Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 1995;91:1861-1871

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vaccarino, V.
Right arrow Articles by Horwitz, R. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vaccarino, V.
Right arrow Articles by Horwitz, R. I.

(Circulation. 1995;91:1861-1871.)
© 1995 American Heart Association, Inc.

Articles

Sex Differences in Mortality After Myocardial Infarction

Is There Evidence for an Increased Risk for Women?

Viola Vaccarino, MD; Harlan M. Krumholz, MD; Lisa F. Berkman, PhD; Ralph I. Horwitz, MD

From Yale University School of Medicine, New Haven, Conn.

Correspondence to Dr. Harlan Krumholz, Section of Cardiology, Yale University School of Medicine, 333 Cedar St, PO Box 8017, New Haven, CT 06520-8017.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Background A number of studies have indicated that women who have a myocardial infarction have higher mortality rates than men. The purpose of the present study was to review the literature on sex differences in mortality after myocardial infarction to determine whether female sex is independently associated with lower survival.

Methods and Results Reports were identified mainly through a MEDLINE search of the English-language literature from January 1966 through June 1994. Studies included were those comparing mortality after myocardial infarction between men and women, controlling at least for age and with more than 30 outcome events. After duplicate patient series were eliminated, 27 reports were included in our review. Crude rates were higher in women than in men during the early phase (in-hospital or first month), but control for age alone or in combination with other factors reduced sex differences in almost all studies. Unadjusted mortality rates among the survivors of the early phase were similar for men and women in most studies, and control for age and other factors resulted in an increased survival rate in women compared with men in several investigations, particularly those with a follow-up of >1 year.

Conclusions Much of the increased early mortality after myocardial infarction in women is explained by the older age and more unfavorable risk characteristics of the women. In the long run, when differences in age and other risk factors are controlled for, women tend to have an improved survival compared with men.


Key Words: myocardial infarction • sex • mortality


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
down arrowResults
down arrowDiscussion
down arrowReferences
 
The results of numerous studies indicate that coronary heart disease morbidity1 2 and mortality1 3 4 occur about 10 years later in women than in men. A paradox has been observed, however, that once women develop a myocardial infarction, they have higher rates of mortality during hospitalization or during the first 30 days after infarction in almost all studies.5 6 7 8 9 10 11 12 13 14 15 16 17 Although abundant research has focused on "female protection" for the development of coronary heart disease, little has been done to clarify why once coronary heart disease becomes clinically manifest, women have worse outcomes. Curiously, female sex is invoked to explain both the decreased risk of coronary disease and the increased risk of mortality after myocardial infarction.12

Can the apparent poorer prognosis of women after myocardial infarction be explained by differences in the clinical epidemiology of coronary heart disease in the two sexes, or does it reflect biological peculiarities of coronary heart disease among women? The clarification of this issue will aid in our understanding of coronary heart disease in women and aid in the development of effective measures to improve their clinical course and outcome.

The purpose of the present review was to determine whether sex differences in survival can be explained by differences in baseline characteristics between men and women. Our hypothesis was that factors related to the epidemiology of coronary heart disease in women, such as women's older age and higher prevalence of comorbidity at the time of the infarction, exert a primary role in the lower survival rates of women and may account for most or all of the observed sex differences in mortality. We expected that studies that adjust for these differences will reduce or eliminate the apparent survival disadvantage of women.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Article Selection
We searched the English-language medical literature for articles that compared mortality rates after myocardial infarction between men and women. Articles were found through a MEDLINE search (from January 1966 through June 1994) by using as search terms the Medical Subject Headings (MeSH) "myocardial infarction," "myocardial infarction mortality," "myocardial infarction complications," and "sex factors." Another source of articles included references in articles found by the above method.

Articles were included if at least one of the following was reported: mortality rates by sex, a sex-related relative risk of mortality, or a significance test for the comparison of mortality rates between men and women. Because age is a crucial confounder when survival rates after myocardial infarction are compared in men and women, studies that did not include age adjustment were not considered. In addition, we excluded articles that considered only prevalent cases of myocardial infarction; in which the myocardial infarction was self-reported or not defined by contemporary criteria of typical symptoms, serial ECG changes, and enzyme elevations; that included patients described in later publications for a similar study end point; and that included fewer than 30 outcome events. After these studies were eliminated, the final literature review included 27 investigations.

End Point Definition
In most reports the outcomes were in-hospital mortality and/or mortality after discharge among hospital survivors. In some studies, the end points were mortality during the first month since the infarction17 18 and mortality among survivors of the first month.18 19 20 21 22 One study used a cutoff point of 6 weeks to define early mortality.23 A number of investigations8 17 18 23 24 25 26 27 28 presented mortality rates over the entire follow-up period, combining the hospital and the postdischarge mortality.

In the present review, two main end points were evaluated: early mortality, defined as in-hospital mortality or mortality during the first 4 or 6 weeks, and late mortality, including mortality among hospital survivors or mortality among survivors of the first month. The available data on cumulative mortality, defined as mortality since admission to the hospital to the end of the follow-up, are also examined. For articles that gave information on more than one length of follow-up for the same end point considered in our review (eg, death at 1, 3, and 5 years after discharge), the time point closest to admission was considered.

Data Extraction
From each article, crude and age-adjusted mortality rates in men and women, crude and age-adjusted female-to-male relative risks of death, or both were extracted for each end point. For articles in which investigators controlled for factors in addition to age, death rates and/or relative risks adjusted simultaneously for age and other predictors were also extracted. When rates, relative risks, or P values were not reported, they were calculated from the data provided, whenever possible. Similarly, when not done by the authors and when sufficient data were provided,5 6 9 17 18 29 summary age-adjusted Mantel-Haenszel odds ratio and statistic were calculated.

Comparison of Study Results
The original plan was to perform a meta-analysis of the study results. However, after the literature was collected and examined, it was clear that a meta-analysis was not feasible because of the heterogeneity in the factors considered in the multivariable analyses across studies, as described in "Results." Therefore, instead of pooling the study results in a summary estimate, we compared them taking into account both direction and magnitude of the relative risks reported, rather than statistical significance, since some of the studies may have lacked power to find significant associations. Because a number of studies considered only age as a control factor, studies were initially compared for age-adjusted results. Studies that also controlled for other differences were further compared for results adjusted simultaneously for age and other covariates.

