Objective To evaluate if female gender is associated with renal insufficiency in patients with ST-elevation myocardial infarction (STEMI) and if there is a gender difference in the prognostic importance of renal insufficiency in STEMI.
Design Single-centre observational study.
Setting One tertiary cardiac centre.
Patients All consecutive patients with STEMI planned for primary percutaneous coronary intervention in one Swedish county in 2005 (98 women and 176 men).
Main outcome measures Logistic regression analyses were conducted to evaluate the predictors of renal insufficiency, associations between estimated glomerular filtration rate (eGFR) and outcome in each gender and a possible interaction between gender and eGFR regarding outcome.
Results Renal insufficiency was defined as eGFR less than 60 ml/min per 1.73 m2. 67% of women had renal insufficiency compared with 26% of men, OR 5.06 (95% CI 2.66 to 9.59) after multivariable adjustment. In women each 10 ml/min per 1.73 m2 increment of eGFR was associated with a 63% risk reduction for 1-year mortality, OR 0.37 (95% CI 0.15 to 0.89). No such association was found in men, OR 1.05 (95% CI 0.63 to 1.76). A trend towards a significant interaction between gender and eGFR regarding 1-year mortality was found, OR 2.05 (95% CI 0.93 to 4.50).
Conclusions A considerable gender difference in the prevalence of renal insufficiency in STEMI was found and renal insufficiency seemed to be a more important prognostic marker in women. These results are important as previous STEMI studies have shown higher multivariable adjusted mortality in women than in men but renal function has seldom been taken into consideration.
- coronary intervention
- myocardial infarction
- percutaneous coronary intervention
- renal disease
- renal insufficiency
- sex factors
Statistics from Altmetric.com
- coronary intervention
- myocardial infarction
- percutaneous coronary intervention
- renal disease
- renal insufficiency
- sex factors
End-stage renal failure is a well known independent risk factor both for the development and poor outcome of coronary artery disease.1 2 Mild to moderate renal insufficiency has recently also been shown to be associated with adverse outcomes among patients with acute coronary syndrome (ACS).3–6 According to previous studies the prevalence of renal insufficiency seems to be higher in women than in men among myocardial infarction (MI) patients.3 7 Although it is not clear whether this is explained by greater age and comorbidity in female compared with male MI patients or if female sex is an independent predictor of renal insufficiency in this setting. Several studies have shown worse outcomes in women than in men with ST-elevation myocardial infarction (STEMI) also after multivariable adjustments, but few studies have investigated the role of renal insufficiency.8–10 We wanted to evaluate if there exist sex differences in the occurrence and importance of renal insufficiency in a population with STEMI, and therefore we chose to study all consecutive STEMI patients in 1 year who arrived at our percutaneous coronary intervention (PCI) unit for primary angioplasty as a result of STEMI.
We hypothesised that:
Female sex is an independent predictor of renal insufficiency in STEMI.
The association between renal insufficiency and outcome after STEMI is stronger in women than in men.
Since 1 January 2005 all patients in the county of Östergötland (425 000 inhabitants) with signs of STEMI (symptoms suspicious of ACS combined with ST-elevation or new bundle branch block on ECG) have been directly transferred to the cath lab at the Heart Centre at Linköping University Hospital for primary PCI. The patients are consecutively logged on to a log sheet at the coronary care unit with information about onset of symptoms and ECG findings. They are also registered in the prospective observational Register of Information and Knowledge about Swedish Heart Intensive Care Admissions11 and in the Swedish Coronary Angiography and Angioplasty Registry.12
For this study, we consecutively included all patients between 1 January and 31 December 2005 who fulfilled the criteria for ST-elevation or bundle branch block on the admission ECG, who underwent acute coronary angiography with the intention of primary PCI because of STEMI and who received a MI diagnosis at discharge. If the patient had more than one STEMI during the study period only the first STEMI was included.
Data regarding ECG on admission, cardiovascular risk factors, Killip class on arrival, medications, results of investigations, complications during hospital care, diagnoses at discharge and the occurrence of a new MI during the first year after the index STEMI were retrieved from the Register of Information and Knowledge about Swedish Heart Intensive Care Admissions. Key times were retrieved from the log sheet at the coronary care unit. Angiographic data, including data about complications after PCI and information about any new angiography or PCI within the first year after the index MI, were retrieved from the Swedish Coronary Angiography and Angioplasty Registry.
