ORIGINAL RESEARCH Sex-Specific Treatment Effects After Primary Percutaneous Intervention: A Study on Coronary Blood Flow and Delay to Hospital Presentation Edina Cenko MD, PhD; Mihaela van der Schaar, PhD; Jinsung Yoon, MSc; Sasko Kedev, MD, PhD; Marija Valvukis, MD; Zorana Vasiljevic, MD, PhD; Milika Asanin, MD, PhD; Davor Milicic, MD, PhD; Olivia Manfrini, MD; Lina Badimon, PhD; Raffaele Bugiardini, MD Background-—We hypothesized that female sex is a treatment effect modifier of blood flow and related 30-day mortality after primary percutaneous coronary intervention (PCI) for ST-segment–elevation myocardial infarction and that the magnitude of the effect on outcomes differs depending on delay to hospital presentation. Methods and Results-—We identified 2596 patients enrolled in the ISACS-TC (International Survey of Acute Coronary Syndromes in Transitional Countries) registry from 2010 to 2016. Primary outcome was the occurrence of 30-day mortality. Key secondary outcome was the rate of suboptimal post-PCI Thrombolysis in Myocardial Infarction (TIMI; flow grade 0–2). Multivariate logistic regression and inverse probability of treatment weighted models were adjusted for baseline clinical covariates. We characterized patient outcomes associated with a delay from symptom onset to hospital presentation of ≤120 minutes. In multivariable regression models, female sex was associated with postprocedural TIMI flow grade 0 to 2 (odds ratio [OR], 1.68; 95% CI, 1.15– 2.44) and higher mortality (OR, 1.72; 95% CI, 1.02–2.90). Using inverse probability of treatment weighting, 30-day mortality was higher in women compared with men (4.8% versus 2.5%; OR, 2.00; 95% CI, 1.27–3.15). Likewise, we found a significant sex difference in post-PCI TIMI flow grade 0 to 2 (8.8% versus 5.0%; OR, 1.83; 95% CI, 1.31–2.56). The sex gap in mortality was no longer significant for patients having hospital presentation of ≤120 minutes (OR, 1.28; 95% CI, 0.35–4.69). Sex difference in post- PCI TIMI flow grade was consistent regardless of time to hospital presentation. Conclusions-—Delay to hospital presentation and suboptimal post-PCI TIMI flow grade are variables independently associated with excess mortality in women, suggesting complementary mechanisms of reduced survival. Clinical Trial Registration-—URL: http://www.clinicaltrials.gov. Unique identifier: NCT01218776. ( J Am Heart Assoc. 2019;8: e011190. DOI: 10.1161/JAHA.118.011190.) Key Words: coronary blood flow • mortality • sex differences • ST-segment–elevation myocardial infarction P ossible sex-based differences with respect to atheroscle- rotic disease have become a subject of intense research and debate. Atherosclerosis tends to affect men and women irrespective of reperfusion modalities.1–7 These findings have raised the physician’s interest in finding out more about the reasons for this large sex gap. Yet, such information is missing differently and at different times in their lives. One of the most because studies on pathophysiological characteristics of intriguing findings is that women presenting with ST-segment– ischemic heart disease have mainly focused on women with elevation myocardial infarction (STEMI) die more often than men the exclusion or underrepresentation of men.8 From the Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, Bologna, Italy (E.C., O.M., R.B.); University of Cambridge, Cambridge, United Kingdom (M.v.d.S.); Department of Electrical and Computer Engineering, University of California, Los Angeles, Los Angeles, CA (J.Y.); Medical Faculty, University Clinic of Cardiology, University “Ss Cyril and Methodius,” Skopje, Macedonia (S.K., M.V.); School of Medicine, University of Belgrade, Belgrade, Serbia (Z.V.,M.A.); Department of Cardiology, Clinical Centre of Serbia, Belgrade, Serbia (M.A.); Department for Cardiovascular Diseases, University Hospital Center Zagreb, University of Zagreb, Zagreb, Croatia (D.M.); and Cardiovascular Program (ICCC), IR-Hospital de la Santa Creu i Sant Pau. CiberCV-Institute Carlos III. Autonomous University of Barcelona, Barcelona, Spain (L.B.). Accompanying Data S1, Tables S1 through S11 and Figures S1, S2 are available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.118.011190 Correspondence to: Raffaele Bugiardini, MD, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, Policlinico Sant’Orsola Malpighi, Padiglione 11, Via Massarenti 9, 40138 Bologna, Italy. E-mail: raffaele.bugiardini@unibo.it Received October 10, 2018; accepted December 13, 2018. ª 2019 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 1 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Clinical Perspective Methods Study Population deos Is New? The authors declare that all supporting data are available • In the current study, compared with men, postprocedural within the article and its supplemental material. Thrombolysis in Myocardial Infarction flow grade rates were The ISACS-TC (International Survey of Acute Coronary lower and 30-day mortality was higher in women. Syndromes in Transitional Countries; ClinicalTrials.gov: • The distribution of covariates was similar between women NCT01218776) registry is a large, prospective, multicenter and men. cohort study designed to record clinical characteristics, • Delay to hospital presentation also modified comparative effectiveness between sexes because the sex gap in treatments, and clinical outcomes for patients with acute mortality was no longer significant for patients presenting coronary syndrome.5,15 The ISACS-TC registry enrolled 5590 to hospital within 120 minutes. patients who underwent primary PCI for STEMI within 12 hours from symptom onset between January 2010 and January 2016. What Are the Clinical Implications? The designated physician collected the registry data at the time • Differences between men and women in post primary of clinical assessment. Patients were asked about the quality of percutaneous coronary intervention mortality rates exist their symptoms. All patients presented with chest pain or even when women and men are balanced in terms of equivalent symptoms, such as dyspnea and fatigue. Data on baseline characteristics, treatment, and time to hospital coronary hemodynamics were available in 2651 of the 5590 presentation. patients. In the remaining patients, the ISACS-TC registry • Delay to hospital presentation and suboptimal postproce- generated reports with summarizing information on PCI dural Thrombolysis in Myocardial Infarction flow grade are procedural success without any further detail. Among the variables independently associated with excess mortality in 2651 patients with data on coronary hemodynamics, those women, suggesting complementary mechanisms of reduced who had an unknown time of symptom onset (n=55) were survival. • Sex as a biological variable should be factored into research excluded, giving a final study population of 2596 patients designs, analyses, and reporting in human studies. (Figure). Patients who were transferred from a PCI noncapable hospital were included in this analysis. Time of symptom onset relied on patient recall and required documentation by each Coronary blood flow is a major determinant of patients’ hospital participating in this registry. Delay time from symptom prognosis.9 Restoration of blood flow after recanalization of onset to hospital presentation included door-in to door-out the infarct-related artery is commonly quantified with the time, defined as the duration of time from arrival to discharge at Thrombolysis in Myocardial Infarction (TIMI) grading system.10 the first referral hospital, and transfer time to a STEMI receiving Failure to account for sex differences in coronary blood flow hospital for primary PCI. Because delay times are a continuous after primary percutaneous coronary intervention (PCI) could variable, we characterized time to reperfusion and patient explain contradictory findings on the rates of mortality among outcomes associated with a time from symptom onset to women and men.11 Abnormal coronary reactivity and hospital presentation of ≤120 minutes and door-to-balloon microvascular dysfunction are more prevalent among time of ≤90 minutes according to prior work.4,14 The local women12 and may be associated with lower TIMI flow grades research ethics committee from each hospital approved the and adverse outcomes. In addition, myocardial regions study. The data-coordinating center has been established at subjected to prolonged ischemia exhibit loss of microvascular the University of Bologna (Bologna, Italy). The study was integrity with endothelial swelling and edema.13 The extent of approved by the local research ethics committee from each lesion depends on the duration of ischemia, and women have hospital. Because patient information was collected anony- longer delays to hospital presentation compared with mously, institutional review boards waived the need for men.4,5,14 individual informed consent. The purpose of our study was to characterize clinical and angiographic variables potentially associated with suboptimal coronary blood flow and their association with outcomes in a Outcomes large cohort of patients undergoing primary PCI. The primary The primary outcome was the