

After adjusted analysis, female sex remained as an independent factor for in-hospital mortality (OR=1.672; 95% CI [1.078–2.593]; p=0.022) but not for one-year mortality.

Regarding in-hospital mortality, several statistically significant variables in univariate analysis – including baseline characteristics such as HBP, diabetes, angina, VHD, HF, history of smoking, cerebrovascular disease, PAD, CKD, cancer, chronic obstructive pulmonary disease, clinical findings with GRACE risk score and STEMI localization, as well as medication use both prior to and during hospitalization and angiographic characteristics like multivessel disease, culprit artery, and time to reperfusion – were incorporated into the regression model. Likewise, the regression model for one-year mortality included HBP, VHD, PAD, CKD, cancer, prior, inpatient and outpatient medications, the use of PCI, angiographic characteristics (multivessel disease), and in-hospital complications.

GRACE risk score, inotropes and LVEF lower than 30% were the most important independent predictors for in-hospital mortality. Guideline-directed medical therapy was associated with a reduced risk of in-hospital mortality. Moreover, older age, in-hospital complications, worse clinical profiles at admission, including GRACE risk score, and lower LVEF were independent predictors for one-year mortality.

Approximately one-quarter of our population was female, consistent with the underrepresentation of women in clinical studies regarding ACS.6,9,15,21

Similar to other European registries – such as in Estonia, Hungary, Norway, Sweden and Vienna – and to other studies,1,3,6–8,10,16 women were older and had a higher frequency of several cardiovascular risk factors. Coronary artery disease appears to be age-dependent in women, especially occurring after menopause6 and typically five to 10 years after men,6,21 due to the protective effects of estrogen.21 Moreover, these risk factors appear to have a more unfavorable role in women – for example, the OR of AMI in diabetic women is higher compared to men (4.26 vs. 2.67).19 Unassessed comorbidities, such as depression,13,21 anxiety and stress21 may be related to worse outcomes. The prevalence of depression after an AMI is estimated to be 20% and doubles in women compared to men.12 A study found higher early complication rates in patients with anxiety or depression and proposes the use of screening tools to manage this comorbidity better.12 These data probably justify the inclusion of these new variables in future registries.

Chest pain is the most common symptom of acute ACS.11 Nevertheless, atypical symptoms appear to be more common among women,1,7,21 in accordance with the present study, which is an alarming feature that can lead to misdiagnosis and undertreatment.1 Furthermore, women appear to wait longer before seeking healthcare3,7,11,13,16,17; this was also observed in the present study. Reducing this delay may have potential benefits on mortality and morbidity.3 In our analysis, despite a higher prevalence of obesity, BNP levels in women were twice as high as those in men, probably corroborating a delayed and worse clinical presentation, as well as the presence of other comorbidities known to increase BNP levels (anemia, CKD). Awareness around clinicians should be raised, and effective educational campaigns targeting the general population should address this issue.

The management of STEMI differed based on sex1,6,7,16; women underwent CAG, PCI, and guideline-directed medical therapy less frequently. Accordingly, a study found lower prescription of ticagrelor and prasugrel in women.22 Lower prescription of dual antiplatelet (DAPT) therapy could be explained by low adherence to guideline recommendations in conservative cases and also due to higher prescription of oral anticoagulants1; this was presumed in the current study due to the higher frequency of atrial fibrillation and anticoagulation at discharge. While the rate of major bleeding in our population was low, concerns about higher bleeding risk in women may also contribute to lower adherence to DAPT.1,22 Thus, these differences may be attributed to older age and frailty,1 which is consistent with our population of frailer women. This characteristic was, however, inferred through other features.

In another recent study among patients with ACS, similar tendencies were observed with less guideline-directed therapy during both prehospital and hospital care in women.11 It is suggested that this difference is greater in prehospital care,11 which is a parameter that was not evaluated in our study. Ours reveals differences in discharge medications that could be partly explained by increased comorbidities in women. Nevertheless, clinicians should be aware of these differences in order to prescribe optimized medical therapy and individualize treatment regardless of the patient's sex.

On the other hand, the lower rates of PCI in our study may be partially explained by the higher number of normal CAG findings in women, while the success of PCI for the most common culprit artery was similar, suggesting that equivalent treatment was offered in cases that underwent PCI. Myocardial infarction with non-obstructive coronary arteries (MINOCA) is a heterogeneous entity more frequently found in women and associated with a higher risk of cardiovascular events19 that should be properly addressed. However, as ProACS does not collect information regarding MINOCA, this analysis could not be performed in the present study.