Serum creatinine (Scr) on arrival was retrieved from patient files. Valid data was available for 270 patients (99%). The estimated glomerular filtration rate (eGFR) was calculated for each patient using the modification of diet in renal disease (MDRD) study equation; GFR (ml/min per 1.73 m2)=186×(Scr/88.4)−1.154×(age)−0.203×(0.742 if female) (SI units).13 14 We assumed that all our patients were non-blacks and the correction factor for race (×1.212 if black) was not used. According to guidelines from the National Kidney Foundation we analysed MDRD eGFR at 90 or more (chronic kidney disease (CKD) stage 1), 60.0 to 89.9 (CKD stage 2), 45.0 to 59.9 (CKD stage 3a), 30.0 to 44.9 (CKD stage 3b), 15.0 to 29.9 (CKD stage 4) and less than 15.0 ml/min per 1.73 m2 (CKD stage 5). Reduced kidney function was defined as below 60 ml/min per 1.73 m2 (≥CKD stage 3).15 As Scr at arrival may be affected by the acute condition with possible acute heart failure, acute bleeding/dehydration, admitted drugs, etc, we performed a sensitivity analysis using Scr values analysed within 1 year before or after the acute event (the first 2 weeks after the acute event excluded). At least one more Scr value was found in 249 of the 274 patients (91%). The correlation between the eGFR on admission and the eGFR based on earlier or later Scr levels was high, r=0.76, p<0.001 (in women r=0.66, p<0.001; in men r=0.72, p<0.001) and the difference between these two eGFR means was not statistically significant, p=0.68 (in women p=0.92; in men p=0.56).
Information regarding other comorbidities was obtained from the National Patient Register, which was also used to find hospitalisations for stroke within 1 year from index STEMI. The cardiac surgery database at the Heart Centre of Östergötland and the Swedish Heart Surgery Register were examined to find patients who had undergone coronary artery bypass grafting after the index STEMI. We obtained vital status information on all Swedish citizens from the National Cause of Death Register. Patients were followed for 1 year regarding death, new MI, new revascularisation and rehospitalisation for stroke. A major adverse cardiac event (MACE) was defined as death, non-fatal MI, stroke or new revascularisation not planned at the index event within the first year after the index STEMI. No patient was lost to follow-up.
Categorical variables are presented as numbers and percentages and compared by the χ2 test. Continuous variables are presented as means with SD or medians with IQR and the t test or the Mann–Whitney U test was used.
Multivariable logistic regression analyses were used to define independent predictors of renal insufficiency (defined as eGFR <60 ml/min per 1.73 m2). Covariates included were: age (in 10-year increments) as a continuous variable and the dichotomous variables sex, active smoker, symptom to needle time more than 3 h, primary PCI performed, diabetes, hypertension, previous MI, previous PCI, previous heart failure, previous stroke, chronic obstructive lung disease, cancer within the past 3 years and therapy on arrival (ACE inhibitors/angiotensin receptor blockers, statins, aspirin, other platelet inhibitors and/or beta-blockers). Backward elimination was used and variables with p values less than 0.05 were kept in the final model.
Multivariable logistic regression analyses were also used to define if there was an interaction between sex and eGFR regarding 1-year mortality and 1-year MACE, respectively. The covariates were the same as above but instead of therapy on arrival, therapy at discharge was used. In addition, the continuous variable eGFR (in 10 ml/min per 1.73 m2 increments) was added and the product of sex and eGFR as an interaction term. Backward elimination was used and variables with p values less than 0.05 were kept in the final model.
Separate analyses on men and women were also performed regarding 1-year mortality and 1-year MACE with the same variables included as above. Backward elimination was then conducted for each sex separately in the same way as previously described.
Kaplan–Meier estimates for women and for men stratified according to CKD stage for 1-year mortality and 1-year MACE were determined and presented as event curves. The log-rank test was used to compare the strata. All p values were two-sided, and a p value of less than 0.05 was considered to indicate statistical significance. Statistical analyses were performed using SPSS version 15.0.
Two hundred and seventy-four patients were included in the study, 98 women and 176 men. The women were older, had a higher prevalence of hypertension and chronic obstructive lung disease and were more often treated with platelet inhibitors, digitalis or diuretics before admission. There was no sex difference in delay times or in admission ECG findings (table 1).
PCI were performed more seldom in women who had a higher incidence of non-obstructive angiography and more often thrombolysis in myocardial infarction (TIMI) 2–3 flow before PCI (table 2).