incidence of 30-day all-cause hypothesis was that post-PCI coronary blood flow might differ mortality. Key secondary outcome was the incidence of for men and women and may partially account for increased suboptimal TIMI flow grade 0 to 2 after PCI. We also studied 30-day mortality among women. We also hypothesized that the likelihood of having ischemic stroke and bleeding compli- the magnitude of the effect on outcomes differs depending on cations. Ischemic stroke was diagnosed on the basis of an delay to hospital presentation. imaging study. Bleeding complications were defined as DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 2 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Figure. Flow diagram. STEMI indicates ST-segment–elevation myocardial infarction. previously reported by the TIMI definitions.16 Decreases in the Pearson’s v² test for categorical variables and the 2- hemoglobin ≥5 g/dL occurring during hospitalization or sample t test for continuous variables. We had complete intracranial hemorrhage classified bleeding as major. The use data on time from symptom onset to PCI and mortality. of medications given at hospital admission was also noted. Some patients had missing data on other variables. We imputed the missing values of the clinical variables whose missing rate was <10% using IVEWARE software.18 Only 1 TIMI Flow Grading variable, Killip class, had missing rates that exceeded 10%. The TIMI flow of the infarct artery was estimated before and We used k-nearest neighbor algorithms as imputation after completion of PCI, according to 4 grades of flow, as method19–21 (Data S1). Estimates of the odds ratios (ORs) previously described.10 In brief, grade 0 perfusion is no and associated 95% CIs were obtained with the use of antegrade flow beyond the point of occlusion; grade 1 is faint multivariable logistic regressions. Fixed covariates included antegrade coronary flow beyond the occlusion with incom- demographic information and baseline clinical characteristics plete filling of the distal coronary bed; grade 2 is complete but (Tables 1 and 2). We stabilized weights to reduce the delayed filling of distal coronary bed; and grade 3 is normal variability and ensure that the estimated treatment effect flow with complete filling of the distal bed at a normal rate. remains balanced.22 Weighted t tests and weighted v2 tests Patients were divided into 2 groups on the basis of the were used in the inverse probability of treatment weighting postprocedural TIMI flow in the ischemia-related artery: TIMI (IPTW) analyses to compare continuous or categorical grades ≤2 (no reflow) and TIMI grade 3. The definition of no variables in women and men (Data S1). Furthermore, to reflow required the absence of coronary dissection or assess significant heterogeneity of outcomes in function of significant (≥25%) residual stenosis that could have caused sex and delay from symptom onset to hospital presentation, a decrease in coronary blood flow.17 we made statistical comparisons across 2 delay cohorts (<120 and ≥120 minutes). We used IPTW because it is the simplest method that adjusts for the confounding effect of Statistical Analysis time-varying covariates.23 Multivariable-adjusted regression Patient and procedure characteristics were examined accord- was inadequate in controlling time-varying confounding24 ing to sex. Statistical testing was performed with the use of and was not used for this task (Data S1). DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 3 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Table 1. Baseline Characteristics of Patients With STEMI Sorted by Sex Characteristics Women (N=673) Men (N=1923) P Value Age, meanSD, y 64.811.5 58.811.1 <0.001 Cardiovascular risk factors, n (%) Family history of CAD 224 (33.3) 595 (30.9) 0.265 Diabetes mellitus 163 (24.2) 392 (20.4) 0.043 Hypertension 500 (74.3) 1198 (62.3) <0.001 Hypercholesterolemia 340 (50.5) 838 (43.6) 0.002 Current smoking 278 (41.3) 1023 (53.2) <0.001 Former smoking 45 (6.7) 286 (14.9) <0.001 Previous cardiovascular disease, n (%) Previous angina pectoris 160 (23.8) 354 (18.4) 0.004 Previous myocardial infarction 36 (5.3) 149 (7.7) 0.024 Previous PCI 12 (1.8) 88 (4.6) <0.001 Previous CABG 0 (0.0) 11 (0.6) 0.001 Peripheral artery disease 5 (0.7) 27 (1.4) 0.121 Previous heart failure 38 (5.6) 95 (4.9) 0.488 Previous stroke 16 (2.4) 46 (2.4) 0.983 Clinical presentation ST-segment elevation in anterior leads, n (%) 232 (34.5) 745 (38.7) 0.047 Killip class ≥2, n (%) 100 (14.9) 215 (11.2) 0.018 Systolic blood pressure at baseline, meanSD, mm Hg 134.626.6 136.624.5 0.094 Heart rate at baseline, meanSD, beats/min 79.317.6 79.417.0 0.854 Serum creatinine at baseline, meanSD, lmol/L 79.044.7 94.452.3 <0.001 Medications administered within 24 h from admission, n (%) Aspirin 667 (99.1) 1914 (99.5) 0.283 Clopidogrel 663 (98.5) 1856 (96.5) 0.001 Unfractionated heparin 377 (56.0) 1118 (58.1) 0.340 Low-molecular-weight heparins 318 (47.3) 933 (48.5) 0.571 Heparins (overall) 627 (93.2) 1809 (94.1) 0.415 Glycoprotein IIb/IIIa inhibitor 121 (18.0) 363 (18.9) 0.604 Outcomes 30-d Mortality, n (%) 40 (5.9) 45 (2.3) <0.001 30-d Mortality, OR (95% CI) 2.64 (1.66–4.17) <0.001 TIMI flow grade ≤2, n (%) 54 (8.0) 91 (4.7) 0.004 TIMI flow grade ≤2, OR (95% CI) 1.76 (1.21–2.52) 0.002 CABG indicates coronary artery bypass graft; CAD, coronary artery disease; OR, odds radio; PCI, percutaneous coronary intervention; STEMI, ST-segment–elevation myocardial infarction; TIMI, Thrombolysis in Myocardial Infarction. Results optimized regression coefficient (b) and constant term (a) for the logistic regression (Table S2). Women presented at an older Similarities and differences between women and men excluded age than men. They were more likely to have diabetes mellitus, from the present analysis are shown in Table S1. Among the hypertension, and congestive heart failure, but less likely to 2596 patients enrolled in the study, there were 1923 men (74%) have a history of smoking. Crude mortality rates among women and 673 women (26%) (Table 1). There was an imbalance of and men were 5.9% versus 2.3%. Angiograms revealed 8.1% of clinical covariates between women and men, as shown by the women and 4% of men with post-PCI TIMI flow grade 0 to 2. DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 4 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Table 2. IPTW: Women Versus Men Characteristics Women (n=673) Men (n=1923) P Value Age, meanSD, y 61.811.6 60.311.4 0.003 Cardiovascular risk factors, % Family history of CAD 34.9 31.9 0.153 Diabetes mellitus 22.5 21.2 0.480 Hypertension 68.4 65.0 0.109 Hypercholesterolemia 46.6 44.0 0.243 Current smoking 47.9 50.2 0.304 Former smoking 9.5 12.6 0.031 Previous cardiovascular disease, % Previous angina pectoris 20.7 19.4 0.466 Previous myocardial infarction 5.6 7.1 0.180 Previous PCI 1.9 3.9 0.013 Previous CABG 0 0.4 0.112 Peripheral artery disease 0.7 1.2 0.286 Previous heart failure 5.2 5.1 0.919 Previous stroke 3.0 2.4 0.398 Clinical presentation at admission ST-segment elevation in anterior leads, % 37.4 37.7 0.890 Killip class ≥2, % 19.2 19.3 0.954 Systolic blood pressure at baseline, meanSD, mm Hg 134.827.6 136.024.3 0.324 Heart rate at baseline, meanSD, beats/min 79.017.2 79.217.4 0.809 Serum creatinine at baseline, meanSD, lmol/L 77.828.2 91.746.4 <0.0001 Outcomes 30-d Mortality, % 4.8 2.5 0.0024 30-d Mortality, OR (95% CI) 2.00 (1.27–3.15) 0.0029 TIMI flow grade ≤2, % 8.8 5.0 0.0003 TIMI flow grade ≤2, OR (95% CI) 1.83 (1.31–2.56) 0.0004 CABG indicates coronary artery bypass graft; CAD, coronary artery disease; IPTW, inverse probability of treatment weighting; OR, odds radio; PCI, percutaneous coronary intervention; TIMI, Thrombolysis in Myocardial Infarction. Angiographic and Procedural Characteristics greatly, as we observed in our population for women versus men. We, therefore, balanced the distribution of covariates Procedural characteristics stratified by sex are illustrated in using a nonparametric balancing strategy by weighting Table S3. Women and men had similar rates of preprocedural (IPTW). The 2 groups (women and men) were well balanced TIMI flow grade 3 (26.4% versus 31.1%; P=0.13). There was no (Table 2). The primary outcome of death occurred in 4.8% sex difference in frequency of multivessel disease, acute of women compared with 2.5% of men (reference, men; OR, vessel closure, acute thrombosis, and bifurcation lesions. 2.00; 95% CI, 1.27–3.15). Moreover, we found a significant Bleeding complications and ischemic strokes occurred infre- sex difference in the secondary outcome of post-PCI TIMI quently and were not significantly different between the 2 flow grade ≤2 (reference, men; OR, 1.83; 95% CI, 1.31– groups. 