Previous studies investigating the influence of gender on ST-segment elevation myocardial infarction have reported conflicting results. The aim of this study was to assess the influence of gender on ischemic times and outcomes after ST-segment elevation myocardial infarction in patients treated with primary percutaneous coronary intervention in modern practice. The present multicenter registry included consecutive patients with ST-segment elevation myocardial infarctions treated with primary percutaneous coronary intervention at 3 hospitals. Adjusted mortality rates were calculated using Cox proportional-hazards analyses. In total, 3,483 patients were included, of whom 868 were women (25%). Women were older, had a higher risk factor burden, and more frequently had histories of malignancy. Men more often had cardiac histories and peripheral vascular disease. Ischemic times were longer in women (median 192 minutes [interquartile range 141 to 286] vs 175 minutes [interquartile range 128 to 279] in men, p = 0.002). However, multivariate linear regression showed that this was due to age and co-morbidity. All-cause mortality was higher at 7 days (6.0% in women vs 3.0% in men, p <0.001) and at 1 year (9.9% in women vs 6.6% in men, p = 0.001). After adjustment, female gender predicted 7 day all-cause mortality (hazard ratio 1.61, 95% confidence interval 1.06 to 2.46) and cardiac mortality (hazard ratio 1.58, 95% confidence interval 1.03 to 2.42) but not 1-year mortality. Moreover, gender was an independent effect modifier for cardiogenic shock, leading to substantially worse outcomes in women. In conclusion, ischemic times remain longer in women because of age and co-morbidity. Female gender independently predicted early all-cause and cardiac mortality after primary percutaneous coronary intervention, and a strong interaction between gender and cardiogenic shock was observed.
Coronary artery disease (CAD) is the leading cause of death in men and women in the Western world. In 2003, the total cost of CAD in the European Union was an estimated €45 billion, with approximately 1 million working years lost because of CAD mortality. To reduce the huge burden of CAD, research aimed at the optimal recognition and treatment of CAD in men and women is essential. Nonetheless, women are underrepresented in clinical trials of CAD. It is known that risk factors for CAD bear different weight between men and women and that women with anginal symptoms more frequently show nonobstructive CAD on coronary angiography compared to men. Moreover, women have consistently shown higher mortality rates after acute coronary syndromes, because of older age and more co-morbidities. However, in the specific ST-segment elevation myocardial infarction (STEMI) population, conflicting results have been reported. Furthermore, previous studies lacked guideline-recommended treatment with primary percutaneous coronary intervention (PCI) or excluded high-risk patients commonly encountered in daily practice. Our goal was to investigate the influence of gender on ischemic times and outcomes after STEMI in patients treated with primary PCI in modern practice.
The present Dutch multicenter registry prospectively included consecutive patients treated with primary PCI for STEMI at 3 tertiary centers in The Netherlands. Two high-volume centers in the north of the Netherlands, Medical Center Alkmaar and Medical Center Leeuwarden, provide 24/7 cardiac care for areas of 450,000 and 650,000 inhabitants, respectively. The Leiden University Medical Center serves an area of approximately 750,000 inhabitants.
Consecutive patients who underwent primary PCI for STEMI from January 2006 to December 2009 were included in the analysis. Interventions included passing of a guidewire through a thrombus, thrombosuction, and/or percutaneous coronary balloon angioplasty with or without the placement of coronary stents. In case of out-of-hospital cardiac arrest, only patients with return of spontaneous circulation at the moment of arrival at the catheterization laboratory were included. Furthermore, patients permanently living outside The Netherlands were excluded to make follow-up through municipality records possible. STEMI was defined as symptoms of angina lasting >30 minutes along with typical electrocardiographic changes (ST-segment elevation ≥0.2 mV in ≥2 contiguous leads, V 1 through V 3 , or ≥0.1 mV in other leads or presumed new left bundle branch block).
Prehospital protocols included triage by 12-lead electrocardiographic findings in the field, faxed to the operator on call, and in-ambulance treatment with aspirin, intravenous heparin bolus, and a loading dose of clopidogrel. Glycoprotein IIb/IIIa inhibitors were administered frequently, using up-front administration at the Leiden University Medical Center and periprocedural administration at the other hospitals. On arrival at the hospital, patients were transferred as soon as possible to the catheterization laboratory. Procedures were performed according to current clinical guidelines. Patients treated at the Leiden University Medical Center were treated according to the institutional MISSION! protocol, a standardized prehospital, in-hospital, and outpatient clinical framework for decision making and treatment. After hospital discharge, these patients were intensively monitored and managed at the outpatient clinic for 1 year, after which they were referred back to their general practitioners or referred to regular, generally regional, cardiologic outpatient clinics. At the other centers, local residents were managed at the outpatient clinics, and patients referred from regional hospitals were referred back for further management after primary PCI by regional cardiologists.