Because of the later onset of clinical coronary heart disease in women, we anticipated that the older age of female patients compared with male patients is a major explanation of the poorer survival of women. Therefore, particular attention was devoted to the analytical strategies used in the different studies to control for age. Age adjustment was considered adequate if one of the following methods was used: stratified analysis or logistic regression using age intervals of <=10 years, standardization of the age distribution of men and women to the age distribution of the entire patient group or to the age distribution of the population from which the sample originated, or matching by age of women and men.

In addition to age, other preinfarction characteristics may differ between women and men and may contribute to an explanation of sex differences in outcome. In studies in which factors in addition to age were controlled for, we examined whether differences between men and women in preinfarction factors related to mortality were taken into account in the multivariable analysis. The control for the following factors was considered particularly important since their distribution usually differs in women and men: previous coronary heart disease, history of diabetes mellitus, history of hypertension, history of congestive heart failure, and history of other major comorbidities (eg, cerebrovascular disease, pulmonary disorders, and renal failure).

In comparing study results, we also considered potential differences in study populations. The results of investigations based either on clinical trials or on community studies in which participation rate was rather low (ie, <80%) were considered potentially less valid than the results based on one of the following types of study samples: consecutive hospital admissions, coronary registries covering virtually all hospitalizations for myocardial infarction in specific populations, or community samples with a response rate of >=80%. These study designs ensured that the majority of men and women hospitalized for myocardial infarction in a specific population or hospital were included in the study and that men and women were equally likely to be included. Randomized trials usually have stringent eligibility criteria and require formal patient consent and therefore may be less representative of the majority of the people hospitalized for myocardial infarction. In addition, not all eligible patients are enrolled in clinical trials (usually because of patient or physician refusal), and the rate of exclusion of eligible patients may differ in men and women. Similarly, the representativeness of community samples depends on the participation rate of the target population, and if the participation rate differs between eligible men and women, sex comparisons may be biased. Unfortunately, usually no information is provided on enrollment rates by sex.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
*Results
down arrowDiscussion
down arrowReferences
 
The studies included in the present review are listed in Table 1Down. Most were based on consecutive hospital admissions5 6 9 10 11 12 13 14 19 20 26 30 31 or were community based.1 Four studies were randomized clinical trials,8 15 23 28 and one study18 was based on participants in a health insurance plan who sought medical care for symptoms of heart attack. As previously mentioned, only studies that reported at least age-adjusted or age-stratified results contributed to specific end points. For example, in the MILIS study,8 although the results at 4 years of follow-up were multivariable adjusted, only unadjusted data were reported for hospital mortality. Therefore, only the 4-year data were considered from the MILIS study. Conversely, in the report by Dittrich et al,9 data were age adjusted for hospital mortality but unadjusted for 1-year mortality. Therefore, this study was considered only for hospital mortality.


View this table:
[in this window]
[in a new window]
  		Table 1. Studies Comparing Mortality After Myocardial Infarction Between Women and Men

Most of the population-based studies were based on registries, including virtually all hospitalized patients with myocardial infarctions in the populations under study.16 17 21 22 24 29 Therefore, no important selection bias is anticipated in these studies. The remaining population studies were based on follow-up studies of community samples.25 27 32 The response rate of these community samples was 82% in one study25 but only 69% in the two reports based on the Framingham Study cohort.27 32 In addition, the Framingham cohort included, together with the community sample, 14% of volunteers who were not part of the original sampling frame. Furthermore, one of these two Framingham reports32 was limited to hospital survivors who attended an examination subsequent to their infarction, which might have occurred as long as 4 years after the infarction.

Several studies excluded older age groups. Four investigations excluded patients aged >=60 years,18 20 21 29 one excluded those aged >=70 years,22 and four excluded those aged >=75 years.8 11 17 23 On the other hand, one study only considered persons aged >=65 years.25

Women were older than men in all studies that provided data on mean age by sex (Table 1Up) or age distribution by sex.2 In addition to age, women and men differed in the distribution of other prognostic factors. Women had a higher prevalence of history of hypertension,3 congestive heart failure,5 8 9 11 12 14 16 23 and diabetes.5 6 8 11 12 16 23 26 27 28 Women also had a higher incidence of mechanical complications during hospitalization, such as congestive heart failure and cardiogenic shock.10 11 12 16 23 28 31 On the other hand, men more frequently had a previous myocardial infarction,9 11 12 14 26 28 cigarette smoking habit,9 10 11 12 27 28 31 and arrhythmic complications.10 12

Early Mortality
Table 2Down includes the 17 studies that compared in-hospital or 1-month mortality rates between men and women. Only 4 of these studies were derived from clinical trials15 23 28 or other selected samples18 ; the remaining investigations were based on either registries16 17 or consecutive hospital admissions.5 6 9 10 11 12 13 14 26 30 31


View this table:
[in this window]
[in a new window]
  		Table 2. Studies Comparing Early Mortality (In-Hospital or First Month) After Myocardial Infarction Between Women and Men

Unadjusted Results
All of the investigations listed in Table 2Up showed higher crude (unadjusted) death rates among women, except for one study,26 which differed from the others in that it was limited to black patients. The increase in mortality in women relative to men ranged from <10%30 to >100%6 15 23 but was >40% in most studies.6

Results Adjusted for Age
Eleven studies presented age-adjusted or age-stratified data (Table 2Up). One study only reported a nonsignificant P value.10 Most studies performed age adjustment by logistic regression or stratification, using age strata of <=10 years, with the exception of four studies.6 12 26 31 In the study by Puletti et al,6 age strata of 15-year intervals were used, and the study population was subdivided into the following age strata: <=40, 41 to 55, 56 to 70, and >70 years. In the report by Greenland et al,12 all patients >=70 years old were grouped together. In the study by Liao et al,26 patients were classified into one of three age categories: <55, 55 to 64, and >=65 years. In the study by Jenkins et al,31 patients were classified in two categories: <65 years or >=65 years. With one exception,18 all of the studies that presented age-adjusted data were based on consecutive hospital admissions or coronary registries. Therefore, potential biases deriving from the selection of the populations probably were not important in these results.