2.56). To determine the relation between sex and severe no reflow complicating primary PCI, we excluded patients who Balancing Covariates: Descriptive Baseline had TIMI flow grade 2. The sex difference in the incidence Comparison of TIMI flow grade ≤1 was similar (OR, 1.51; 95% CI, 1.02– Standard parametric adjustment by regression could be 2.26) to that seen for the entire cohort (TIMI flow grade ≤2) sensitive to model misspecification when groups differ (Table S4). DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 5 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Multivariable Analysis of Baseline Clinical Factors although smaller in effect size than, those seen in the overall Associated With Outcomes population. There were sex differences for late STEMI presentations (delay to hospital presentation ≥120 minutes), To reinforce our data, we also estimated the multivariable- with women having worse outcomes (3.9% versus 2.0%; OR, adjusted effect of PCI on TIMI flow grade 0 to 2 and 30-day 1.94; 95% CI, 1.06–3.57). In contrast, there was no significant mortality. Although the odds of a multivariable-adjusted difference in the rates of death in women and men with timely regression cannot calculate the absolute risk, female sex (<120-minute) hospital presentation (1.1% versus 1.4%; OR, was still a predictor variable for the association with post- 0.74; 95% CI, 0.13–4.26). procedural TIMI flow grade 0 to 2 (OR, 1.68; 95% CI, 1.15– 2.44) (Table S5) and 30-day mortality (OR, 1.72; 95% CI, 1.02– 2.90) (Table S6). It follows that linear regression Delay to Treatment and Mortality in Patients With adjustment resulted in conclusions similar to those obtained TIMI Flow Grade 0 to 2 using IPTW methods. Next, we focused on patients who did not benefit from PCI intervention (TIMI flow grade 0–2). Figure S2 summarizes the Delay to Hospital Presentation relationship between delay to hospital presentation and Median time from symptom onset to hospital presentation mortality in patients with post-PCI TIMI flow grades 0 to 2. was longer for women (280 versus 240 minutes), with only The incidence of post-PCI TIMI flow grades 0 to 2 in the 23.2% of women versus 29.1% of men having a <120-minute overall population was 8.0% in women and 4.7% in men delay from symptom onset to admission (P=0.002) (Table S7). (Table 1). The unadjusted 30-day mortality rates associated There were no differences between sexes in median door-to- with post-PCI TIMI flow grades 0 to 2 were 2.5% in women balloon time (40 versus 38 minutes). The unadjusted risk of and 0.5% in men. Sex differences in patients with early, mortality increased with increasing time from symptom onset within 120 minutes, hospital presentations were comparable to hospital presentation (Figure S1). We therefore explored to those of the overall population (Table 3). There was a whether female sex could be a contributing factor that large difference in the incidence of post-PCI TIMI flow grades increases the risk of mortality when hospital presentation 0 to 2 between women and men (7.9% versus 1.7%). Sex times were comparable among women and men. differences were less pronounced in late hospital presenta- tions, in which the incidence of post-PCI TIMI flow grades 0 to 2 was 9.4% in women and 6.3% in men. Unadjusted Sex Differences in Outcomes in Function of Delay mortality rates in patients with TIMI flow grades 0 to 2 were to Hospital Presentation >2-fold higher in women compared with men in either early We assessed how differences in time to hospital presentation or late hospital presentation (28.6% versus 10.0% and 31.9% could translate into sex-specific PCI outcomes using IPTW versus 11.1%, respectively). TIMI flow grades 0 to 2 (Table 3). For delayed time to hospital presentation ≥120- remained a correlate of higher 30-day mortality in women minute benchmark, the mortality rates were 5.5% in women versus men even after removing confounding by IPTW and 2.8% in men. When women and men presented to hospital (Tables S8 and S9). within 120 minutes, their outcomes were equally positive (2.0% versus 1.6%). On the contrary, the incidence of post-PCI Age Groups and TIMI Flow Grade 0 to 2 TIMI flow grade ≤2 rate was higher in women compared with men regardless of time to hospital presentation: 9.4% versus We investigated the role of sex and age on TIMI flow by 6.3% in patients with delay to hospital presentation >120 stratifying patients into 2 groups: aged <60 and ≥60 years. minutes and 7.9% versus 1.6% in those with delay <120 TIMI flow grades 0 to 2 remained an independent correlate of minutes. female sex only in patients aged ≥60 years (OR, 1.56; 95% CI, 1.08–2.26) (Tables S10 and S11). Delay to Treatment and Mortality in “Flow Benefiters” Discussion We wanted to create a challenge in defining the effect of delay Sex differences in mortality can be seen only in patients with to hospital presentation on the primary outcome of 30-day STEMI.25–33 Many of the differences in outcomes have been mortality independently of the occurrence of suboptimal attributed to sex differences in treatment because women blood flow (grade 0–2). We, therefore, restricted our analysis present to hospital later than men and are less likely to to patients who fully benefited from PCI intervention (TIMI undergo primary PCI.30,31 There are few studies addressing flow grade 3) (Table 4). The results were consistent with, the issue of sex differences in outcome after primary DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 6 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Table 3. IPTW: Women Versus Men Sorted by Delay Cohorts Timely (<120-min) PCI Delayed (≥120-min) PCI Variable Women (n=156) Men (n=560) P Value Women (n=517) Men (n=1363) P Value Age, meanSD, y 61.211.6 58.211.1 0.003 62.111.5 61.111.4 0.093 Cardiovascular risk factors, % Family history of CAD 34.3 34.5 0.963 33.9 31.0 0.228 Diabetes mellitus 17.7 16.3 0.678 23.8 23.1 0.748 Hypertension 68.7 59.9 0.045 69.3 66.9 0.321 Hypercholesterolemia 45.2 43.6 0.722 48.0 44.3 0.150 Current smoking 58.2 56.7 0.738 44.4 47.6 0.214 Former smoking 4.0 10.6 0.011 11.2 13.4 0.202 Previous cardiovascular disease, % Previous angina pectoris 20.0 17.6 0.491 20.9 20.1 0.701 Previous myocardial infarction 7.0 9.4 0.351 6.0 6.4 0.749 Previous PCI 3.1 6.1 0.145 1.6 3.0 0.089 Previous CABG 0 0.3 0.4813 0 0.5 0.101 Peripheral artery disease 0.8 0.9 0.9041 0.7 1.4 0.217 Previous heart failure 6.8 4.0 0.140 5.4 5.6 0.866 Previous stroke 4.1 1.9 0.112 2.8 2.7 0.905 Clinical presentation at admission ST-segment elevation in anterior leads, % 42.4 42.7 0.946 34.9 35.9 0.686 Killip class ≥2, % 15.6 13.7 0.547 20.2 21.5 0.538 Systolic blood pressure at baseline, meanSD, mm Hg 137.226.4 135.723.7 0.503 134.927.6 136.224.5 0.354 Heart rate at baseline, meanSD, beats/min 77.113.5 77.8169 0.669 79.117.6 79.717.5 0.475 Serum creatinine at baseline, meanSD, lmol/L 72.819.4 90.735.4 <0.0001 78.228.8 92.150.2 <0.001 Outcomes 30-d Mortality, % 2.0 1.6 0.713 5.5 2.8 0.004 30-d Mortality, OR (95% CI) 1.28 (0.35–4.69) 0.714 2.03 (1.23–3.33) 0.005 TIMI flow grade ≤2, % 7.9 1.7 0.0001 9.4 6.3 0.02 TIMI flow grade ≤2, OR (95% CI) 4.82 (2.04–11.41) 0.0003 1.54 (1.07–2.23) 0.021 CABG indicates coronary artery bypass graft; CAD, coronary artery disease; IPTW, inverse probability of treatment weighting; OR, odds radio; PCI, percutaneous coronary intervention; TIMI, Thrombolysis in Myocardial Infarction. PCI.1–7,32,34–37 Yet, most of these studies still documented sex-by-PCI treatment interaction. Furthermore, these findings persistently higher rates of adverse events in women demonstrate that delay to hospital presentation modifies the compared with men.1–7 The unsolved issue remains why? effects of PCI between sexes. Delay and suboptimal postpro- cedural TIMI flow grade are variables independently associ- ated with excess mortality in women, suggesting The Current Study complementary mechanisms of reduced survival. These The main findings of the current study are: (1) women have outcomes raise several significant methodological and clinical lower procedural success as assessed by post-PCI TIMI 3 flow issues. regardless of time to hospital presentation; (2) time to hospital presentation correlates with 30-day mortality; (3) for late STEMI presentations, women have higher mortality rates Women Have Lower Procedural Success than men; (4) when women and men present for STEMI within We know that failure to achieve post-PCI TIMI 3 flow 120 minutes of symptom onset, their outcomes are equally correlates with mortality.38,39 The different twist of this study positive. These findings suggest the existence of an important is that it demonstrates that failure to achieve post-PCI TIMI 3 DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 7 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Table 4. IPTW: PCI Treatment Effect on 30-Day Mortality in “Flow Benefiters” Timely (<120-min) PCI Delayed (≥120-min) PCI Characteristics Women (N=143) Men (N=549) P Value Women (N=469) Men (N=1278) P Value Age, meanSD, y 61.711.1 57.911.0 0.0003 62.311.7 60.911.4 0.023 Cardiovascular risk factors, % Family history of CAD 33.6 35.0 0.754 34.4 31.7 0.285 Diabetes mellitus 20.1 16.1 0.256 24.8 23.2 0.485 Hypertension 69.2 59.5 0.033 69.7 66.6 0.220 Hypercholesterolemia 45.0 43.6 0.764 48.3 44.2 0.127 Current smoking 59.0 57.1 0.683 43.9 47.8 0.148 Former smoking 4.7 10.5 0.033 10.9 13.3 0.181 Previous cardiovascular disease, % Previous angina pectoris 21.0 17.3 0.306 21.1 19.8 0.548 Previous myocardial infarction 7.5 9.9 0.382 5.2 6.3 0.392 Previous PCI 0.3 6.3 0.126 1.5 3.2 0.052 Previous CABG 0 0.3 0.507 0 0.5 0.132 