The aim of this thesis was to evaluate baseline characteristics, management and outcome in real life ST-elevation myocardial infarction [STEMI] cohorts from a gender perspective. We aimed to evaluate the total STEMI population as well as certain subgroups, such as the youngest. Moreover we aimed to analyse gender differences in renal function, and the prognostic impact of reduced renal function in men and women with STEMI.
In Paper I all STEMI patients registered in RIKS-HIA between 1st Jan 1995 and 31st Dec 2006 were included, in total 54 146 patients, 35% women. Women were 7 years older than men, with 30 min longer median symptomto-door time. They had higher prevalence of co-morbidities such as diabetes, hypertension and heart failure whereas men were more often smokers, had a previous myocardial infarction [MI] or were previously revascularised. During hospital care, fewer women than men, 63% vs. 72%, p<0.001, received acute reperfusion therapy, odds ratio [OR] 0.83 (95% confidence interval [CI] 0.79 – 0.88) after multivariable adjustment. Inhospital mortality was 13% vs. 7%, women vs. men, p<0.001. After multivariable adjustments women had 22% higher risk of in-hospital death, OR 1.22 (95% CI 1.11 – 1.33). Adding reperfusion therapy to the adjustment model did not change the odds of death, OR 1.21 (1.11 – 1.32). Stratifying the cohort into four age-groups revealed increased mortality with increasing age as well as higher mortality in women than in men in all groups. The multivariable adjusted risk in women relative to men was highest amongst the youngest, OR 1.45 (95% CI 0.98 – 2.14). The long term prognosis was assessed in women vs. men with Cox proportional regression analyses, follow-up time 1 to 13 years. Women had 8% lower risk of long term mortality after multivariable adjustments, and after age-stratifying, women had better long term survival in all age-groups, except the youngest.
Previous studies based on mixed MI cohorts had found a gender-age interaction with higher risk of death women relative to men in the youngest group. In Paper II we included all STEMI patients <46 years old registered in RIKS-HIA between 1st Jan 1995 and 31st Dec 2006, 1748 men and 384 women. Cardiovascular risk factors were common, and women had more often clustering of risk factors compared to men. The most prevalent risk factor was smoking, 64% of the women compared to 58% of the men were current smokers. There was no gender difference in delay times or in rate of reperfusion. Almost 60% of both women and men underwent coronary angiography within one week. There was no gender difference in prevalence of non-obstructive disease, (p=0.64), but men had had multi-vessel/left main disease much more often than women (33.6% vs. 19.2%; p<0.001). In-hospital mortality was low, 3% in women vs. 1% in men, crude OR women vs. men 2.83 (95% CI 1.32 – 6.03). Female gender appeared as an independent predictor in the multivariable model of in-hospital mortality, OR 2.85 (95% CI 1.31– 6.19). When the cohort was followed up to 10 years (mean 5.4 years) the risk of mortality was not higher in women (hazard ratio [HR] 0.93, 95% CI 0.60 – 1.45; p=0.75), and men had significantly higher risk of a second new MI during the following 10 years, HR 1.82 (95% CI 1.25 – 2.65; p=0.002).
In the beginning of the 21st century there was a shift in reperfusion strategy with a decline in use of fibrinolytic therapy and an increase in use of primary PCI. We hypothesised that the gender differences noticed during the fibrinolytic era with lower chance of receiving reperfusion therapy and higher risk of early mortality in women, would have diminished during the new primary PCI era, as this is a better reperfusion strategy, especially for women. In Paper III we included STEMI patients from two time periods with different dominating reperfusion strategies in order to compare management and outcome between genders in both periods. Patients in the early period (n=15 697, 35% women) were registered in RIKS-HIA between 1st Jan 1998 and 31st Dec 2000 and those in the late period (n=14 380, 35% women) between 1st Jan 2004 and 31st Dec 2006. Among patients treated with reperfusion therapy 9% in the early compared to 68% in the late period were treated with primary PCI. The use of reperfusion therapy increased between the two periods, in men from 70.9% to 75.3%, in women from 63.1% to 63.6%. After multivariable adjustment, women were 14% and 20% less likely than men to receive reperfusion therapy, early and late periods, respectively. Heart failure, cardiogenic chock and major bleedings were more common in women compared to men. Evidence-based secondary preventive therapies were prescribed more often in the late compared to the early period in both genders, but more seldom to women in both periods. After multivariable adjustments women still had less chance of receiving ACE-inhibitors/ARBs but higher chance of receiving statins in the early period. In the late period women had 14 – 25% less chance of receiving any of the evidence-based secondary preventive therapies.