Control for age reduced sex differences in mortality in all studies. After age adjustment, the sex difference in mortality became <20% and not statistically significant in all investigations that presented age-adjusted relative risks, except two.6 12 On the other hand, Puletti et al6 and Greenland et al12 reported age-adjusted female-to-male relative risks of 1.4 and 1.6, respectively. However, these two studies are among those in which age adjustment might have been incomplete, according to the criteria specified in "Methods." All of the studies in which age adjustment was performed according to our criteria5 9 13 16 17 18 30 showed a decrease in the relative risk to <1.2.

Results Adjusted for Age and Other Covariates
Only nine studies compared death rates in men and women after controlling for factors other than age,9 11 12 14 15 16 23 28 31 and of these, only six presented multivariable-adjusted rates or relative risks.11 12 16 23 28 31 The comparison of the results of these studies is difficult because the factors that were controlled for varied across studies. Although all investigations took into account previous infarctions and history of hypertension, five studies12 15 16 28 31 did not consider history of congestive heart failure, and two studies12 14 did not control for history of diabetes mellitus. Only three investigations controlled for a complete set of the following major historical factors: previous coronary heart disease, history of diabetes, history of hypertension, and history of congestive heart failure.9 11 23 None of the studies controlled for comorbid diseases other than previous coronary heart disease and coronary heart disease risk factors. In addition to medical history, some studies considered severity features on admission. Seven studies controlled for location of the infarction,9 11 12 15 16 23 31 five took into account congestive heart failure on admission,9 11 12 15 16 and six considered peak levels of creatinine kinase.9 11 12 15 16 31 Three studies also adjusted for complications that might have occurred at any time after hospital admission through discharge, such as congestive heart failure12 16 and shock.12 16 31

Three studies were secondary analyses of randomized trials in which all patients met eligibility criteria and consented to receive thrombolytic therapy.15 23 28 All of the other investigations were based on consecutive hospital admissions9 11 12 14 31 or were population based.16

Of the nine studies that controlled simultaneously for age and other differences between men and women, one did not provide information on the magnitude of the effect of covariable adjustment on sex differences in mortality.15 All of the remaining studies except two12 31 either showed11 16 23 28 or indicated in the text9 14 that adjustment for baseline differences in risk factors lowered the risk of death for women compared to the unadjusted or age-adjusted results. Nonetheless, in all of the six studies that reported relative risks adjusted for age and other covariates,11 12 16 23 28 31 women still had an increased mortality, which was >25% compared with men in four studies11 12 23 31 and statistically significant in one12 and possibly a second23 study that used 99% confidence intervals. Two of these studies adjusted for the important historical conditions, including previous coronary heart disease, congestive heart failure, hypertension, and diabetes.11 23 Also, these two studies reported a relative risk of >1. It should be noted, however, that the results of some of these studies may be limited by the selected characteristics of the study population,23 28 as well as by inadequate age12 31 and covariable adjustment,12 28 31 as described above.

Late Mortality
The 12 studies in which investigators compared mortality rates between men and women among survivors of the hospitalization or of the first month are listed in Table 3Down. All of these studies were either community studies or based on consecutive hospital admissions, with two exceptions.15 18 All of the community reports were likely to be representative of the populations studied, except the study by Wong et al,32 which included a selected sample of hospital survivors, as previously noted.


View this table:
[in this window]
[in a new window]
  		Table 3. Studies Comparing Late Mortality Between Women and Men Who Survived the Early Phase (Hospitalization or First Month) After Myocardial Infarction

Unadjusted Results
Of the 12 studies, 4 reported higher unadjusted death rates in women than in men,12 15 19 30 and 1 showed significantly higher death rates among men.21 All of the remaining 7 studies observed either similar crude mortality rates in men and women or a trend for lower mortality in women.

Results Adjusted for Age
Age adjustment was performed according to the criteria we specified in "Methods," with the exception of the study by Greenland et al,12 in which all people aged >=70 years were included in a single age stratum. As for early mortality, control for age resulted in a decrease of the female-to-male relative risk of death. Only Greenland et al12 reported a significantly higher late mortality in women after age adjustment, and this discrepant finding may be due to incomplete age adjustment. All of the other studies that presented age-adjusted data reported either similar mortality in men and women16 29 30 or a trend toward a lower mortality in women.18 21 32

Results Adjusted for Age and Other Covariates
Eight studies presented relative risks or P values adjusted for a number of predictors in addition to age. The predictors included in the multivariable analysis varied considerably across studies, ranging from just traditional coronary heart disease risk factors and diabetes32 to clinical data collected on admission11 15 19 22 or clinical admission data and hospital complications.12 16 20 22 Only half of these studies considered all major variables such as history of coronary heart disease, history of diabetes, history of hypertension, and history of congestive heart failure or heart failure during hospitalization.11 15 16 22 None of the studies took into account presence of noncardiac comorbidity, with the exception of the study by Martin et al,22 which considered history of stroke.

Of these eight studies, six reported adjusted relative risks.11 12 16 20 22 32 The study by Greenland et al12 is the only one showing a higher mortality in female hospital survivors after multivariable adjustment for age and other differences in comorbidity and risk factors. The remaining five studies showed either an increased survival in women11 16 22 32 or similar death rates in men and women.20 All of the investigations that controlled simultaneously for all major historical factors, ie, history of coronary heart disease, of hypertension, or of diabetes and congestive heart failure before the infarction or during hospitalization, found a significantly increased survival in women.11 16 22

Because follow-up varied considerably among the studies that evaluated late survival (between 6 months and 15 years), we compared the study results taking into account the length of follow-up. Since infarction-related mortality occurs mostly in the first year after the myocardial infarction,21 33 we were interested in examining whether the relation between sex and mortality after discharge varies when follow-ups of >1 year are used rather than follow-ups of <=1 year. Of the seven studies with a follow-up of >1 year in which adjusted relative risks were reported,11 16 18 20 21 22 32 six studies found a female-to-male relative risk, adjusted for age alone or in combination with other factors, of <1,11 16 18 21 22 32 which was significant in five studies,11 16 21 22 32 indicating an improved survival rate for women. Conversely, of the three studies that followed patients up to 1 year and in which relative risks were reported,12 29 30 none found a better outcome for women.