Peripheral artery disease 0.8 0.9 0.909 1.0 1.1 0.855 Previous heart failure 7.5 4.1 0.089 5.2 5.7 0.686 Previous stroke 3.6 1.7 0.157 3.3 2.7 0.505 Clinical presentation at admission ST-segment elevation in anterior leads, % 39.0 41.5 0.588 33.8 34.6 0.755 Killip class ≥2, % 15.1 13.0 0.512 19.4 19.8 0.852 Systolic blood pressure at baseline, meanSD, mm Hg 137.228.2 135.823.5 0.555 134.828.0 136.624.4 0.210 Heart rate at baseline, meanSD, beats/min 76.613.2 77.416.5 0.616 79.518.2 79.617.3 0.911 Serum creatinine at baseline, meanSD, lmol/L 73.319.2 89.833.8 <0.001 79.028.8 90.245.9 <0.001 Outcomes 30-d Mortality, % 1.1 1.4 0.732 3.9 2.0 0.029 30-d Mortality, OR (95% CI) 0.74 (0.13–4.26) 0.732 1.94 (1.06–3.57) 0.032 CABG indicates coronary artery bypass graft; CAD, coronary artery disease; IPTW, inverse probability of treatment weighting; OR, odds radio; PCI, percutaneous coronary intervention. is one of the keys to explain mortality difference between cannot be explained by established cardiovascular risk factors sexes. The reasons for these procedure-associated differences or whether patients received conventional treatment by in TIMI flow rates cannot be explained by the data of this primary PCI. registry and should be further investigated. We can just assume that suboptimal reperfusion after primary PCI is related to microvascular disorders resulting from coronary vasospasm Time to Hospital Presentation Correlates With secondary to endothelial injury and/or distal embolization,40 30-Day Mortality and women have a higher frequency of coronary plaque erosion Time to reperfusion is clearly linked with myocardial salvage, and microembolization that could result in more microvascular and it makes sense that more rapid reperfusion should impact dysfunction than men.41 The central point we must underline, outcome. Myocardial cell death starts soon after arterial however, is that women with STEMI have a worse outcome, and occlusion, and the myocardial salvage one can expect from this is likely attributable to a combination of factors. Indeed, reperfusion decreases sharply between 2 and 3 hours.42 After only 2.5% of women and 0.5% of men died with a post-PCI TIMI 120 minutes, the benefit with respect to myocardial salvage flow grade ≤2, whereas the overall mortality rates were much is of limited magnitude. Previous studies have documented a greater (ie, 5.9% and 2.3%, respectively). Thus, other mecha- strict relationship between total ischemic time and 1-year nisms for excess mortality in women are operating, which mortality in patients treated with primary PCI43: every 30- DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 8 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH minute delay before reperfusion, the relative risk of mortality that delay to hospital presentation and suboptimal postpro- increases by 7.5%. North American and European guidelines cedural coronary blood flow are independent mechanisms recommend primary PCI as the preferred reperfusion strategy responsible for sex differences in outcomes. As so, female sex over thrombolysis in patients with STEMI, provided it can be may be a biological variable.50 performed within 120 minutes from diagnosis.44,45 The American Heart Association established “Mission: Lifeline” to develop STEMI systems of care that would improve timely Female Sex as a Biological Variable access to PCI.46 However, despite the increased focus on Little attention has been paid to provide experimental designs delays in STEMI care, it is difficult to translate guideline for comparative assessment of sex-specific modifications in recommendations into clinical practice. The NCDR (National myocardial tissue and blood flow during myocardial infarction. Cardiovascular Data Registry) found that only 15% of patients Few explanations, therefore, may account for our findings. One are treated within 120 minutes from symptom onset.47 In the possibility is that the initial process of infarct healing after current study, we examined the prehospital and hospital STEMI is different among women and men. Female animals components of the total ischemic time: time from symptom show a greater number of macrophages in their ischemic onset to hospital presentation and door-to-balloon time. We areas.51 Macrophages facilitate inward remodeling in arterial found that the median time from symptom onset to hospital vessels, causing further reduction in blood flow with poorer presentation ranged from 280 minutes in women to 240 min- myocardial perfusion. Another possibility is that impaired utes in men and that the median door-to-balloon times of myocardial perfusion after PCI may be facilitated in women women and men were 40 and 38 minutes, respectively. Much because they have smaller coronary artery size. A report from more important, we found that the proportions of men and the Zwolle investigators52 demonstrated a significant associ- women with time from symptom onset to hospital presenta- ation between small vessel size (<3 mm) and poor perfusion. tion within 120 minutes were only 23.2% and 29.1%, respec- As a final point, susceptibility to ischemia injury may also be tively. Women and men with the longer delay time had the sex dependent.53 Perhaps reduced collateral flow in women higher risk of 30-day mortality compared with those with may account for more vulnerability to prolonged untreated shorter delay times. Given these data, we focused more on ischemia after coronary obstruction and more reperfusion time from symptom onset to hospital presentation rather than injury. This may lead to more complications, including a higher focusing on door-to-balloon time. Indeed, the median door-to- incidence of shock, ventricular septal rupture, and severe balloon times of our patients were generally within the mitral regurgitation. guideline-endorsed time frame of ≤90 minutes. Door-to- balloon time only measures the delay to reperfusion once patients reach the hospitals, which constitutes a smaller Testing Hypotheses by IPTW Models fraction of the total ischemic time. Women were more likely to Estimating the effect of treatments on sex-related responses is exceed reperfusion times than men in the prehospital phase one of the crucial problems in resolving gender/sex issues. of the total ischemic time. Indeed, treatment disparities may account for a portion of adverse outcomes, leaving uncertainty about the size of the problem and comparability of different series reported.28 Early Versus Late STEMI Presentation and Sex Matched-pairs designs are commonly implemented in the field Difference in Outcomes of medical research to eliminate confounding and improve The main concern was that sex is correlated with delays, and study efficiency. We matched patients using IPTW models. sex effect may have been masked by the delay effect. Patients IPTW is based on the propensity score and has specific were, therefore, stratified in 2 delay cohorts: those with time advantages in comparison with regression adjustment. Advan- of symptom onset to hospital presentation up to 120 minutes tages include the ability to present a balance of covariates, or later and those with time within 120 minutes. The cutoff of which removes most of the confounding, a property that would 120 minutes was based on previous work.4,14,27,33,48,49 The be expected under randomization,22 and the ability to control main finding of such analysis can be summarized as follows. for many covariates. Matching by IPTW reduces statistical For late STEMI presentations, women have higher mortality power to a lower extent than adjustments in the model.54,55 rates than men (OR, 2.03; 95% CI, 1.23–3.33). As the delay to hospital presentation decreased to <120 minutes, the car- diovascular risk of women became similar to that of men (OR, Multiple Mediators of Excess Mortality in Women 1.28; 95% CI, 0.35–4.69). Interestingly, women have lower We examined whether the combination of suboptimal post- procedural success, as assessed by post-PCI TIMI 3 flow, PCI blood flow and longer delays from symptom onset to regardless of time to hospital presentation, which suggests hospital presentation may increase the risk of death in women DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 9 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH compared with men, over and beyond the separate effect of the current study only reflect those of patients who reach the each single factor. We defined 4 clinical scenarios. Patients hospital. As an observational study, we cannot completely admitted to hospital within 120 minutes having post-PCI exclude residual confounding, even using IPTW analyses. On complete restoration of blood flow (scenario 1) were the most the other hands, one cannot randomize patients to sex. robust patients because they had the lowest mortality Because catheterization data were mandated as part of the (Table 4). Patients admitted to hospital later and having registry protocol only in centers with PCI facilities, patient suboptimal postprocedural flow (scenario 2) were the frailest selection may have introduced potential referral bias. How- subjects (Table S9). However, their outcomes did not differ so ever, the external