Sex Differences in Mortality after Myocardial Infarction: A Retrospective Cohort Study

BACKGROUND: We aimed to examine sex differences in mortality after myocardial infarction. METHODS: This retrospective cohort study included all rst admitted patients 50 years or older with acute myocardial infarction hospitalized in Rajaei hospital of Karaj city, Iran, between 23 th March 2013 and 1 th January 2020. Data was retrieved from the hospital information system (HIS) database, including patient’s demographic and socioeconomic characteristics, medical history, acute myocardial infarction (AMI) type, treatment and procedures, and outcome of hospitalization. Simple and multivariate cox regression models were used to assess the association of gender with in-hospital mortality after AMI. Results were presented as crude and adjusted hazard ratios along with their 95% condence interval (HR (95% CI). RESULTS: Results from the multivariable Cox regression analysis revealed that females had a higher risk of death than males after ST segment Elevation MI (STEMI) (adjusted HR (95% CI): 1.64 (1.15 – 2.36), P=0.007). However, in subgroup analysis by age groups, this signicant increased risk was observed only in female patients aged 50 to 64 years than their male counterparts. There was no signicant differences between males and females after non-STEMI and unspecied MI. CONCLUSIONS: Based on our ndings, women aged 50 to 64 years may be more likely to die during hospitalization after STEMI than men.


Introduction
Cardiovascular disease (CVD) is a main cause of death worldwide, especially in developing countries [1].
In Iran, CVD is the rst cause of death, accounted for about half of all-cause mortality [2]. Although, like some other countries [3] a decreasing trend of mortality rate for CVDs has been observed in both sexes during two last decades [4], partly explained by better management of the diseases after implementation of novel revascularization strategies [3] but it remains high.
Furthermore, contradicting evidence suggest that women compared to men may experience a higher rate of mortality after myocardial infarction (MI), a major form of CVDs [5]. However, most of these evidence comes from the studies performed before the widespread use of revascularization strategies [6].
Up to our knowledge, a few studies have addressed sex differences in mortality after myocardial infarction in Iranian patients, especially within last 5 years [7]. Hence, we aimed to examine if there is a sex differences in mortality after myocardial infarction using data from a large sample of patients.

Study design
We conducted a single-center retrospective cohort study using data of rst-admitted patients 50 years or older with ICD-10-CM codes of acute myocardial infarction (AMI) (ICD-10 code I21) that extracted from the hospital information system (HIS) of Shahid Rajaei hospital in Karaj city, Iran, over a 4-year period.

The study population
The study population was all inpatients 50 years or older who were admitted for the rst time with a diagnosis of AMI (with or without ST segment elevation) in our hospital from 23th March 2013 to 1th January 2020. As novel revascularization techniques including percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) have been performed in our hospital since 2013, the study was restricted to this period.
Karaj city is the center of Alborz Province, Iran. Karaj, with about 1.97 million inhabitants as recorded in the 2016 census, is the fourth-largest city in Iran. Also, Karaj is located 20 kilometers (12 mi) west of Tehran and is a large suburb of it.
Shahid Rajaei is a referral hospital providing outpatient, inpatient and emergency services for patients with AMI in Karaj city.
Management of acute myocardial infarction in Shahid Rajaei hospital is according to the most updated guidelines of the American College of Cardiology and American Heart Association and of the European Society of Cardiology.

Data Sources and Collection
Data on demographic characteristics, medical history, and ICD-10-CM codes of diagnoses and procedures for each patient are recorded and stored in the hospital information system (HIS) by trained staffs.
In this study, we retrieved some baseline characteristic of included patients, including age (years), gender, presence of comorbidities, type of MI, performed procedures, discharge status (dead or alive), date of admission, discharge or death from the HIS database. Patient's age was categorized into three groups: 50-64, 65-74, and ≥ 75 years. Length of the stay in the hospital was calculated in days for each patient.
The clinical data were extracted by using International Classi cation of Diseases, tenth Revision, Clinical Modi cation (ICD-10-CM) codes.
The type of MI was determined according to the ICD-10-CM classi cation as follows: STEMI (codes I21.0-3), unspeci ed (code I21.9) and NSTEMI (code I21.4) Using the ICD-10-CM codes, we identi ed the presence or absence of speci c comorbidities at the time of the admission including heart failure (code I50), hypertension (code I10), diabetes (codes E10-E14), and chronic kidney disease (code N18). Besides, we also used ICD-10-CM codes 00.66and 36.01 to 36.09 to identify patients who underwent PCI, and ICD-9-CM codes 36.10 to 36.19 to identify patients who underwent CABG surgery.

The outcome measurement
The main outcome was in-hospital mortality in patients with AMI. As data on the exact cause of death was not available, in-hospital death was de ned as death due to any cause during the hospital stay.

Statistical analyses
We used descriptive statistics, mean (SD) for continuous variables and as number (percentage) for categorical variables to summarize demographic and clinical data of the cohort. Characteristics of males and females were compared using two-tailed t-tests or chi-square tests, as appropriate.
We also used Univariable and Multivariable Cox proportional hazard regression models to assess the effect of being female on the in-hospital mortality after AMI. The results is presented as a hazard ratio (HR) 95% con dence interval [CIs]). Proportional hazard assumption was evaluated using both graphical methods (ln-ln S [t] graphs) and statistical tests containing continuous time-interaction terms (Cox tests) Various cox regression models were constructed according to adjustment strategy: model I, unadjusted; model II, adjusted for patient age; and model III, adjusted for age, diabetes, arterial hypertension.
We also compared the survival distribution between males and females using the log-rank test. Follow-up period de ned as time (days) from the date of admission until the date of discharge or in-hospital death whichever occurred rst. The probability of survival for males and females was depicted in Kaplan-Meier plots. For all analyses a p-value < 0.05 (two-tailed) was considered statistically signi cant. Analyses were performed using SPSS version 18.0 (SPSS, Chicago, IL, USA) and STATA V.12.0 (Stata Corp, College Station, Texas, USA).