Cumulative Mortality Since Hospital Admission
Nine studies that reported sex differences in cumulative mortality from hospital admission to the end of follow-up are listed in Table 4Down. These studies differ greatly in the selection of the study population, being based on randomized trials8 23 28 or other selected samples,18 based on community samples,25 27 composed of consecutively admitted patients,26 or derived from coronary registries.17 24 Only four of these studies presented age-adjusted data.17 18 24 27 Three of them found very similar mortality rates in women and men after age was controlled for.17 18 24


View this table:
[in this window]
[in a new window]
  		Table 4. Studies Comparing Cumulative Mortality Since Hospital Admission Between Women and Men After Myocardial Infarction

Six studies reported results adjusted for age and other factors.8 23 25 26 27 28 Although most of them adjusted for a number of historical characteristics, such as history of previous coronary heart disease, diabetes, and hypertension, only two studies took into account history of congestive heart failure8 or heart failure on admission.25 The results of these studies appear inconsistent. Three studies indicate an increased mortality rate for women,8 23 27 two suggest a lower mortality rate in women,25 26 and one reports a similar outcome in men and women.28 Heterogeneity of the study populations as well as incomplete control for baseline differences may be responsible for these discrepant findings.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
up arrowResults
*Discussion
down arrowReferences
 
Several studies have provided data on sex differences in mortality after myocardial infarction. These studies consistently show an increased mortality in women during the early phase after the infarction (hospitalization or first 4 to 6 weeks) in unadjusted analysis. In most studies, much of this early increase in female mortality is explained by the older age and higher prevalence of unfavorable baseline conditions of women. Of 15 studies that presented short-term mortality data adjusted for age alone or in combination with other factors,4 10 reported a decrease in the magnitude of the female-to-male relative risk to <1.2.5 On the other hand, after the early phase only a few studies reported an increased mortality in women in unadjusted analysis.12 15 19 30 When allowance is made for age and other differences, women actually show an increased survival rate compared with men in several studies.11 16 18 21 22 32

Early Mortality
A major explanatory factor for the increased short-term mortality rate after myocardial infarction for women is the older age of the women compared with the men. Age adjustment alone was able to reduce sex differences in early mortality to <20% in 9 of 11 studies that presented age-adjusted relative risks and in eight studies that controlled adequately for age.

In addition to a more advanced age, however, women hospitalized for myocardial infarction may have a higher prevalence of other unfavorable premorbid conditions that could influence the severity of the infarction and/or the outcome after the infarction. Virtually all studies that reported baseline characteristics by sex indicated that women more often have a history of congestive heart failure, diabetes, and hypertension. Although the reasons for these differences need to be investigated, these differences, if accounted for, might decrease the risk of dying of women compared with men even further. However, only approximately half of the studies controlled for baseline differences in medical history. Of these, only three adjusted simultaneously for major factors such as history of coronary heart disease, diabetes, hypertension, and congestive heart failure.9 11 23 Two additional studies considered congestive heart failure on admission, rather than history of heart failure, in addition to historical factors.15 16 The inclusion of these control factors in the analysis resulted in a decrease of the sex differences in survival after infarction. However, in all of the six studies that presented covariable-adjusted relative risks, these relative risks were still >1 after multivariable adjustment. In three of these studies, the increased mortality in women was statistically significant or almost significant.12 23 31 Therefore, even though baseline differences in age and other characteristics explain a good portion of the increased early mortality in women, women may be still at a higher risk of death than men.

It is difficult, however, to draw firm conclusions on the role of sex on short-term mortality from the current literature for two main reasons. First, studies that included an extensive multivariable analysis may represent a selected group in which sex differences were more pronounced. Only few studies adjusted for differences other than age, and even fewer presented adjusted risk estimates. In general, investigators who found negligible sex differences in mortality after controlling just for age did not proceed to adjust for additional variables.10 13 18 26 30 For the same reason, other studies omitted to report relative risks adjusted for age and other covariates, even though a multivariable analysis including age and other factors was performed.9 14 Therefore, the studies more likely to report relative risks adjusted for baseline characteristics in addition to age were probably those in which the sex differences in mortality were most marked.

The composition and selection characteristics of the study sample of some of these studies may contribute to their finding of a higher mortality in women. For example, the study by Becker et al23 was a randomized trial in which all patients enrolled were treated with a specific type of thrombolytic therapy, ie, recombinant tissue-type plasminogen activator (rTPA). In this study, women were still 50% more likely to die at 6 weeks and 40% more likely to die at 1 year than were men after multivariable adjustment. Two other thrombolytic therapy trials, in which patients were treated with both rTPA and streptokinase, did not find marked differences in mortality between men and women.15 28 The reason for the increased mortality in women in the study by Becker et al23 compared with the other thrombolytic therapy trials15 28 may lie in the fact that more women than men suffer stroke and bleeding complications after rTPA but not after streptokinase.28 It should also be noted that the study samples of thrombolytic therapy trials are likely to differ from those of other studies that did not select patients according to feasibility of thrombolytic therapy. Fewer women than men are likely to be eligible for thrombolytic therapy, since women are older, have more comorbidity, and delay more before seeking medical care for severe chest pain than men.11 34 35 36 Although the age and comorbidity eligibility criteria may mitigate sex differences in outcome, it is unknown how the exclusion of delayed patients might influence the study results. In addition, even among eligible patients, women are less likely to receive thrombolytic therapy than men.37 The effect of this selection bias on the study results also is unknown.

Another study that suggested a poorer outcome in women than in men after multivariable analysis and in which selection bias may be an issue is the MILIS trial.8 Subgroup analyses led the authors to conclude that the poorer prognosis of women in this study was due mostly to the higher mortality of black women (which constituted almost one third of the female sample) compared with black and white men, as well as white women. In contrast, there was little difference in mortality between white men and white women.8 A recent investigation, however, did not confirm the adverse prognosis of black women, with respect to mortality in black men.26 In this study, after a similar length of follow-up, black women were found to have an improved survival compared with black men.26 The study by Tofler et al8 was based on a clinical trial in which as many as 45% of the eligible patients were not enrolled, mostly because of patient or physician refusal.38 The rate of enrollment by sex and race was not reported. Thus, a selection bias, perhaps leading to the enrollment of more severely affected women, particularly black women, compared with men cannot be excluded.