validity of this study is confirmed by the much from those of patients presenting within 120 minutes proportion of patients with post-PCI TIMI flow grade 0 to 2 from their symptom onset, but still with post-PCI TIMI flow in the overall population, which is similar to previous grades 0 to 2 (scenario 3) (Table S8). Scenario 4 (Table 4) estimates.58 Furthermore, the demographics, characteristics, included patients presenting to hospital later than 120 minutes and 30-day mortality rates of patients in this study were but with post-PCI complete restoration of blood flow. In these similar to those reported by recent research.59 Finally, results patients, mortality rates were twice as high as those reported in may not be definitive without replication. scenario 1, but consistently less than those seen in patients with TIMI flow grades 0 to 2 (scenarios 2 and 3). These findings indicate that female sex is associated with multiple indepen- Conclusions dent pathways leading to excess mortality after primary PCI for To date, most studies focused only on female sex, whereas STEMI. It is remarkable that post-PCI TIMI flow grade 0 to 2 is little or no specific data on comparison between sexes are associated with a much higher risk of mortality compared with available.6,60,61 The current study is the first investigation that the risk for patients with normal post-PCI TIMI flow grade. This estimated and reported heterogeneity of treatment effects relationship persists regardless of time to hospital presenta- after primary PCI among women and men. We provided data tion, and its magnitude is much greater for women than men. disaggregated by sex and gave insights into how sex may be Suboptimal post-PCI blood flow accounts for a part of the relevant as a biological variable. Female sex has a distinct observed sex differences in outcomes. Further studies are cardiovascular blood flow reactivity after primary PCI, regard- required to clarify the mechanisms underlying the persisting less of time to hospital presentation. Furthermore, women are female disadvantage in patients with longer delays to hospital more vulnerable to prolonged untreated ischemia. For late presentation. This information is of foremost importance to STEMI presentations, women have a 2-fold likelihood of reduce the sex gap in mortality. mortality compared with men. These results provide motiva- tion for adequate consideration of both sexes in experiments Myocardial Infarction in Younger Versus Older and disaggregation of data by sex, which may allow for sex- Women based comparisons and may inform clinical interventions. Sex differences in mortality rates have been shown to be especially higher in the younger age group (<60 years).5,28–30,56 Disclosures Accordingly, we studied sex differences in post-PCI TIMI flow None. sorted by younger versus older age. Women had higher rates of post-PCI TIMI flow grade 0 to 2 than men only in patients aged ≥60 years. This is consistent with the hypothesis that the References association between female sex and excess mortality in STEMI is 1. Kunadian V, Qiu W, Lagerqvist B, Johnston N, Sinclair H, Tan Y, Ludman P, not attributable to only 1 factor (namely, post-PCI TIMI flow). An James S, Sarno G; National Institute for Cardiovascular Outcomes Research and Swedish Coronary Angiography and Angioplasty Registries. Gender important reason for the relatively higher mortality rates in differences in outcomes and predictors of all-cause mortality after percuta- neous coronary intervention (data from United Kingdom and Sweden). Am J younger women may be the occurrence of coronary spasm or Cardiol. 2017;119:210–216. dysfunction a few days after PCI.57 Diagnosis can be difficult, 2. Heer T, Hochadel M, Schmidt K, Mehilli J, Zahn R, Kuck KH, Hamm C, B€ ohm M, Ertl G, Hoffmeister HM, Sack S, Senges J, Massberg S, Gitt AK, Zeymer U. Sex given the transience of coronary spasm, and might require more differences in percutaneous coronary intervention: insights from the coronary sophisticated diagnostic approaches than those used in current angiography and PCI registry of the German Society of Cardiology. J Am Heart Assoc. 2017;6:e004972. DOI: 10.1161/JAHA.116.004972. practice, as reflected by our registry study. 3. Pancholy SB, Shantha GP, Patel T, Cheskin LJ. Sex differences in short-term and long-term all-cause mortality among patients with ST-segment elevation myocardial infarction treated by primary percutaneous intervention: a meta- analysis. JAMA Intern Med. 2014;174:1822–1830. Limitations 4. Bugiardini R, Ricci B, Cenko E, Vasiljevic Z, Kedev S, Davidovic G, Zdravkovic We cannot rule out that more men and women died before M, Milicic D, Dilic M, Manfrini O, Koller A, Badimon L. Delayed care and mortality among women and men with myocardial infarction. J Am Heart presentation to the hospitals; therefore, the mortality rates of Assoc. 2017;6:e005968. DOI: 10.1161/JAHA.117.005968. DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 10 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH 5. Cenko E, Yoon J, Kedev S, Stankovic G, Vasiljevic Z, Krljanac G, Kalpak O, Ricci acute myocardial infarction treated with thrombolytic therapy. JAMA. B, Milicic D, Manfrini O, van der Schaar M, Badimon L, Bugiardini R. Sex 1996;275:777–782. differences in outcomes after STEMI: effect modification by treatment strategy 27. De Luca L, Marini M, Gonzini L, Boccanelli A, Casella G, Chiarella F, De Servi S, and age. JAMA Intern Med. 2018;178:632–639. Di Chiara A, Di Pasquale G, Olivari Z, Caretta G, Lenatti L, Gulizia MM, 6. Bavishi C, Bangalore S, Patel D, Chatterjee S, Trivedi V, Tamis-Holland JE. Short Savonitto S. Contemporary trends and age-specific sex differences in and long-term mortality in women and men undergoing primary angioplasty: a management and outcome for patients with ST-segment elevation myocardial comprehensive meta-analysis. Int J Cardiol. 2015;198:123–130. infarction. J Am Heart Assoc. 2016;5:e004202. DOI: 10.1161/JAHA.116. 004202. 7. Mehta LS, Beckie TM, DeVon HA, Grines CL, Krumholz HM, Johnson MN, Lindley KJ, Vaccarino V, Wang TY, Watson KE, Wenger NK; American Heart 28. Vaccarino V, Parsons L, Every NR, Barron HV, Krumholz HM. Sex-based Association Cardiovascular Disease in Women and Special Populations differences in early mortality after myocardial infarction: National Registry of Committee of the Council on Clinical Cardiology, Council on Epidemiology Myocardial Infarction 2 Participants. N Engl J Med. 1999;341:217–225. and Prevention, Council on Cardiovascular and Stroke Nursing, and Council on 29. Lichtman JH, Wang Y, Jones SB, Leifheit-Limson EC, Shaw LJ, Vaccarino V, Quality of Care and Outcomes Research. Acute myocardial infarction in Rumsfeld JS, Krumholz HM, Curtis JP. Age and sex differences in inhospital women: a scientific statement from the American Heart Association. complication rates and mortality after percutaneous coronary intervention Circulation. 2016;133:916–947. procedures: evidence from the NCDR(). Am Heart J. 2014;167:376–383. 8. Merz CN, Kelsey SF, Pepine CJ, Reichek N, Reis SE, Rogers WJ, Sharaf BL, 30. Jneid H, Fonarow GC, Cannon CP, Hernandez AF, Palacios IF, Maree AO, Wells Sopko G. The Women’s Ischemia Syndrome Evaluation (WISE) study: protocol Q, Bozkurt B, Labresh KA, Liang L, Hong Y, Newby LK, Fletcher G, Peterson E, design, methodology and feasibility report. J Am Coll Cardiol. 1999;33:1453– Wexler L; Get With The Guidelines Steering Committee and Investigators. Sex 1461. differences in medical care and early death after acute myocardial infarction. 9. Frye RL, Gibbons RJ, Schaff HV, Vlietstra RE, Gersh BJ, Mock MB. Treatment of Circulation. 2008;118:2803–2810. coronary artery disease. J Am Coll Cardiol. 1989;13:957–968. 31. Bangalore S, Fonarow GC, Peterson ED, Hellkamp AS, Hernandez AF, Laskey 10. TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) trial: phase W, Peacock WF, Cannon CP, Schwamm LH, Bhatt DL; Get With The Guidelines I findings. N Engl J Med. 1985;312:932–936. Steering Committee and Investigators. Age and gender differences in quality of care and outcomes for patients with ST-segment elevation myocardial 11. Duvernoy CS, Smith DE, Manohar P, Schaefer A, Kline-Rogers E, Share D, infarction. Am J Med. 2012;125:1000–1009. McNamara R, Gurm HS, Moscucci M. Gender differences in adverse outcomes after contemporary percutaneous coronary intervention: an analysis from the 32. Jackson EA, Moscucci M, Smith DE, Share D, Dixon S, Greenbaum A, Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) Grossman PM, Gurm HS. The association of sex with outcomes among percutaneous coronary intervention registry. Am Heart J. 2010;159:677– patients undergoing primary percutaneous coronary intervention for ST 683.e1. elevation myocardial infarction in the contemporary era: insights from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2). Am 12. Bugiardini R, Bairey Merz CN. Angina with “normal” coronary arteries: a Heart J. 2011;161:106–112.e1. changing philosophy. JAMA. 2005;293:477–484. 33. De Luca G, Suryapranata H, Dambrink JH, Ottervanger JP, van ‘t Hof AW, 13. Kloner RA, Ganote CE, Jennings RB. The “no-reflow” phenomenon after Zijlstra F, Hoorntje JC, Gosselink AT, de Boer MJ. Sex-related differences in temporary coronary occlusion in the dog. J Clin Invest. 