In Paper IV all consecutive patients who fulfilled the criteria for ST-elevation or bundle branch block on admission ECG and who were planned to undergo immediate coronary angiography with the intention to perform primary PCI at the Department of Cardiology in Linköping were included, 98 women and 176 men. Estimated glomerular filtration rate [eGFR] according to Modification of Diet in Renal Disease study [MDRD] was calculated for all patients and they were staged into CKD stages 1-5. Estimated GFR was lower in women than in men, mean eGFR 54 vs. 68 mL/min/1.73m2, p<0.001. Ten men but no woman were classified belonging to the best CKD stage 1(eGFR >90 mL/min/1.73m2). In total 67% of women compared to 27% of men were classified as having renal insufficiency [RI] (eGFR <60 mL/min/1.73m2) and female sex was a strong independent factor associated with RI, OR 5.06 (95% CI 2.66 – 9.59). Reduced eGFR per 10 mL/min decline was independently associated to higher risk of death and MACE (death, new MI or stroke) within one year in women whereas we found no such associations in men. There was a borderline significant interaction between gender and eGFR regarding one year mortality (p=0.08) but not regarding MACE (p=0.11).
As we found a remarkable gender difference in RI prevalence in Paper IV, we analysed an updated SWEDEHEART database including the years since S-creatinine became a mandatory variable to register. In Paper V all STEMI patients registered between 1st of Jan 2003 and 31st of Dec 2009 were included, in total 37 991 patients (36% women). RI was present in 38% in women vs. 19% in men according to MDRD and in 50% of men vs. 22% of men according to Cockcroft Gault [CG] (p<0.001 for both comparisons). Female gender was independently associated with RI regardless of used formula. In both genders, RI patients were older, had higher co-morbidity, suffered from more complications and had lower chance of receiving reperfusion therapy and evidence-based therapy at discharge compared to non RI patients. Among both RI and non RI patients, men had significantly higher chance than women of getting these therapies. In-hospital mortality was four to five times higher in RI vs. non RI patients. RI compared to non RI patients had approximately doubled risk of inhospital mortality in women and 2.5 times higher risk in men after multivariable adjustment. Regardless of used formula, the risk of dying at hospital increased with approximately 30% and the risk of long term mortality with approximately 10% in both genders per 10 mL/min decline of eGFR. There was no significant interaction between gender and eGFR regarding short- or long term outcome according to any of the formulas. Women had twice as high in-hospital and also higher cumulative long term mortality than men. After multivariable adjustments including all confounders except kidney function women had 7% lower risk of long term mortality but still 11% higher risk of in-hospital mortality. If eGFR according to any of the formulas was also included, there was no longer a gender difference regarding in-hospital mortality and women had lower risk of long term mortality. This was also the case if only adjusting for eGFR according to CG.
Conclusion: In the real life STEMI setting, women were older with higher co-morbidity, longer delay, more complications and twice as high in-hospital mortality. They had significantly less chance of receiving acute reperfusion therapy, also after adjusting for possible confounders. During the fibrinolytic era women had higher risk of severe bleedings. We hypothesised that the gap in management would have decreased during the new primary PCI era, with a less time-dependent regime with less risk of fatal complications. Our hypothesis failed, and future studies ought to further scrutinise this gender difference in management. The less chance of reperfusion therapy did anyhow not explain the higher in-hospital mortality in women, which was 10-20% higher after multivariable adjustments, consistent with previous findings. Moderate to severe chronic kidney disease was very common in women with STEMI, 50% according to the Cockcroft Gault formula. Estimated GFR has seldom been taken into account in studies evaluating gender differences in outcome. If adjustment for eGFR was done, alone or added to the all other co-variates, women had no longer higher risk of in-hospital mortality. Adjusted long term outcome was better in women than in men, which was also the case in the youngest cohort when studied separately.
Linköping: Linköping University Electronic Press , 2012. , 100 p.
2012-02-10, Berzeliussalen, Universitetssjukhuset, Campus US, Linköpings Universitet, Linköping, 09:00 (Swedish)
Swahn, Eva, ProfessorStenestrand, Ulf, Associate ProfessorAlfredsson, Joakim, Dr.