The second reason why the current literature does not give a definite answer to the question whether early mortality is increased in women after correction for baseline differences is the incomplete control for age and/or other important preinfarction conditions in many studies. Incomplete age adjustment may have contributed to spuriously high multivariable-adjusted relative risks in two studies.12 31 Greenland et al12 classified all patients >=70 years into a single age stratum. These patients accounted for 48% of the women and 29% of the men in the study. Because mortality after myocardial infarction increases exponentially with age in these age brackets as in the other age groups,15 almost half of the female sample may not have received adequate age adjustment in this study. Jenkins et al31 used a dichotomous cutoff of age (<65 and >=65), which may not have appropriately accounted for the different age distribution of men and women.

Several investigations did not take into account important comorbid conditions. History of congestive heart failure was not adjusted for in five studies,12 15 16 28 31 although some of these authors controlled for heart failure on admission.12 15 16 Because women hospitalized for myocardial infarction more often have a history of congestive heart failure than men,5 8 9 11 12 14 16 23 premorbid left ventricular dysfunction may contribute to the higher death rates of women. Congestive heart failure on admission should be considered more appropriately an indicator of severity of the infarction rather than a potential confounder that needs adjustment. Other investigators did not control for history of diabetes.12 14 Also, diabetes is more common in women5 6 8 11 12 16 23 26 27 28 and is a stronger prognostic factor after myocardial infarction in women,12 39 40 possibly because of a more adverse effect of diabetes on triglyceride and lipoprotein cholesterol concentrations in women.41 In the study by Greenland et al,12 history of diabetes was more common among women as well as a strong predictor of mortality in women. When diabetics were excluded from the analysis, sex was marginally associated with hospital mortality and no longer predicted postdischarge mortality. Therefore, the inclusion of diabetes among the control factors may have lowered the women's likelihood of dying compared with men in this study. Noncardiac comorbidities, such as cerebrovascular disease, renal failure, and pulmonary disorders, are also likely to be more prevalent in women, given their older age, but age adjustment may not totally account for their effects. The presence of noncardiac comorbid diseases may increase the frailty of female patients with myocardial infarction compared with men and thus limit their ability to overcome the acute phase of the infarction. However, almost none of the studies in the literature have taken into account comorbidities other than previous coronary heart disease, diabetes, and hypertension.

Late Mortality
After hospital discharge or after the first month, only in about half of the studies was the relative risk >1, ie, the crude death rates were higher in women.11 12 15 19 29 30 After age alone or in combination with other covariables was controlled for, only two studies showed an increased mortality in women,12 30 which was significant in one of them.12 The other studies found either similar death rates in women and men15 19 20 29 or lower mortality rates in women.11 16 18 21 22 32 The improved survival was observed particularly when the length of follow-up was >1 year after discharge from the hospital or the first month. Of the seven studies with a follow-up of >1 year that reported relative risks, only one showed an adjusted relative risk that was slightly >1.20 In all of the other six investigations,11 16 18 21 22 32 women had lower adjusted death rates than men, which were statistically significant in five studies.11 16 21 22 32

The lower long-term mortality in women does not appear to be due to their higher short-term case fatality rate, which may result in a cohort of female survivors at a lower risk of death than their male counterparts.42 Only a few studies showed adjusted sex differences in short-term mortality that might have been great enough to be responsible for such a bias.11 12 23 31 It is more likely that the improved long-term survival of women reflects the well-established survival advantage of women in the general population.3 In fact, the impact of the infarction on survival occurs mostly during the first year after infarction.21 33 In the first year after the acute phase, women have either similar20 29 or slightly increased12 30 mortality rates compared with men. However after the first year, when the effects of the infarction are no longer present, women show an improved survival compared with men,11 16 18 21 22 32 as they do in the general population.

In summary, after the first year, when the initial impact of the myocardial infarction has been overcome, women show a survival advantage compared with men in multivariable analysis. However, in the acute phase, and possibly also in the first year afterward, women have a higher unadjusted mortality that is reduced by adjusting for baseline differences, although not totally eliminated. This higher mortality in women may be due to baseline differences between men and women that have not been adjusted for in the literature. Unaccounted sex differences in comorbidity, as well as incomplete control for age, are likely to play a major role in these findings, as discussed above. However, a number of other features may explain this observation.

One of these may be a lower likelihood of women to undergo invasive treatment compared with men. If women are treated less aggressively after myocardial infarction and if invasive treatment improves prognosis, then women may be disadvantaged compared with men. A number of studies have indicated that women are less likely to undergo invasive procedures for coronary heart disease.43 44 However, most of this difference in patient care has been ascribed to the lower probability of severe coronary artery disease in women with coronary symptoms compared with men.45 After myocardial infarction, a condition known to have a prevalence of coronary atherosclerosis approaching 100%, results have been conflicting.13 14 43 46 47 In addition, the impact of invasive treatment after myocardial infarction on prognosis has not been extensively addressed. Therefore, additional research is warranted to investigate further sex differences in the use of coronary procedures after myocardial infarction, as well as the impact of such differences on outcome.

Another aspect requiring consideration is the well-established delay in seeking medical care of women with coronary symptoms.11 23 31 34 35 36 This delay may also contribute to increase the risk of early mortality of women. The reasons for this increased delay of women compared with men are unclear. This finding is unlikely to be due to sex differences in use of health care services because women use health services for cardiovascular conditions as well as for all causes more than men.48 Among patients receiving thrombolytic therapy, delay in reaching the hospital was found to be associated with mortality.49 However, the prognostic role of delay among other patient groups with myocardial infarction needs to be established. Nevertheless, late hospital arrival may play a role in explaining sex differences in mortality. A related issue is the higher prevalence of unrecognized and silent myocardial infarctions among women.50 Although the reasons for this finding are also unknown, it is possible that in addition to delaying their hospital arrival, women underestimate their cardiac symptoms to the point of not seeking any medical care. Because unrecognized infarctions carry prognostic risk,50 they may also contribute to sex differences in outcome after myocardial infarction.