1974;54:1496–1508. outcome after ST-segment elevation myocardial infarction treated by primary 14. D’Onofrio G, Safdar B, Lichtman JH, Strait KM, Dreyer RP, Geda M, Spertus JA, angioplasty: data from the Zwolle Myocardial Infarction study. Am Heart J. Krumholz HM. Sex differences in reperfusion in young patients with ST- 2004;148:852–856. segment-elevation myocardial infarction: results from the VIRGO study. 34. Benamer H, Tafflet M, Bataille S, Escolano S, Livarek B, Fourchard V, Caussin Circulation. 2015;131:1324–1332. C, Teiger E, Garot P, Lambert Y, Jouven X, Spaulding C; CARDIO-ARHIF Registry 15. Bugiardini R, Badimon L; ISACS-TC Investigators and Coordinators. The Investigators. Female gender is an independent predictor of in-hospital International Survey of Acute Coronary Syndromes in Transitional Countries mortality after STEMI in the era of primary PCI: insights from the greater Paris (ISACS-TC): 2010–2015. Int J Cardiol. 2016;217:S1–S6. area PCI Registry. EuroIntervention. 2011;6:1073–1079. 16. Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, 35. Mehilli J, Kastrati A, Dirschinger J, Pache J, Seyfarth M, Blasini R, Hall D, Neumann FJ, Ardissino D, De Servi S, Murphy SA, Riesmeyer J, Weerakkody G, Neumann FJ, Sch€ omig A. Sex-based analysis of outcome in patients with acute Gibson CM, Antman EM; TRITON-TIMI 38 Investigators. Prasugrel versus myocardial infarction treated predominantly with percutaneous coronary clopidogrel in patients with acute coronary syndromes. N Engl J Med. intervention. JAMA. 2002;287:210–215. 2007;357:2001–2015. 36. de Boer SP, Roos-Hesselink JW, van Leeuwen MA, Lenzen MJ, van Geuns RJ, 17. Cenko E, Ricci B, Kedev S, Kalpak O, C^alm^ac L, Vasiljevic Z, Knezevic B, Dilic Regar E, van Mieghem NM, van Domburg R, Zijlstra F, Serruys PW, Boersma E. M, Milicic D, Manfrini O, Koller A, Dorobantu M, Badimon L, Bugiardini R. The Excess mortality in women compared to men after PCI in STEMI: an analysis of no-reflow phenomenon in the young and in the elderly. Int J Cardiol. 11,931 patients during 2000–2009. Int J Cardiol. 2014;176:456–463. 2016;222:1122–1128. 37. Gevaert SA, De Bacquer D, Evrard P, Convens C, Dubois P, Boland J, Renard M, 18. Raghunathan TE, Solenberger PW, Van Hoeyk J. IVEware: Imputation and Beauloye C, Coussement P, De Raedt H, de Meester A, Vandecasteele E, Variance Estimation Software. Ann Arbor, MI: Survey Research Center, Institute Vranckx P, Sinnaeve PR, Claeys MJ. Gender, TIMI risk score and in-hospital for Social Research University of Michigan; 2002. ftp://ftp.isr.umich.edu/ mortality in STEMI patients undergoing primary PCI: results from the Belgian pub/src/smp/ive/ive_user.pdf. Accessed August 4, 2017. STEMI registry. EuroIntervention. 2014;9:1095–1101. 19. Troyanskaya O, Cantor M, Sherlock G, Brown P, Hastie T, Tibshirani R, Botstein 38. Terkelsen CJ, Sørensen JT, Maeng M, Jensen LO, Tilsted HH, Trautner S, Vach D, Altman RB. Missing value estimation methods for DNA microarrays. W, Johnsen SP, Thuesen L, Lassen JF. System delay and mortality among Bioinformatics. 2001;17:520–525. patients with STEMI treated with primary percutaneous coronary intervention. JAMA. 2010;304:763–771. 20. Beretta L, Santaniello A. Nearest neighbor imputation algorithms: a critical evaluation. BMC Med Inform Decis Mak. 2016;16(suppl 3):74. 39. Mehta RH, Ou FS, Peterson ED, Shaw RE, Hillegass WBJr, Rumsfeld JS, Roe MT; American College of Cardiology-National Cardiovascular Database Registry 21. Kononenko I, Simec E, Robnik-Sikonja M. Overcoming the myopia of inductive Investigators. Clinical significance of post-procedural TIMI flow in patients with learning algorithms with RELIEFF. Appl Intell. 1997;7:39–55. cardiogenic shock undergoing primary percutaneous coronary intervention. 22. Austin PC, Stuart EA. Moving towards best practice when using inverse JACC Cardiovasc Interv. 2009;2:56–64. probability of treatment weighting (IPTW) using the propensity score to 40. Rezkalla SH, Kloner RA. No-reflow phenomenon. Circulation. 2002;105:656– estimate causal treatment effects in observational studies. Stat Med. 662. 2015;34:3661–3679. 41. Burke AP, Farb A, Malcom GT, Liang Y, Smialek J, Virmani R. Effect of risk 23. Mansournia MA, Etminan M, Danaei G, Kaufman JS, Collins G. Handling time factors on the mechanism of acute thrombosis and sudden coronary death in varying confounding in observational research. BMJ. 2017;359:j4587. women. Circulation. 1998;97:2110–2116. 24. Robins JM, Hernan MA, Brumback B. Marginal structural models and causal 42. Reimer KA, Lowe JE, Rasmussen MM, Jennings RB. The wavefront phe- inference in epidemiology. Epidemiology. 2000;11:550–560. nomenon of ischemic cell death, 1: myocardial infarct size vs duration of 25. Zhang Z, Fang J, Gillespie C, Wang G, Hong Y, Yoon PW. Age-specific gender coronary occlusion in dogs. Circulation. 1977;56:786–794. differences in in-hospital mortality by type of acute myocardial infarction. Am J 43. De Luca G, Suryapranata H, Ottervanger JP, Antman EM. Time delay to Cardiol. 2012;109:1097–1103. treatment and mortality in primary angioplasty for acute myocardial infarction: 26. Weaver WD, White HD, Wilcox RG, Aylward PE, Morris D, Guerci A, Ohman EM, every minute of delay counts. Circulation. 2004;109:1223–1225. Barbash GI, Betriu A, Sadowski Z, Topol EJ, Califf RM; ; GUSTO-I Investigators. 44. O’Gara PT, Kushner FG, Ascheim DD, Casey DE Jr, Chung MK, de Lemos JA, Comparisons of characteristics and outcomes among women and men with Ettinger SM, Fang JC, Fesmire FM, Franklin BA, Granger CB, Krumholz HM, DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 11 Sex, STEMI, and Post-PCI Flow Cenko et al ORIGINAL RESEARCH Linderbaum JA, Morrow DA, Newby LK, Ornato JP, Ou N, Radford MJ, Tamis- 52. De Luca G, Suryapranata H, de Boer MJ, Ottervanger JP, Hoorntje JC, Gosselink Holland JE, Tommaso CL, Tracy CM, Woo YJ, Zhao DX, Anderson JL, Jacobs AK, AT, Dambrink JH, van’t Hof AW. Impact of vessel size on distal embolization, Halperin JL, Albert NM, Brindis RG, Creager MA, DeMets D, Guyton RA, myocardial perfusion and clinical outcome in patients undergoing primary Hochman JS, Kovacs RJ, Kushner FG, Ohman EM, Stevenson WG, Yancy CW; angioplasty for ST-segment elevation myocardial infarction. J Thromb Throm- American College of Cardiology Foundation/American Heart Association Task bolysis. 2009;27:198–203. Force on Practice Guidelines. 2013 ACCF/AHA guideline for the management 53. Tamis JE, Thompson TD, Weaver WD, White HD, Van de Werf F, Aylward P, of ST-elevation myocardial infarction: a report of the American College of Topol EJ, Califf RM; Global Use of Strategies to Open Occluded Coronary Cardiology Foundation/American Heart Association Task Force on Practice Arteries in Acute Coronary Syndromes IIb Investigators. Sex, clinical Guidelines. Circulation. 2013;127:e362–e425. presentation, and outcome in patients with acute coronary syndromes. N 45. Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno H, Engl J Med. 1999;341:226–232. Caforio ALP, Crea F, Goudevenos JA, Halvorsen S, Hindricks G, Kastrati A, 54. Brookhart MA, Wyss R, Layton JB, St€urmer T. Propensity score methods for Lenzen MJ, Prescott E, Roffi M, Valgimigli M, Varenhorst C, Vranckx P, confounding control in nonexperimental research. Circ Cardiovasc Qual Widimsky P; ESC Scientific Document Group. 2017 ESC guidelines for the Outcomes. 2013;6:604–611. management of acute myocardial infarction in patients presenting with ST- segment elevation: the Task Force for the management of acute myocardial 55. Gutman R, Rubin DB. Estimation of causal effects of binary treatments in infarction in patients presenting with ST-segment elevation of the European unconfounded studies. Stat Med. 2015;34:3381–3398. Society of Cardiology (ESC). Eur Heart J. 2018;39:119–177. 56. Ricci B, Cenko E, Vasiljevic Z, Stankovic G, Kedev S, Kalpak O, Vavlukis M, 46. The American Heart Association. Mission: lifeline. http://www.heart.org/ Zdravkovic M, Hinic S, Milicic D, Manfrini O, Badimon L, Bugiardini R. Acute HEARTORG/Professional/MissionLifelineHomePage/Mission-Lifeline-Home- coronary syndrome: the risk to young women. J Am Heart Assoc. 2017;6: Page_UCM_305495_SubHomePage.jsp. Accessed July 18, 2018. e007519. DOI: 10.1161/JAHA.117.007519. 47. Chakrabarti A, Krumholz HM, Wang Y, Rumsfeld JS, Nallamothu BK; National 57. el-Tamimi H, Davies GJ, Hackett D, Sritara P, Bertrand O, Crea F, Maseri A. Cardiovascular Data Registry. Time-to-reperfusion in patients undergoing Abnormal vasomotor changes early after coronary angioplasty: a quantitative interhospital transfer for primary percutaneous coronary intervention in the arteriographic study of their time course. Circulation. 1991;84:1198–1202. U.S: an analysis of 2005 and 2006 data from the National Cardiovascular Data 58. De Luca G, Suryapranata H, Zijlstra F, van ‘t Hof AW, Hoorntje JC, Gosselink Registry. J Am Coll Cardiol. 2008;51:2442–2443. AT, Dambrink JH, de Boer MJ; ZWOLLE Myocardial Infarction Study Group. 