Our analysis revealed a more significant role of sex in in-hospital mortality. There is consistent evidence that women have higher short-term mortality, while there are conflicting results for long-term mortality.6,7 In a recent metanalysis, the effect of difference in sex on short-term mortality was significant, in contrast to long-term mortality, even after attenuation with adjusted analysis.7 In contrast, women had higher short and long-term mortality rates in several registries, although both these differences seem to attenuate after adjusted analysis in European countries,1,16 but not in India.9 The justification for these differences is not known, although frailty1 and worse cardiovascular profiles7 may play a role.

An international study regarding thirty-day mortality found statistically higher mortality in women after adjusted analysis but that these differences were stronger in younger patients, and not related to treatment variations.13 Other studies in younger women had similar results.21 Conversely, a study focusing on differences in STEMI patients who underwent PCI, did not find differences in thirty-day and one-year mortality.10

Most of these studies are based on registries and observational data and one-year follow-up.6 A subanalysis from the EXAMINATION-EXTENDED trial, with a 10-year follow-up, encountered a higher incidence of all-cause and cardiac death in women. After adjusted analysis of its primary endpoint, a composite of all-cause death, myocardial infarction and revascularization, sex was not a predictor of events, while age was the most important one.6 The authors suggest that HF and cardiogenic shock may lead to worse outcomes in women, as they are related to higher in-hospital and thirty-day mortality,6 similar to our findings. Other complications, including bleeding and mechanical complications were also reported as more common among women.7

In our analysis, lower LVEF was associated with mortality, which is similar to previous studies in which LVEF was an independent predictor of in-hospital complications.12 On the other hand, with regard to noncardiac comorbidities and death,6 the delay in seeking medical observation and in diagnosis,13 may play a role in all-cause death. In this context, one study focused on in-hospital STEMI and found there were fewer disparities in baseline characteristics between sexes, although longer times, less guideline-medical therapy in women, and no significant differences in mortality.2

There are several recommendations to improve outcomes in women, including education campaigns for the general population, particularly women, education campaigns for healthcare providers and the implementation of protocols for STEMI management.17 In fact, STEMI management protocols may reduce sex differences,10,14,17 and one study observed a reduction in thirty-day mortality in women.17 Further research is of utmost importance, and the use of artificial intelligence may help reduce existing bias.17

There are sex-based differences in the pathophysiology of acute myocardial infarction (AMI)1,5,8,13,14 which may play a role in mortality.13 While plaque rupture and thrombus formation are established mechanisms in men, plaque erosion,4 inflammation, endothelial dysfunction, coronary vasospasm, spontaneous coronary dissection and connective tissue disorders may have a role in women.1 Furthermore, women have smaller vessels, fewer collaterals and increased vascular stiffness.13 Estrogen may act as a plaque stabilizer, and the small vessel size and higher blood flow may lead to increased endothelial shear stress, thus contributing to plaque erosion.4

Differences in thrombogenicity between sexes may also be a plausible cause related to sex hormones; a study found that women with high thrombus burden in an acute setting had higher cardiovascular mortality compared to men. Moreover, another prospective study found higher clopidogrel resistance in women. High thrombus burden appears to be important in prognosis and may indicate the need for a more aggressive treatment.15 These factors were not accounted for in our study.

Despite the conflicting results, sex-based differences in AMI cannot be denied.7

There are important sex-related differences in Portuguese STEMI patients. Compared with men, women were older and had a worse cardiovascular risk profile. In terms of clinical management, women were less likely to receive reperfusion therapy as well as other important medications with impact on prognosis. In-hospital mortality was higher in women. Nevertheless, this difference disappeared after attenuation analysis at the end of the first year.

According to our results there is room to improve the clinical management of women with STEMI. Efforts should be made toward achieving an equal standard of care. Standardized diagnosis and treatment protocols, as well as education campaigns targeting the general population and clinicians, may help reduce this sex bias. Noncardiac comorbidities should also be accounted for.

For a better understanding of sex differences in STEMI patients, further studies are needed, as they may help to improve medical therapy tailored according to sex and to reduce AMI mortality.

During the preparation of this work the author(s) used ChatGPT 3.5 to correct grammar errors. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.

The ProACS registry is approved by the Portuguese Data Protection Authority (No. 3140/2010) and registered at ClinicalTrials.gov (NCT01642329).

The authors have no conflicts of interest to declare.