Finally, a number of psychosocial factors, such as social isolation,51 52 53 emotional support,25 socioeconomic resources,54 55 and depression,56 57 have been found to be related to prognosis after myocardial infarction but have not been taken into account in explaining sex differences in outcome after myocardial infarction. Because female patients with myocardial infarction have a more unfavorable psychosocial risk profile,53 57 these features may also play a role in explaining sex differences in mortality.

More research is needed to determine whether the slight increase in early mortality in women found in the current literature is explained by unaccounted baseline differences in comorbidity and psychosocial factors, as well as differences in treatment. Alternatively, this increased mortality in women may be due to peculiar characteristics of the infarction in women that place women at a higher risk of complications and death. For example, it is well established that women experience a number of complications after myocardial infarction more often than men, such as congestive heart failure,10 11 12 30 cardiogenic shock,10 11 12 28 30 infarct extension,58 and cardiac rupture.59 Nevertheless, some commonly used indexes of infarct severity and left ventricular systolic function, such as peak creatine kinase,8 10 11 location of the infarction,6 8 9 10 11 12 and left ventricular ejection fraction,8 9 suggest a less extensive infarction and a greater preservation of left ventricular systolic function in women. The increased occurrence of congestive heart failure in women might result from impairment of left ventricular diastolic function, possibly related to the greater prevalence of diabetes and hypertension in women.8 60 A few reports based on small samples indicate that women may be more likely than men to develop congestive heart failure due to isolated diastolic dysfunction.61 62 However, little is known about the prevalence and the predictors of diastolic dysfunction in women.

A major limitation of the present analysis is the heterogeneity of the investigations reviewed, which makes comparisons of study results difficult. Some studies were based on consecutive hospital admissions, others were population based, and others were secondary analyses of randomized clinical trials. Studies also differed in sample size, and many of them may have lacked sufficient power to detect significant associations of sex with mortality. Although a meta-analysis might have solved this problem, a meta-analysis was not feasible because studies also differed greatly in the numbers and types of characteristics considered in the multivariable analysis. We therefore performed a more qualitative type of comparison of the study results in which we considered the direction and the magnitude of the relative risks reported rather than the statistical significance. Differences in study populations and in covariables included in the multivariable analyses were also taken into account using a qualitative-type approach.

Another limitation of the present study was that the analysis was limited to published literature. Therefore, the results of this review may be affected by publication bias, ie, studies that found sex differences in survival may have been more likely to be reported in literature. However, this limitation applies to any review or meta-analysis. If publication bias is present, sex differences in mortality after myocardial infarction may be less important than reported by the current literature.

Despite these limitations, this review provides important information for understanding the role of sex on prognosis after myocardial infarction. There is a common clinical perception that myocardial infarction is associated with a worse prognosis in women. However, the available data indicate that women fare worse than men only in the short term, and most of this higher initial mortality corresponds to the higher mortality of older patients with more comorbidity. After the early postinfarction period (hospitalization or first 4 to 6 weeks after infarction), sex differences in mortality are less evident. After the first year, women even show an improved survival compared with men.

This review also indicates that more research needs to be done before firm conclusions can be drawn on sex differences in outcome after myocardial infarction. Many of the available studies do not provide definite information because of incomplete adjustment for sex differences in preinfarction conditions. In addition, differences in treatment and psychosocial factors were rarely considered. Finally, the potential existence of unfavorable characteristics of the infarction or of the cardiac function that might be specific for women is largely unexplored. However, the available literature indicates that even if an increased mortality in women does exist, this increased mortality is limited to the first few weeks, possibly up to the first year, after the infarction. Furthermore, this difference in mortality is small once differences in age and other baseline characteristics are taken into account.


*    Acknowledgments
 
This work was supported, in part, by National Institute of Mental Health training grant 5T32-MH-14235-17 and National Institute of Aging contract N01-AG-02105.


*    Footnotes
 
1 References 16, 17, 21, 22, 24, 25, 27, 29, 32. Back

2 References 5, 13, 15, 17-19, 21, 22, 24, 29. Back

3 References 5, 6, 8-12, 14, 16, 23, 26-28, 31. Back

4 References 5, 6, 9, 11-14, 16-18, 23, 26, 28, 30, 31. Back

5 References 5, 9, 13, 14, 16-18, 26, 28, 30. Back

6 References 5, 6, 9-12, 14-16, 23, 26, 28, 31. Back

Received October 12, 1994; accepted November 8, 1994.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
up arrowResults
up arrowDiscussion
*References
 