48. Milavetz JJ, Giebel DW, Christian TF, Schwartz RS, Holmes DR Jr, Gibbons RJ. Symptom-onset-to-balloon time and mortality in patients with acute myocar- Time to therapy and salvage in myocardial infarction. J Am Coll Cardiol. dial infarction treated by primary angioplasty. J Am Coll Cardiol. 2003;42:991– 1998;31:1246–1251. 997. 49. Brodie BR, Webb J, Cox DA, Qureshi M, Kalynych A, Turco M, Schultheiss HP, 59. Berg J, Bj€ orck L, Nielsen S, Lappas G, Rosengren A. Sex differences in survival Dulas D, Rutherford B, Antoniucci D, Stuckey T, Krucoff M, Gibbons R, Lansky after myocardial infarction in Sweden, 1987–2010. Heart. 2017;103:1625– A, Na Y, Mehran R, Stone GW; EMERALD Investigators. Impact of time to 1630. treatment on myocardial reperfusion and infarct size with primary percuta- 60. Wenger NK. Women and coronary heart disease: a century after Herrick: neous coronary intervention for acute myocardial infarction (from the understudied, underdiagnosed, and undertreated. Circulation. 2012;126:604– EMERALD Trial). Am J Cardiol. 2007;99:1680–1686. 611. 50. NIH policy on sex as a biological variable. https://orwh.od.nih.gov/sex-gender/ 61. Vaccarino V, Sullivan S, Hammadah M, Wilmot K, Al Mheid I, Ramadan R, Elon nih-policy-sex-biological-variable. Accessed July 18, 2018. L, Pimple PM, Garcia EV, Nye J, Shah AJ, Alkhoder A, Levantsevych O, Gay H, 51. Cavasin MA, Tao Z, Menon S, Yang XP. Gender differences in cardiac function Obideen M, Huang M, Lewis TT, Bremner JD, Quyyumi AA, Raggi P. Mental during early remodeling after acute myocardial infarction in mice. Life Sci. stress-induced-myocardial ischemia in young patients with recent myocardial 2004;75:2181–2192. infarction: sex differences and mechanisms. Circulation. 2018;137:794–805. DOI: 10.1161/JAHA.118.011190 Journal of the American Heart Association 12 SUPPLEMENTAL MATERIAL Data S1. SUPPLEMENTAL METHODS Inverse Probability of Treatment Weighting Analysis We used Inverse Propensity of Treatment Weighting (IPTW) to balance the distribution of covariates between two patient groups. If e denotes the estimated propensity score (i.e. e=\hat{P}(Z=1 | x), where the patient x is included in patient group 1; then, 1-e = \hat{P}(Z=0 | x)), then the original sample is weighted by the following weights: Z/e+(1−Z)/ 1−e where Z represents the patient group. For instance, women (Z=1) are assigned a weight equal to the reciprocal of the propensity score (1/e), while men (Z=0) are assigned a weight equal to the reciprocal of one minus the propensity score (1/1-e). The weighting procedure for each sample balances the covariate distributions between two patient groups. Nearest neighbor imputation algorithms Nearest neighbor (NN) imputation algorithms are efficient methods to fill in missing data where each missing value on some records is replaced by a value obtained from related cases in the whole set of records. Thus, imputation for clinical features, whose missing rate exceeded 10%, was conducted using the average of measured values from k records (kNN) [1, 2] NN algorithms are similarity-based methods that rely on distance metrics and results may change in relation to the similarity measure used to evaluate the distance between recipients and donors. In our work, we used the following norm as metric to evaluate distance: (∑ni=1|xi−yi|p)1/p Before imputation of the recipient Xi, the full set with no missing data C(X) was filtered to select a subset of features relevant to the missing variable to be imputed (Xi_miss). To this end, C(X) was considered as a dataset in the context of a regression problem, where the variable with the missing data (Xmiss) was set as the class variable and the other q variables (X1, X2, …, Xq) as predictors. We also applied the RReliefF algorithm [3] The set was, therefore, filtered to select a subset Cs(X) ⊂ C(X) where (X1, X2, …, Xs) ⊂ (X1, X2, …, Xq) and s < q. In the present context, we set the number of neighbors for RReliefF equal to 10 and set s as 10 %, 20 % or 30 % of q. As C(X) is invariant to Xi, the filtering step was performed only once before the NN imputation step that, on the contrary was performed separately for each Xi. Table S1. Characteristics of patients undergoing primary PCI within 12 hours from symptoms onset and excluded from the analysis. Overall Women Men Characteristics population P-value N=856 N=2138 N=2994 Age, years 60.3 ± 11.3 63.9 ± 10.9 58.9 ± 11.2 <0.001 Median age, years (IQR) 60 (53 – 68) 64 (57 – 72) 59 (51 – 66) <0.001 Cardiovascular risk factors, n (%) Family history of CAD 1115 (37.2) 330 (38.6) 785 (36.7) 0.350 Diabetes 680 (22.7) 269 (31.4) 411 (19.2) <0.001 Hypertension 1952 (65.2) 635 (74.2) 1317 (61.6) <0.001 Hypercholesterolemia 1195 (39.9) 338 (39.5) 857 (40.1) 0.763 Current smoking 1237 (41.3) 261 (30.5) 976 (45.7) <0.001 Former smoking 59 (2.0) 8 (0.9) 51 (2.4) 0.002 Previous cardiovascular disease, n (%) Previous angina pectoris 279 (9.3) 104 (12.1) 175 (8.2) 0.002 Previous myocardial infarction 437 (14.6) 105 (12.3) 332 (15.5) 0.017 Previous PCI 649 (21.7) 190 (22.2) 459 (21.5) 0.664 Previous CABG 27 (0.9) 6 (0.7) 21 (1.0) 0.429 Peripheral artery disease 36 (1.2) 10 (1.2) 26 (1.2) 0.913 Previous heart failure 40 (1.3) 15 (1.8) 25 (1.2) 0.249 Previous stroke 118 (3.9) 48 (5.6) 70 (3.3) 0.008 Clinical presentation, n (%) ST-segment elevation in anterior leads 606 (20.2) 181 (21.1) 425 (19.9) 0.441 Killip Class > 2 263 (8.8) 92 (10.7) 171 (8.0) 0.023 Systolic BP at baseline, mmHg 145.3 ± 19.1 145.6 ± 18.5 145.1 ± 19.4 0.551 Heart rate at baseline, beats/min 80.3 ± 13.0 80.5 ± 12.8 80.3 ± 13.0 0.709 Serum creatinine at baseline, µmol/liter 86.8 ± 47.9 86.0 ± 42.5 87.1 ± 49.7 0.630 Data are presented as number (%). CABG, coronary-artery bypass graft; CAD, coronary artery disease; IQR, interquartile range, PCI, percutaneous coronary intervention; SD, standard deviation Table S2. General logistic regression and regression coefficients in the propensity score model in the overall study population of women versus men. 𝛃 SE T statistics p-value Constant term (α) -1.695 0.517 -3.281 0.0010 Age, decades 0.538 0.053 10.197 <0.0001 Cardiovascular risk factors Family history of CAD 0.390 0.108 3.610 0.0003 Diabetes -0.010 0.119 -0.088 0.9299 Hypertension 0.530 0.115 4.599 <0.0001 Hypercholesterolemia 0.262 0.103 2.551 0.0107 Current smoking -0.423 0.114 -3.701 0.0002 Former smoking -1.211 0.182 -6.640 <0.0001 Previous cardiovascular disease Previous angina pectoris 0.165 0.125 1.326 0.1848 Previous myocardial infarction 0.281 0.237 -1.184 0.2365 Previous PCI -0.570 0.375 -1.520 0.1285 Previous CABG 0.000 0.000 0.000 - Peripheral artery disease -0.332 0.524 -0.614 0.5392 Previous heart failure -0.361 0.232 -1.555 0.1200 Previous stroke -0.025 0.323 -0.077 0.9384 Clinical presentation ST-segment elevation in anterior leads -0.315 0.107 -2.955 0.0031 Killip Class> 2 0.403 0.131 3.071 0.0021 Systolic blood pressure at baseline, mmHg -0.007 0.002 -3.477 0.0005 Heart rate at baseline, beats/min -0.001 0.003 -0.253 0.8002 Serum creatinine at baseline, µmol/liter -0.022 0.002 -9.686 <0.0001 Optimized regression coefficient (ß) and constant term(α) for the logistic regression. CABG, coronary artery bypass graft; CAD, coronary artery disease; PCI, percutaneous coronary intervention Table S3. Angiographic and Procedural Characteristics. Women Men P-value (n=673) (n=1923) Multivessel disease, n (%) 305 (45.3) 857 (44.6) 0.735 Acute vessel closure, n (%) 3 (0.4) 5 (0.3) 0.510 Pre-procedural TIMI flow grade 3, n (%) 174 (25.9) 601 (31.3) 0.007 Bifurcation lesion, n (%) 11 (1.6) 40 (2.1) 0.448 Acute thrombosis, n (%) 5 (0.7) 10 (0.5) 0.547 Ischemic stroke, n (%) 3 (0.4) 4 (0.2) 0.391 Minor bleeding, n (%) 7 (1.0) 18 (0.9) 0.817 Major bleeding, n (%) 3 (0.4) 7 (0.4) 0.779 TIMI, thrombolysis In Myocardial Infarction Table S4. TIMI flow grade ≤ 1: inverse probability of treatment weighting: women versus men. Women Men Characteristics P value (n=673) (n=1923) Age, years 61.8 ± 11.6 60.3 ± 11.4 0.003 Cardiovascular risk factors, % Family history of CAD 34.9 31.9 0.153 Diabetes 22.5 21.2 0.480 Hypertension 68.4 65.0 0.109 Hypercholesterolemia 46.6 44.0 0.243 Current smoking 47.9 50.2 0.304 Former smoking 9.5 12.6 0.031 Previous cardiovascular disease, % Previous angina pectoris 20.7 19.4 0.466 Previous myocardial infarction 5.6 7.1 0.180 Previous PCI 1.9 3.9 0.013 Previous CABG 0 0.4 0.112 Peripheral artery disease 0.7 1.2 0.286 Previous heart failure 5.2 5.1 0.919 Previous stroke 3.0 2.4 0.398 Clinical presentation at admission, % ST-segment elevation in anterior leads 37.4 37.7 0.890 Killip Class ≥2 19.2 19.3 0.954 Systolic blood pressure at baseline, mmHg 134.8 ± 27.6 136.0 ± 24.3 0.324 Heart rate at baseline, beats/min 79.0 ± 17.2 79.2 ± 17.4 0.809 Serum creatinine at baseline, µmol/liter 77.8 ± 28.2 91.7 ± 46.4 <0.0001 Outcomes TIMI flow grade ≤ 1, % 5.8 3.9 0.0397 TIMI flow grade ≤ 1, OR (95% CI) 1.51 (1.02 – 2.26) 0.0411 Values are %, mean ± SD or odd ratio (95% confidence intervals) CABG, denotes coronary-artery bypass graft; CAD, coronary artery disease; CI confidence interval; OR, odds radio; PCI, percutaneous coronary intervention; SD, standard deviation; TIMI, Thrombolysis In Myocardial Infarction Table S5. Multivariate analysis of factors associated with post PCI TIMI