There was no sex difference in Killip class on admission (table 1) or in complications during hospital care. Left ventricular assessment was more seldom conducted in women. At discharge, statins and antiplatelet therapy was prescribed more often to men, whereas diuretics and anticoagulants were prescribed more often to women (table 2).
The inhospital (table 2) as well as the 1-year mortality was almost twice as high in women as in men, but the differences were not statistically significant. There was no sex difference in MACE within the first year after the index infarction (table 3).
The mean Scr level was higher in men than in women, (table 1) but after estimating the GFR for each patient, renal insufficiency was more common in women (67%) than in men (26%), 5.06 (95% CI 2.66 to 9.59) p<0.001 after multivariable adjustments (table 4).
Kaplan–Meier curves with log-rank tests showed that a higher CKD stage was significantly associated with higher 1-year mortality in women but not in men (figures 1 and 2). Kaplan–Meier curves with log-rank tests regarding 1-year MACE showed that a higher CKD stage was significantly associated with 1-year MACE in women; the same association had borderline significance in men (figures 3 and 4). In order to adjust for possible confounding factors, logistic regression analyses were conducted. Using backward elimination analysing each sex separately, the final models showed that each 10 ml increment of eGFR was associated with a 63% risk reduction of 1-year death and 39% reduction of 1-year MACE in women. In men, no such associations were found (table 5).
The interaction tests showed borderline significant interaction between sex and eGFR regarding 1-year mortality; OR 2.05 (95% CI 0.93 to 4.50, p=0.08) but no significant interaction regarding 1-year MACE; OR 1.42 (95% CI 0.93 to 2.17, p=0.11).
As far as the investigators are aware, this is the first study to investigate gender differences regarding the influence of renal insufficiency on outcome after STEMI. Women have been shown to have 10–20% higher odds of early mortality in STEMI even after adjustment for all kinds of possible confounders.8–10 16 Our findings of female sex being an independent strong predictor of renal insufficiency in STEMI patients and its possibly higher prognostic impact in women could be an important explanation for this observed sex difference.
Prevalence of renal insufficiency
By calculating the eGFR based on the MDRD formula, a more accurate method of evaluating renal function than Scr values, we found that women had renal insufficiency far more often than men. The mean eGFR in women was 54 ml/min per 1.73 m2 compared with 68 ml/min per 1.73 m2 in men. This could be compared with a MDRD eGFR reference material of non-diseased Caucasians in which the mean eGFR was above 60 ml/min per 1.73 m2 in all age groups in both sexes except for women over 85 years whose mean eGFR was 59 ml/min per 1.73 m2.17 Several studies have suggested that a cut off value for an eGFR of less than 60 mL/min/1.73 m2 is predictive of adverse cardiovascular outcomes, i.e. CKD stage 3 or greater.13 18 19 In our study, we found a dramatic gender difference where 67% of women had renal insufficiency compared with only 26% of men. In the population with eGFR greater than 60mL/min per 1.73 m2, only 20% were women compared with 83% in the population with eGFR less than 30 mL/min per 1.73 m2. In concordance with our findings, Sadeghi et al20 showed from the CADILLAC trial, using the Cockgroft Gault formula, that the proportion of women was 51% in the group with renal insufficiency compared with 21% in the group without. In the ExTRACT-TIMI 25 Trial, also using the Cockgroft Gault formula, 12% were women in the group with eGFR >90 ml/min compared to 59% in the group with eGFR <30 ml/min.4 These findings have been attributed to their higher age and incidence of hypertension and diabetes, but our study showed that female sex is a strong independent predictor for impaired renal function in MI patients, even after adjustment for these variables. In fact, female sex was associated with a five times higher risk of renal insufficiency than in men after multivariable adjustment.