  1. Lerner DJ, Kannel WB. Patterns of coronary heart disease morbidity and mortality in the sexes: a 26-year follow-up of the Framingham population. Am Heart J. 1986;111:383-390. [Medline] [Order article via Infotrieve]
  2. Coronary heart disease incidence, by sex: United States, 1971–1987. MMWR Morb Mortal Wkly Rep. 1992;41:526-529. [Medline] [Order article via Infotrieve]
  3. Wingard DL, Cohn BA, Kaplan GA, Cirillo PM, Cohen RD. Sex differentials in morbidity and mortality risks examined by age and cause in the same cohort. Am J Epidemiol. 1989;130:601-610. [Abstract/Free Full Text]
  4. Keil JE, Gazes PC, Loadholt B, Tyroler HA, Sutherland S, Gross AJ, Knowles M, Rust PF. Coronary heart disease mortality and its predictors among women in Charleston, South Carolina. In: Eaker E, Packard B, Wenger NK, Clarkson TB, Tyroler HA, eds. Coronary Heart Disease in Women: Proceedings of an NIH Workshop. New York, NY: Haymarket Doyma; 1987:90-98.
  5. Henning R, Lundman T. The Swedish Cooperative Study, Part I: a description of the early stage. Acta Med Scand. 1975;586(suppl): 27-29.
  6. Puletti M, Sunseri L, Curione M, Erba SM, Borgia C. Acute myocardial infarction: sex-related differences in prognosis. Am Heart J. 1984;108:63-66. [Medline] [Order article via Infotrieve]
  7. Kannel WB, Sorlie P, McNamara PM. Prognosis after initial myocardial infarction: the Framingham Study. Am J Cardiol. 1979;44:53-59. [Medline] [Order article via Infotrieve]
  8. Tofler GH, Stone PH, Muller JE, Willich SN, Davis VG, Poole WK, Strauss HW, Willerson JT, Jaffe AS, Robertson T, Passamani E, Braunwald E. Effect of gender and race on prognosis after myocardial infarction: adverse prognosis for women, particularly black women. J Am Coll Cardiol. 1987;9:473-482. [Abstract]
  9. Dittrich H, Gilpin E, Nicod P, Cali G, Henning H, Ross J Jr. Acute myocardial infarction in women: influence of gender on mortality and prognostic variables. Am J Cardiol. 1988;62:1-7. [Medline] [Order article via Infotrieve]
  10. Robinson K, Conroy RM, Mulchay R, Hickey N. Risk factors and in-hospital course of first episode of myocardial infarction or acute coronary insufficiency in women. J Am Coll Cardiol. 1988;11: 932-936.
  11. Fiebach NH, Viscoli CM, Horwitz RI. Differences between women and men in survival after myocardial infarction: biology or methodology? JAMA. 1990;263:1092-1096. [Abstract]
  12. Greenland P, Reicher-Reiss H, Goldbourt U, Behar S. In-hospital and 1-year mortality in 1,524 women after myocardial infarction: comparison with 4,315 men. Circulation. 1991;83:484-491. [Abstract/Free Full Text]
  13. Krumholz HM, Douglas PS, Lauer MS, Pasternak RC. Selection of patients for coronary angiography and coronary revascularization early after myocardial infarction: is there evidence for a gender bias? Ann Intern Med. 1992;116:785-790.
  14. Maynard C, Litwin PE, Martin JS, Weaver WD. Gender differences in the treatment and outcome of acute myocardial infarction: results from the Myocardial Infarction Triage and Intervention Registry. Arch Intern Med. 1992;152:972-976. [Abstract]
  15. Maggioni AP, Maseri A, Fresco C, Franzosi MG, Mauri F, Santoro E, Tognoni G. Age-related increase in mortality among patients with first myocardial infarctions treated with thrombolysis. N Engl J Med. 1993;329:1442-1448. [Abstract/Free Full Text]
  16. Goldberg RJ, Gorak EJ, Yarzebski J, Hosmer DW, Dalen P, Gore JM, Alpert JS, Dalen JE. A communitywide perspective of sex differences and temporal trends in the incidence and survival rates after acute myocardial infarction and out-of-hospital deaths caused by coronary heart disease. Circulation. 1993;87:1947-1953.[Abstract/Free Full Text]
  17. Hammar N, Larsen FF, Sandberg E, Alfredsson L, Theorell T. Time trends in survival from myocardial infarction in Stockholm County 1976-1984. Int J Epidemiol. 1992;21:1090-1096. [Abstract/Free Full Text]
  18. Weinblatt E, Shapiro S, Frank CW. Prognosis of women with newly diagnosed coronary heart disease: a comparison with causes of disease among men. Am J Public Health. 1973;63:577-593. [Free Full Text]
  19. Pardaens J, Lesaffre E, Willems JL, De Geest H. Multivariate survival analysis for the assessment of prognostic factors and risk categories after recovery from acute myocardial infarction: the Belgian situation. Am J Epidemiol. 1985;122:805-819. [Abstract/Free Full Text]
  20. Robinson K, Conroy RM, Mulcahy R, Hickey N. The 15-year prognosis of a first acute coronary episode in women. Eur Heart J. 1992;13:67-69. [Abstract/Free Full Text]
  21. Pohjola S, Siltanen P, Romo M. Five-year survival of 728 patients after myocardial infarction: a community study. Br Heart J. 1980;43:176-183. [Abstract/Free Full Text]
  22. Martin CA, Thompson PL, Armstrong BK, Hobbs MST, De Klerk N. Long-term prognosis after recovery from myocardial infarction: a nine year follow-up of the Perth Coronary Register. Circulation. 1983;68:961-969. [Abstract/Free Full Text]
  23. Becker RC, Terrin M, Ross R, Knatterud GL, Desvigne-Nickens P, Gore JM, Braunwald E. Comparison of clinical outcomes for women and men after acute myocardial infarction. Ann Intern Med. 1994;120:638-645. [Abstract/Free Full Text]
  24. Orencia A, Bailey K, Yawn BP, Kottke TE. Effect of gender on long-term outcome of angina pectoris and myocardial infarction/sudden unexpected death. JAMA. 1993;269:2392-2397. [Abstract]
  25. Berkman LF, Leo-Summers L, Horwitz RI. Emotional support and survival following myocardial infarction: a prospective, population-based study of the elderly. Ann Intern Med. 1992;117: 1003-1009.
  26. Liao Y, Cooper RS, Ghali JK, Szocka A. Survival rates with coronary artery disease for black women compared with black men. JAMA. 1992;268:1867-1871. [Abstract]
  27. Murabito JM, Evans JC, Larson MG, Levy D. Prognosis after the onset of coronary heart disease: an investigation of differences in outcome between the sexes according to initial coronary disease presentation. Circulation. 1993;88:2548-2555. [Abstract/Free Full Text]