flow grade ≤2. Characteristics OR 95% CI p-value Women 1.68 1.15 – 2.44 0.003 Age, decades 1.33 1.12 – 1.58 0.0006 Cardiovascular risk factors Family history of CAD 0.73 0.49 – 1.09 0.061 Diabetes 0.90 0.60 – 1.37 0.317 Hypertension 1.05 0.70 – 1.57 0.413 Hypercholesterolemia 0.97 0.68 – 1.39 0.441 Current smoking 1.34 0.88 – 2.05 0.083 Former smoking 1.48 0.87 – 2.54 0.076 Previous cardiovascular disease Previous angina pectoris 1.07 0.70 – 1.64 0.373 Previous myocardial infarction 0.82 0.36 – 1.86 0.320 Previous PCI 0.34 0.07 – 1.54 0.080 Previous CABG 0.00 0.00 – Inf - Peripheral artery disease 5.07 1.98- 12.99 0.0004 Previous heart failure 0.50 0.21 – 1.15 0.051 Previous stroke 0.81 0.29 – 2.31 0.348 Clinical presentation ST-segment elevation in anterior leads 1.88 1.32 – 2.69 0.0003 Killip Class ≥ 2 2.28 1.55 – 3.34 <0.001 Systolic blood pressure at baseline, mmHg 0.99 0.99 – 1.00 0.054 Heart rate at baseline, beats/min 1.01 1.00 – 1.01 0.0002 Serum creatinine at baseline, µmol/liter 1.00 1.00 – 1.01 0.0002 CABG, coronary-artery bypass graft; CAD, coronary artery disease; CI, confidence interval; OR, odds radio; PCI, percutaneous coronary intervention; TIMI, thrombolysis In Myocardial Infarction Table S6. Multivariate analysis of factors associated with 30-day mortality. Characteristics OR 95% CI p-value Women 1.72 1.02 – 2.90 0.022 Age, decades 1.82 1.41 – 2.36 <0.001 Cardiovascular risk factors Family history of CAD 1.33 0.76 – 2.32 0.161 Diabetes 1.76 1.04 – 2.98 0.018 Hypertension 1.51 0.82 – 2.77 0.092 Hypercholesterolemia 0.51 0.29 – 0.88 0.008 Current smoking 1.17 0.64 – 2.14 0.306 Former smoking 0.47 0.17 – 1.31 0.073 Previous cardiovascular disease Previous angina pectoris 1.05 0.58 – 1.90 0.434 Previous myocardial infarction 0.91 0.35 – 2.37 0.424 Previous PCI 2.48 0.72 – 8.52 0.074 Previous CABG 0.00 0.00 - Inf - Peripheral artery disease 1.27 0.24 – 6.76 0.389 Previous heart failure 0.78 0.33 – 1.83 0.282 Previous stroke 2.67 0.95 – 7.45 0.031 Clinical presentation ST-segment elevation in anterior leads 0.89 0.53 – 1.51 0.335 Killip Class ≥ 2 5.19 3.03 – 8.88 <0.001 Systolic blood pressure at baseline, mmHg 0.97 0.96 – 0.98 <0.001 Heart rate at baseline, beats/min 1.01 1.00 – 1.02 0.021 Serum creatinine at baseline, µmol/liter 1.00 1.00 – 1.01 <0.001 CABG, coronary-artery bypass graft; CAD, coronary artery disease; CI, confidence intervals; OR, odds radio; PCI, percutaneous coronary intervention Table S7. Delay to reperfusion for STEMI sorted by sex. Women Men Characteristics P-value N=673 N=1923 Median time from symptoms onset to admission (IQR), 280 (170 - 498) 240 (145 - 430) 0.383 min Time from symptoms onset to admission<120 minutes, n 156 (23.2) 560 (29.1) 0.002 (%) Median time from door to balloon (IQR), min 40 (20 – 75) 38 (20 – 75) 0.677 Data are presented as median (IQR) or number (%) IQR denotes interquartile range Table S8. Inverse probability of treatment weighting: timely (<120‐minutes) PCI treatment effect on 30- day mortality in in patients with TIMI flow grade 0-2. Women Men Characteristics p-value (N=13) (N=11) Age, years 70.5 ± 16.3 63.3 ± 12.0 0.223 Cardiovascular risk factors, % Family history of CAD 38.5 18.2 0.287 Diabetes 7.7 27.3 0.240 Hypertension 69.2 72.7 0.858 Hypercholesterolemia 46.2 36.4 0.645 Current smoking 30.8 63.6 0.119 Former smoking 7.7 18.2 0.478 Previous cardiovascular disease, % Previous angina pectoris 30.8 18.2 0.493 Previous myocardial infarction 0 0 - Previous PCI 7.7 0 0.337 Previous CABG 0 0 - Peripheral artery disease 0 0 - Previous heart failure 7.7 0 0.337 Previous stroke 7.7 0 0.337 Clinical presentation ST-segment elevation in anterior leads, % 61.5 90.9 0.094 Killip Class ≥ 2, % 53.8 9.1 0.016 Systolic blood pressure at baseline, mmHg 130.4 ± 16.1 139.0 ± 32.7 0.439 Heart rate at baseline, beats/min 87.3 ± 17.4 90.5 ± 21.9 0.696 Serum creatinine at baseline, µmol/liter 91.2 ± 52.3 134.9 ± 88.0 0.169 Outcomes Primary outcome: 30-day mortality, % 30.8 9.1 0.194 Odd ratio (95% CI) 4.19 (0.33 – 239.96) 0.327 Data are presented as percentages (%). CABG, coronary-artery bypass graft; CAD coronary artery disease; CI confidence intervals; OR odds radio; PCI percutaneous coronary intervention; SD, standard deviation Table S9. Inverse probability of treatment weighting: delayed (>120‐minutes) treatment effects on 30-day mortality in in patients with TIMI flow grade 0-2. Women Men Characteristics p-value (N=48) (N=85) Age, years 69.1 ± 10.5 62.9 ± 10.9 0.001 Cardiovascular risk factors, % Family history of CAD 20.8 23.5 0.720 Diabetes 25.0 23.5 0.851 Hypertension 77.1 68.2 0.269 Hypercholesterolemia 45.8 44.7 0.901 Current smoking 25.0 50.6 0.003 Former smoking 8.3 18.8 0.076 Previous cardiovascular disease, % Previous angina pectoris 22.9 22.4 0.941 Previous myocardial infarction 4.2 7.1 0.475 Previous PCI 0 1.2 0.320 Previous CABG 0 0.4 - Peripheral artery disease 4.2 7.1 0.475 Previous heart failure 6.2 3.5 0.505 Previous stroke 4.2 2.4 0.590 Clinical presentation at admission ST-segment elevation in anterior leads, % 47.9 52.9 0.582 Killip Class ≥ 2, % 58.3 35.3 0.011 Systolic blood pressure at baseline, mmHg 130.0 ± 31.3 132.3 ± 24.2 0.666 Heart rate at baseline, beats/min 85.9 ± 21.2 80.7 ± 18.7 0.161 Serum creatinine at baseline, µmol/liter 112.9 ± 112.1 115.6 ± 105.5 0.892 Outcomes Primary outcome: 30-day mortality, % 27.1 12.9 0.061 Odds Ratio (95% CI) 2.48 (0.92 – 6.80) 0.059 Data are presented as percentages (%).CABG, coronary-artery bypass grafting; CAD coronary artery disease; CI confidence intervals; OR odds radio; PCI percutaneous coronary intervention; SD, standard deviation Table S10. Inverse probability of treatment weighting: women versus men in patients aged 60 year or over. Women Men Characteristics P value (n=462) (n=878) Age, years 69.4 ± 7.5 69.4 ± 7.4 0.908 Cardiovascular risk factors, % Family history of CAD 22.4 22.1 0.900 Diabetes 25.3 25.9 0.811 Hypertension 76.6 74.4 0.376 Hypercholesterolemia 43.3 43.2 0.972 Current smoking 31.7 33.0 0.553 Former smoking 12.3 14.0 0.385 Previous cardiovascular disease, % Previous angina pectoris 21.9 22.7 0.739 Previous myocardial infarction 5.0 7.6 0.071 Previous PCI 2.0 3.9 0.062 Previous CABG 0 0.8 0.047 Peripheral artery disease 1.8 1.6 0.785 Previous heart failure 5.9 6.7 0.571 Previous stroke 3.7 3.3 0.702 Clinical presentation at admission, % ST-segment elevation in anterior leads 38.7 37.2 0.591 Killip Class ≥2 17.0 16.7 0.889 Systolic blood pressure at baseline, mmHg 136.4 ± 29.2 136.5 ± 25.3 0.936 Heart rate at baseline, beats/min 79.5 ± 18.0 79.4 ± 19.2 0.901 Serum creatinine at baseline, µmol/liter 84.5 ± 25.6 95.2 ± 48.5 <0.0001 Outcomes TIMI flow grade ≤ 2, % 12.1 8.1 0.0179 TIMI flow grade ≤2, OR (95% CI) 1.56 (1.08 – 2.26) 0.0188 Values are %, mean ± SD or odd ratio (95% confidence intervals) CABG, denotes coronary-artery bypass graft; CAD, coronary artery disease; CI confidence interval; OR, odds radio; PCI, percutaneous coronary intervention; SD, standard deviation; TIMI, Thrombolysis In Myocardial Infarction Table S11. Inverse probability of treatment weighting: women versus men in patients aged < 60 year. Women Men Characteristics P value (n=211) (n=1045) Age, years 51.1 ± 6.1 50.9 ± 6.4 0.563 Cardiovascular risk factors, % Family history of CAD 45.3 40.4 0.187 Diabetes 18.0 16.5 0.595 Hypertension 53.2 55.6 0.523 Hypercholesterolemia 45.2 43.9 0.729 Current smoking 70.5 67.1 0.336 Former smoking 7.3 11.3 0.085 Previous cardiovascular disease, % Previous angina pectoris 15.3 16.1 0.773 Previous myocardial infarction 6.4 6.4 1.000 Previous PCI 3.0 3.8 0.574 Previous CABG 0 0.1 0.608 Peripheral artery disease 0 1.0 0.137 Previous heart failure 3.9 3.4 0.719 Previous stroke 2.1 1.5 0.524 Clinical presentation at admission, % ST-segment elevation in anterior leads 36.3 38.0 0.642 Killip Class ≥2 7.6 7.8 0.921 Systolic blood pressure at baseline, mmHg 134.7 ± 24.2 135.3 ± 23.5 0.741 Heart rate at baseline, beats/min 78.3 ± 15.2 79.3 ± 16.1 0.397 Serum creatinine at baseline, µmol/liter 67.0 ± 14.2 86.9 ± 42.8 <0.0001 Outcomes TIMI flow grade ≤ 2, % 2.1 3.7 0.2459 TIMI flow grade ≤2, OR (95% CI) 0.56 (0.21 – 1.51) 0.2522 Values are %, mean ± SD or odd ratio (95% confidence intervals) CABG, denotes coronary-artery bypass graft; CAD, coronary artery disease; CI confidence interval; OR, odds radio; PCI, percutaneous coronary intervention; SD, standard deviation; TIMI, Thrombolysis In Myocardial Infarction Figure S1. 30-day mortality in early presenters (within 2 hours from symptom onset to admission) and late presenters (≥120 minutes) from symptoms onset to admission) in women versus men. Figure S2. Delay to treatment and mortality in patients with TIMI flow grade 0-2. Supplemental References: 1. Troyanskaya O, Cantor M, Sherlock G, Brown P, Hastie T, Tibshirani R, Botstein D, Altman RB. Missing value estimation methods for DNA microarrays. Bioinformatics. 2001;17:520-5. 2. Beretta L, Santaniello A. Nearest neighbor imputation algorithms: a critical evaluation. BMC Med Inform Decis Mak. 2016;16 Suppl 3:74. 3. Kononenko I, Simec E, Robnik-Sikonja M. Overcoming the myopia of inductive learning algorithms with RELIEFF. Appl Intell. 1997;7:39-55.