Patient Characteristics
A total of 1,730 rst-admitted patients with AMI older than 50 years were recorded in our HIS database from March 2013 through January 2019. The mean age (standard deviation (SD)) of patients was 64.5 (10.4) and 455 (26.3%) were female. In order by frequency, type of MI was STEMI in 1,176 patients (68.0%), NSTEMI in 330 (19.1%), and unspeci ed MI in 224 (12.9%).
Overall, 7 patients (0.4%) dead in the early hours of hospitalization, 1,430 (82.6%) underwent angiography, and 293 (17.0%) received medical therapies. Among patients who underwent angiography, the performed procedure was PCI in 1,069 patients (74.8%), CABG in 73 (5.1%), PCI & CABG in 13 (0.9%), and "only angiography" in 275 (19.2%). Table 1 shows the patients' characteristics and treatment and procedures performed for them during hospitalization according to the gender. The mean age (SD) of females was signi cantly higher than males (68.3 (10.9) vs. 63.3 (9.9), respectively; P<0.001). Compared to males, a higher percentage of females had a history of hypertension and diabetes. (Both P < 0.001) There was no signi cant difference between both sexes regarding prevalence of comorbidities CHF and CKD.  These ndings were not con rmed by the multivariate Cox regression analysis after considering potential confounder age (adjusted HR (95% CI): 1.30 (0.98 -1.74), P= 0.067) However, in subgroup analysis by age group, the signi cant adjusted effect of being female on in-hospital mortality after AMI was observed only in patients aged 50 to 64 years. (Figure 1) Table 2 Comparison of the all-cause mortality between males and females, totally and by subgroups  analysis by age group, this signi cant effect of being female on in-hospital mortality after AMI was observed only in patients aged 50 to 64 years. (Figure 1) We also performed a log-rank test to determine if there were differences in the survival distribution for males and females in the different age groups: 50-64, 65-74, and 75 + years. The survival distributions for males and females were signi cantly different in the age group 50-64 years (χ 2 (1) = 7.689, p = 0.006), but not in the older age groups. (Figure 2& 3)

Discussion
In this study, we evaluated sex differences in in-hospital mortality following AMI in rst admitted patients.
Based on our ndings, female patients aged 50 to 64 years were more likely to die following STEMI, considering potential confounders age, presence of diabetes and hypertension, than males in the same age group; however, this gender effect was no longer seen in older age groups. Also, in patients hospitalized for non-STEMI, both sexes revealed similar risk for in-hospital mortality.
Some other studies con rm our ndings [6,9]. Recently, in line with our research, Zhang et al. have reported that higher in-hospital mortality risk in females compared to males is more evident in younger patients with STEMI [10]. Also, a systematic review evaluated gender differences in short and long term mortality in patients after STEMI. Six out of eleven included studies focused on in-hospital mortality as the main outcome reported adjusted results, of which two studies found signi cant effects of female gender on higher in-hospital mortality [9b].
The gender differences in the early clinical manifestations of STEMI, more prominent in younger age groups [11] may explain the higher mortality rate in females than males. While speci c symptoms are believed to play a key role in patients' help-seeking behavior, early diagnosis, and timely treatment [12], but based on the available evidence, female patients are more likely to present with atypical symptoms compared to males [13].
However, we were unable to examine this justi cation in the present study due to the lack of data on the initial symptoms, and the time interval between the onset of symptoms and treatment.
There is evidence that physicians are less likely to comply with current guidelines recommendations on treatment options for female mi patients due to the underscoring of their risk [9c]. Indeed higher prevalence of micro-vascular involvement in females makes them less likely to undergo angiography and receive speci c medications [9c]. Leurent et al., accounted "under treatment or conservative treatment of female patients" as the main explanation for higher in-hospital mortality in them compared to males [12].
During perimenopause and early post-menopause, being female may be an independent predictor of inhospital mortality following STEMI. At this age, more involvement of microvascular than macrovascular, and having different etiology of STEMI in females compared to males could put them at a higher risk of Noteworthy, while routine practice and guidelines are mainly determined based on the ndings of clinical trials, a systematic review recently revealed the underrepresentation of females in contemporary acute coronary syndrome (ACS) trials, limiting the generalizability of their nding to the male than female population [14].
In line with our studies, previous studies also found that patients hospitalized for non-STEMI, both sexes had similar risk for in-hospital mortality [15].
In contrast to STEMI that is mainly caused by acute plaque rupture, NSTEMI often resulted from a moderate coronary stenosis [16].

Conclusions
Our ndings suggest that female inpatients aged 50 to 64 years may be more likely to die than males following STEMI, considering potential confounders age, presence of diabetes, and hypertension. Declarations Funding Figure 1 Forest plot of the effect of the being female on in-hospital mortality by type of MI in different age groups, considering potential confounder diabetes, and hypertension: Cox regression analysis CI: con dence interval. ES: hazard ratio. STEMI: ST segment elevation myocardial infarction. AMI: acute myocardial infarction Figure 2 Survival Curve in patients with acute myocardial infarction (AMI) by gender in different age groups (Kaplan Meier survival curve & log rank test)