Hospitals prospectively included patients treated with primary PCI, registering baseline and procedural data. Definitions of variables were synchronized among centers. Cardiogenic shock was defined as systolic blood pressure <90 mm Hg with signs of tissue hypoperfusion requiring treatment in form of resuscitation, inotropic agents, or assistive devices. Symptom-to-balloon time was defined as the time between the onset of symptoms and balloon inflation. Diagnosis-to-balloon time was defined as the time between the first diagnostic electrocardiographic assessment, mostly the ambulance triage studies, and balloon inflation. Door-to-balloon time was the time between patient arrival at the tertiary hospital and balloon inflation. Close cooperation with regional emergency medical system providers supplied prehospital times. Vital status was obtained using municipality records.
The 3 databases were pooled into a patient-level database, and stratification was done according to gender. Continuous variables are presented as mean ± SD or as medians with interquartile ranges and were compared using Student’s t test for means and nonparametric tests for medians. Categorical variables are expressed as counts and percentages and were compared using Pearson’s chi-square test. All statistical tests were 2 tailed, and p values <0.05 were considered statistically significant. Time to end point was analyzed using Kaplan-Meier plots, and the log-rank test was applied to compare cumulative incidences of the end point between groups.
Linear regression models were used to analyze variables predictive of log-transformed treatment delay. Univariate predictors of delay were added into multivariate linear regression models using a cut-off p value of 0.10. To evaluate the effect of gender as an independent predictor of mortality, multivariate Cox proportional-hazards models were constructed using a forward stepwise method. A cut-off p value of <0.10 was applied to enter significant univariate predictors of outcome into the multivariate models. Gender was forced to stay in the multivariate models to allow the calculation of adjusted hazard ratios (HRs) for all outcomes.
During the inclusion period, 3,483 consecutive patients with STEMI were treated with primary PCI, of whom 868 were women (24.9%) and 2,615 were men (75.1%). Baseline characteristics ( Table 1 ) showed that women were on average older and had a higher risk factor burden, with a higher prevalence of insulin-dependent diabetes mellitus and hypertension. Furthermore, women more often had histories of malignancy. In contrast, men had had previous myocardial infarctions more frequently, and a larger proportion of men had undergone PCI or bypass surgery. Additionally, peripheral vascular disease was more common in men, and their median peak creatine kinase level was higher compared to that of women.
|Variable||Men (n = 2,615)||Women (n = 868)||p Value|
|Age (yrs)||61.8 ± 11.9||67.6 ± 13.1||<0.001|
|Body mass index (kg/m 2 )||26.6 ± 3.7||26.3 ± 4.8||0.110|
|Non-insulin-dependent diabetes mellitus||214 (8.3%)||80 (9.3%)||0.346|
|Insulin-dependent diabetes mellitus||50 (1.9%)||42 (4.9%)||<0.001|
|Hypertension ∗||841 (32.5%)||394 (45.9%)||<0.001|
|Hypercholesterolemia †||608 (23.6%)||187 (21.8%)||0.282|
|Family history of cardiovascular disease ‡||994 (40.2%)||335 (41.2%)||0.646|
|Current smoking||1,222 (47.8%)||344 (40.6%)||0.001|
|Number of risk factors||1.52 ± 1.05||1.60 ± 1.10||0.036|
|Previous myocardial infarction||314 (12.1%)||61 (7.1%)||<0.001|
|Previous PCI||238 (9.2%)||52 (6.0%)||0.004|
|Previous coronary artery bypass grafting||76 (2.9%)||9 (1.0%)||0.002|
|Previous peripheral vascular disease||136 (5.3%)||28 (3.3%)||0.016|
|Previous cerebrovascular disease||157 (6.1%)||59 (6.9%)||0.418|
|Previous malignancy||142 (5.5%)||68 (8.0%)||0.011|
|Previous renal insufficiency §||86 (3.3%)||41 (4.8%)||0.053|
|Anemia on admission, moderate to severe ||||41 (1.6%)||14 (1.6%)||0.911|
|Out-of-hospital cardiac arrest||176 (6.7%)||48 (5.5%)||0.212|
|Cardiogenic shock||162 (6.2%)||65 (7.5%)||0.181|
|Intra-aortic balloon pump placement||144 (4.4%)||35 (4.0%)||0.680|
|Creatine phosphokinase peak (U/L)||1,420 (649–2,715)||1,170 (566–2,335)||<0.001|
|Symptom-to-balloon time (minutes)||175 (128–279)||192 (141–286)||0.002|
|Diagnosis-to-balloon time (minutes)||78 (64–99)||81 (66–101)||0.037|
|Door-to-balloon time (minutes)||46 (33–67)||46 (33–68)||0.405|
Time between the onset of symptoms and balloon inflation (ischemic time) was significantly longer in women ( Table 1 ). In addition, time between first diagnosis of STEMI and balloon inflation was marginally longer. Multivariate linear regression analysis of log-transformed ischemic time revealed that age per 10-year increase (β = 0.03, 95% confidence interval [CI] 0.01 to 0.05, p = 0.001), history of diabetes mellitus (β = 0.10, 95% CI 0.03 to 0.18, p = 0.006), and history of renal insufficiency (β = 0.15, 95% CI 0.02 to 0.27, p = 0.020) were independent predictors of longer ischemic time, whereas gender was not (β = 0.03, 95% CI −0.03 to 0.08, p = 0.295).