  28. White HD, Barbash GI, Modan M, Simes J, Diaz R, Hampton JR, Heikkilä J, Kristinsson A, Moulopoulos S, Paolasso EAC, Van der Werf T, Pehrsson K, Sandøe E, Wilcox RG, Verstraete M, von der Lippe G, Van de Werf F. After correcting for worse baseline characteristics, women treated with thrombolytic therapy for acute myocardial infarction have the same mortality and morbidity as men except for a higher incidence of hemorrhagic stroke. Circulation. 1993;88[pt 1]:2097-2103.
  29. Johansson S, Bergstrand R, Ulvenstam G, Vedin A, Wilhelmsson C, Wedel H, Wilhelmsen L, Aberg A. Sex differences in preinfarction characteristics and long-term survival among patients with myocardial infarction. Am J Epidemiol. 1984;119:610-623. [Abstract/Free Full Text]
  30. Peter T, Harper R, Luxton M, Pennington C, Sloman G. Acute myocardial infarction in women. Med J Aust. 1978;1:189-191. [Medline] [Order article via Infotrieve]
  31. Jenkins JS, Flaker GC, Nolte B, Prince LA, Morris D, Kurz J, Petroski GF. Causes of higher in-hospital mortality in women than in men after acute myocardial infarction. Am J Cardiol. 1994;73: 319-322.
  32. Wong ND, Cupples LA, Ostfeld AM, Levy D, Kannel WB. Risk factors for long-term coronary prognosis after initial myocardial infarction: the Framingham study. Am J Epidemiol. 1989;130: 469-480.
  33. Kannel WB, Sorlie P, McNamara PM. Prognosis after initial myocardial infarction: the Framingham Study. Am J Cardiol. 1979;44: 53-59.
  34. Schmidt SB, Borsch MA. The prehospital phase of acute myocardial infarction in the era of thrombolysis. Am J Cardiol. 1990;65:1411-1415. [Medline] [Order article via Infotrieve]
  35. Alonzo AA. The impact of the family and lay others on the care-seeking during life-threatening episodes of suspected coronary artery disease. Soc Sci Med. 1986;22:1297-1311.
  36. Turi ZG, Stone PH, Muller JE, Parker C, Rude RE, Raabe DE, Jaffe AS, Hartwell TD, Robertson TL, Braunwald E. Implications for acute intervention related to time of hospital arrival in acute myocardial infarction. Am J Cardiol. 1986;58:203-209. [Medline] [Order article via Infotrieve]
  37. Maynard C, Althouse R, Cerquiera M, Olsufka M, Kennedy JW. Underutilization of thrombolytic therapy in eligible women with acute myocardial infarction. Am J Cardiol. 1991;68:529-530. [Medline] [Order article via Infotrieve]
  38. Mukharji J, Rude RE, Poole WK, Gustafson N, Thomas LJ, Strauss HW, Jaffe AS, Muller JE, Roberts R, Raabe DS, Croft CH, Passamani E, Braunwald E, Willerson JT. Risk factors for sudden death after acute myocardial infarction: two year follow-up. Am J Cardiol. 1984;54:31-36. [Medline] [Order article via Infotrieve]
  39. Abbott RD, Donahue RP, Kannel WB, Wilson PWF. The impact of diabetes on survival following myocardial infarction in men versus women: the Framingham Study. JAMA. 1988;260:3456-3460. [Abstract]
  40. Donahue RP, Goldberg RJ, Chen Z, Gore JM, Alpert JS. The influence of sex and diabetes mellitus on survival following acute myocardial infarction: a community-wide perspective. J Clin Epidemiol. 1993;46:245-252. [Medline] [Order article via Infotrieve]
  41. Walden CE, Knopp RH, Wahl PW, Beach KW, Strandness E. Sex differences in the effect of diabetes mellitus on lipoprotein triglyceride and cholesterol concentrations. N Engl J Med. 1984;311: 953-959.
  42. Sackett DL. Bias in analytic research. J Chron Dis. 1979;32:51-63. [Medline] [Order article via Infotrieve]
  43. Ayanian JZ, Epstein AM. Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med. 1991;325:221-225. [Abstract]
  44. Steingart RM, Packer M, Hamm P, Coglianese ME, Gersh B, Geltman EM, Sollano J, Katz S, Moyé L, Basta LL, Lewis SJ, Gottlieb SS, Bernstein V, McEwan P, Jacobson K, Brown EJ, Kukin ML, Kantrowitz NE, Pfeffer MA. Sex differences in the management of coronary artery disease. N Engl J Med. 1991;325: 226-230.
  45. Mark DB, Shaw LK, DeLong ER, Califf RM, Pryor DB. Absence of sex bias in the referral of patients for cardiac catheterization. N Engl J Med. 1994;330:1101-1106. [Abstract/Free Full Text]
  46. Udvarhelyi IS, Gatsonis C, Epstein AM, Pashos CL, Newhouse JP, McNeil BJ. Acute myocardial infarction in the Medicare population: process of care and clinical outcomes. JAMA. 1992;268: 2530-2536.
  47. Chiriboga DE, Yarzebski J, Goldberg RJ, Chen Z, Gurwitz J, Gore JM, Alpert JS, Dalen JE. A community-wide perspective of gender differences and temporal trends in the use of diagnostic and revascularization procedures for acute myocardial infarction. Am J Cardiol. 1993;71:268-273. [Medline] [Order article via Infotrieve]
  48. Harlan W. Cardiovascular disease care for women: service utilization, disability and costs from the National Medical Care Utilization and Expenditure Survey. In: Eaker ED, Packard B, Wenger NK, Clarkson TB, Tyroler HA, eds. Coronary Heart Disease in Women: Proceedings of an NIH Workshop. New York, NY: Haymarket Doyma; 1987.
  49. Maynard C, Althouse R, Olsufka M, Ritchie JL, Davis KB, Kennedy JW. Early versus late hospital arrival for acute myocardial infarction in the Western Washington Thrombolytic Therapy Trials. Am J Cardiol. 1989;63:1296-1300. [Medline] [Order article via Infotrieve]
  50. Kannel WB, Abbott RD. Incidence and prognosis of unrecognized myocardial infarction: an update on the Framingham Study. N Engl J Med. 1984;311:1144-1147. [Abstract]
  51. Chandra V, Szklo M, Goldberg R, Tonascia J. The impact of marital status on survival after an acute myocardial infarction: a population-based study. Am J Epidemiol. 1983;117:320-325. [Abstract/Free Full Text]
  52. Ruberman W, Weinblatt E, Goldberg JD, Chaudhary BS. Psychosocial influences on mortality after myocardial infarction. N Engl J Med. 1984;311:552-559. [Abstract]
  53. Case RB, Moss AJ, Case N, McDermott M, Eberly S. Living alone after myocardial infarction: impact on prognosis. JAMA. 1992;267: 515-519.