Prognostic importance of renal insufficiency
The association between mild to moderate renal insufficiency and poor outcome after ACS and/or PCI has become evident from various studies during recent years. For example, the CADILLAC study showed that in patients undergoing primary PCI a reduced eGFR was found to be a powerful independent predictor of 30-day and 1-year mortality.20 Data from four well-known ACS trials demonstrated that mild to moderate renal insufficiency was an independent predictor of adverse outcome with an approximately 20% risk reduction for death in 180 days for each 10 ml/min increment in eGFR.5 Data from the HERS I and II studies, including postmenopausal women, showed that baseline renal function was among the strongest predictors of cardiovascular events, especially long term.21 The ExTRACT TIMI 25 trial including STEMI patients found a 30% higher RR for 30-day mortality per 10ml/min decline in eGFR.4
An association between mild to moderate renal insufficiency and ACS outcome both in men and women thus seems evident, but there is a lack of studies comparing the genders. One Swedish single-centre study of mixed ACS patients showed that a Scr value on admission above the normal upper limit was an independent risk factor for long-term mortality in women but not in men.22 Chen et al7 studied an unselected cohort of patients undergoing coronary angiography, and found that at all levels of CKD the multivariate adjusted HR of 7-year mortality were higher in women than in men but were statistically significant in both sexes. In concordance with Chen et al,7 we found a stronger association between renal insufficiency and poor outcome in women than in men, both regarding 1-year mortality and MACE, when analysing the sexes separately with logistic regression analyses and Kaplan–Meier curves with log-rank tests. In order to confirm these findings we performed an interaction test between sex and the effect of reduced eGFR per 10 ml/min decline that did not show significance regarding MACE and showed borderline significance regarding mortality.
The pathophysiological mechanisms behind the association between renal insufficiency and poor outcome after ACS and PCI are not clearly understood. Many possible explanations could be found23 such as the high prevalence of diabetes mellitus and hypertension in these patients24 as well as their higher risk of contrast induced nephropathy [CIN] after PCI.25 The undertreatment with less revascularisation and less medical therapy in RI patients6 26–27 may also contribute to reduced survival. Besides this, renal failure is associated with metabolic abnormalities23 such as oxidative stress, hypoalbuminaemia, hyperhomocysteinaemia, hyperfibrinogenaemia, insulin resistance, lipid abnormalities, inflammation, and derangements in the calcium–phosphate homeostasis may all contribute to an independent excess cardiovascular risk.28–33 The reason why women have renal insufficiency more often than men among STEMI patients, even after multivariable adjustment, is unknown. Ischaemic heart disease risk factors are the same in women as in men34 35 but diabetes as well as smoking have previously been shown to be more powerful risk factors in women,36 and according to our data this could also be true for renal insufficiency.
Gender differences in therapy
Previous MI studies have found less intensive treatment in women.37 38 This was also noticed in this small cohort regarding statins, antiplatelet therapy and primary PCI. The latter primarily reflects the higher incidence of normal angiograms and non-obstructive disease in women also seen in this cohort. The former probably reflects the more benign lipid status in women compared with men with ACS. STEMI patients are now recommended statin therapy regardless of cholesterol levels,39 and future studies will hopefully not find this gender difference in statin therapy. Regarding antiplatelet therapy, women were treated more often with anticoagulants probably due to the higher prevalence of atrial fibrillation and/or previous strokes. There was no gender difference in therapy if antiplatelet and anticoagulant therapy was analysed combined (data not shown). None of these gender differences in therapy should affect the incidence or importance of renal insufficiency, but were accounted for in the logistic regression analyses.
This is a small single-centre study with few mortality events, but because of this, we also studied a cardiovascular composite outcome. The calculations of eGFR were based on creatinine on presentation, which may not have been at a steady state, and thus may not represent a true estimation of patients' baseline kidney function. A sensitivity analysis was performed using Scr values taken either before or after the STEMI event (within 1 year) with no significant difference compared with the Scr level taken on admission, p=0.68 (in women p=0.92, in men p=0.56).
Our study is unique as it reflects the current management of STEMI in patients at different stages of CKD based on an unselected cohort, and reveals, to the best of our knowledge for the first time, gender as an independent predictor of renal insufficiency and a possible gender difference in the prognostic impact of renal insufficiency in STEMI patients.
Renal insufficiency in STEMI patients was found to be much more common in women than in men, and female gender was found to be a strong independent predictor of renal insufficiency in STEMI. This is important as renal insufficiency affects the dosage of many drugs and increases the risks of contrast-induced nephropathy and bleeding after PCI. We also found a stronger association between adverse outcomes and decline in eGFR in women. Testing the interaction between gender and eGFR regarding adverse outcomes only showed borderline significance, and a possible gender difference regarding the prognostic importance of renal insufficiency has to be confirmed in studies with more patients and events.
The authors would like to thank statistician Karl Wahlin, Ph.D., and Director of Studies at the Statistics and Data Analysis Programme at the Department of Computer and Information Science, Linköping University, for his statistical advice.
Competing interests None.
Ethics approval This study complies with the Declaration of Helsinki and was approved by the local ethics committee at Uppsala University.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.