Procedurally, a higher percentage of women presented with the left anterior descending coronary artery as the culprit vessel, balanced by a lower percentage with the circumflex coronary artery as the culprit vessel ( Table 2 ). Abciximab treatment was more common in men, and Thrombolysis In Myocardial Infarction (TIMI) flow before and after the procedure was similar between men and women. Beta-blocking agents were more frequently prescribed in men; other medications were balanced between the genders.
|Variable||Men (n = 2,615)||Women (n = 868)||p Value|
|Coronary culprit vessel||0.049|
|Left anterior descending coronary artery||1,026 (39.3%)||381 (43.9%)||0.016|
|Left circumflex coronary artery||435 (16.6%)||118 (13.6%)||0.033|
|Right coronary artery||1,082 (41.4%)||352 (40.6%)||0.657|
|Left main coronary artery||36 (1.4%)||11 (1.3%)||0.807|
|Bypass graft||34 (1.3%)||6 (0.7%)||0.144|
|Number of vessels narrowed >50%||0.152|
|1||1,204 (46.1%)||433 (49.9%)|
|2||837 (32.0%)||257 (29.6%)|
|3||571 (21.9%)||178 (20.5%)|
|Stenting||2,503 (95.8%)||828 (95.4%)||0.616|
|Drug-eluting stents||1,806 (72.6%)||603 (73.6%)||0.584|
|Bare-metal stents||688 (27.7%)||221 (27.0%)||0.701|
|Multivessel intervention||282 (10.8%)||97 (11.2%)||0.743|
|Abciximab treatment||1,964 (75.9%)||596 (69.2%)||<0.001|
|Preprocedural TIMI flow grade||0.078|
|0||1,809 (69.3%)||566 (65.2%)|
|1||274 (10.5%)||112 (12.9%)|
|2||298 (11.4%)||100 (11.5%)|
|3||229 (8.8%)||90 (10.4%)|
|Postprocedural TIMI flow grade ≥2||2,558 (98.0%)||846 (97.6%)||0.488|
|Admission duration (days)||3.7 ± 6.2||4.0 ± 6.9||0.415|
|Aspirin/warfarin derivative||2,502 (99.2%)||808 (99.1%)||0.766|
|Clopidogrel||2,472 (98.1%)||798 (97.9%)||0.745|
|β blocker||2,277 (90.7%)||714 (88.0%)||0.029|
|Angiotensin-converting enzyme inhibitor/angiotensin II antagonist||1,791 (71.4%)||578 (71.3%)||0.938|
|Statin||2,345 (93.4%)||743 (91.6%)||0.093|
One-year survival status was known in 3,479 patients. All-cause and cardiac mortality were more common in women compared to men during the entire follow-up period ( Table 3 ). Landmark analysis, with a cut-off point at 7 days ( Figure 1 ), showed that this was due to higher early mortality, with similar prognoses for men and women after this period.
|Variable||Men (n = 2,615)||Women (n = 868)||p Value|
|7 days||78 (3.0%)||52 (6.0%)||<0.001|
|Unadjusted HR (95% CI)||0.49 (0.35–0.70)||2.04 (1.43–2.89)||<0.001|
|Adjusted HR (95% CI)||0.62 (0.41–0.95)||1.61 (1.06–2.46)||0.027|
|1 year||173 (6.6%)||86 (9.9%)||0.001|
|Unadjusted HR (95% CI)||0.65 (0.50–0.84)||1.54 (1.19–1.99)||0.001|
|Adjusted HR (95% CI)||0.98 (0.73–1.32)||1.02 (0.76–1.37)||0.900|
|7 days||77 (2.9%)||50 (5.8%)||<0.001|
|Unadjusted HR (95% CI)||0.50 (0.35–0.72)||1.98 (1.39–2.83)||<0.001|
|Adjusted HR (95% CI)||0.63 (0.41–0.97)||1.58 (1.03–2.42)||0.037|
|1 year||132 (5.1%)||75 (8.7%)||<0.001|
|Unadjusted HR (95% CI)||0.57 (0.43–0.76)||1.75 (1.32–2.32)||<0.001|
|Adjusted HR (95% CI)||0.79 (0.57–1.10)||1.26 (0.91–1.75)||0.168|