Canto JG, Goldberg RJ, Hand MM, Bonow RO, Sopko G, Pepine CJ, Long T. Symptom Presentation of Women With Acute Coronary SyndromesMyth vs Reality. Arch Intern Med. 2007;167(22):2405–2413. doi:10.1001/archinte.167.22.2405
Optimal diagnosis and timely treatment of patients with an acute coronary syndrome (ACS) depends on distinguishing differences between popular “myths” about ischemic symptoms in women and men. Chest pain or discomfort is regarded as the hallmark symptom of ACS, and its absence is regarded as “atypical” presentation. This review describes the presenting symptoms of ACS in women compared with men and ascertains whether women should have a symptom message that is separate or different from that for men.
MEDLINE (1970-2005), bibliographies of articles, and pertinent abstracts were reviewed, focusing on studies of ACS presentation, especially those reporting differences in symptoms by sex. This analysis included 69 of 361 possible studies. Data regarding symptom presentation were recorded.
The published literature lacks standardization in characterizing ACS presentation, data collection, and reporting of symptoms. Approximately one-third of patients in the large cohort studies and one-quarter of patients in the smaller reports and direct patient interviews presented without chest pain or discomfort. The absence of chest pain or discomfort with ACS was noted more commonly in women than in men in both the cumulative summary from large cohort studies (37% vs 27%) and the single-center and small reports or interviews (30% vs 17%).
Women are significantly less likely to report chest pain or discomfort compared with men. These differences, however, are not likely large enough to warrant sex-specific public health messages regarding the symptoms of ACS at the present time. Further research must systematically investigate sex differences in the clinical presentation of ACS symptoms and must include standardized data collection efforts.
Coronary heart disease is the leading cause of morbidity and mortality among women in the United States.1 A woman is 5.5 times more likely to die from heart disease during her lifetime than from breast cancer (National Center for Health Statistics, 2003). Despite the burden of heart disease in women, many lay persons do not recognize heart disease as an important health issue for women. Recent national surveys of the general public indicate that the awareness of heart disease as the leading cause of death in women has increased in recent years.2 Surveyed women often have described myocardial infarction (MI) as a “male problem” and more often attributed the symptoms of MI to other chronic noncardiac conditions.3 Historically, the description of symptoms associated with MI was based on the presentation characteristics of men. Women's symptoms of MI are often labeled as “atypical” and different from the “classic” MI symptoms noted in men4,5 and include a constellation of associated symptoms, usually without chest pain or discomfort.6- 9
Although the focus of this report is on women, accurate recognition of the symptoms of MI in both men and women is crucial for a number of reasons. First, the patient's inability to recognize MI symptoms will inevitably lead to delay in receiving timely therapy, which in turn may lead to a larger infarct and worse prognosis.10 Women, especially those older than 65 years, wait longer than men before seeking treatment for acute coronary syndrome (ACS) symptoms11 primarily because of symptom-related factors, which comprised more than half of the reasons.11 Second, the inability of health care providers to recognize an evolving MI may lead to an incorrect diagnosis and delays in treatment. Young age and the absence of chest discomfort are among the strongest predictors of a missed diagnosis of MI and inappropriate discharge from the emergency department.12- 14
Optimal diagnosis and timely treatment of ACS-MI, especially reducing patient-associated delay in seeking acute medical care, is critical. Both patients and health care providers must distinguish differences between popular myths and the reality about ACS-MI presentations in both women and men. The major objectives of this review were to examine the literature of the presenting symptoms of ACS-MI in women compared with men and, based on this evidence, to ascertain whether sex differences in ACS-MI presentation are significant enough to warrant a public health message for ACS symptom awareness in women that is separate or different from that for men.
We performed a formal search of the electronic literature (MEDLINE) from 1970 to 2005 using the following search terms: chest pain, myocardial infarction, unstable angina, angina pectoris, acute coronary syndromes, heart disease, gender, sex, and women. This search yielded 361 studies published in the English literature of human subjects. The bibliographies of these articles were reviewed, as were the scientific session abstracts in Circulation and Journal of the American College of Cardiology from 2000 through 2005.
Two individuals (J.G.C. and M.M.H.) separately reviewed each article following a prespecified protocol for inclusion in the review. All studies reporting ACS-MI symptom presentation, especially those reporting the frequency of various symptoms in men and women, were included. This included studies from large cohorts or registries, single-center reports, or studies based on personal interviews. We excluded studies that did not primarily focus on symptom presentation or reported on patients with stable coronary artery disease or chronic angina. Also, we excluded studies of ACS prodromal symptoms because data on differences by sex are lacking. We defined ACS as a diagnosis of either acute MI (ST segment elevation or non-ST segment elevation) or unstable angina. We also included studies that reported on the prevalence of silent or unrecognized MI. Of the 361 studies identified, 69 were included in this analysis.
Data extraction was conducted by several investigators (J.G.C. and M.M.H.) and confirmed by another investigator (R.J.G.) following a prespecified protocol. Information on ACS symptoms, such as the presence or absence of chest pain or discomfort and other associated symptoms, was recorded for the overall population and, whenever available, by sex. For each article reviewed, a brief description of the study, patient population, years under study, sample size, demographic characteristics, and whether the study adjusted for possible differences in symptom presentation between men and women and other possible confounders was recorded. The studies were categorized as typical or atypical chest pain or discomfort, ACS symptoms in the absence of chest pain or discomfort, associated symptoms, and silent or unrecognized MI. Unless otherwise specified, chest pain or discomfort refers to the presenting symptoms of ACS. Given the considerable heterogeneity of the studies analyzed, there were no formal meta-analytic techniques performed except for the reporting of simple descriptive statistics.
Women are more likely to present with unstable angina and are less likely to present with MI than men.15- 17 However, in the setting of acute MI, women more commonly present with Q-wave rather than non–Q-wave MI17 and have less marked ST segment elevations.18 Women are more likely to have cardiac chest pain syndromes not directly associated with obstruction of the large epicardial coronary vessels such as Printzmetal angina, syndrome X, and mitral valve prolapse.4 On average, women are almost a decade older than men at the time of their initial MI19- 22 and are more likely to have comorbid conditions such as diabetes mellitus, hypertension, and heart failure.19- 22 Generally, women are less likely to be referred for coronary angiography than men20- 23 and are less likely to receive fibrinolytic therapy, a percutaneous coronary intervention, or coronary artery bypass surgery.20- 22 Among those referred for coronary angiography, women are more likely to have normal angiographic findings.24- 26 Overall, in-hospital and long-term mortality after MI is higher in women than in men, but this may be partially attributed to women's older age and greater number of comorbid conditions at the time of presentation.27 However, in younger women with MI, short- and long-term mortality may be worse than mortality in younger men, even after adjusting for several prognostic characteristics.28,29
The Rose angina questionnaire of the World Health Organization30 is widely used as a standardized tool in screening for angina. The “typical” (textbook) symptoms of myocardial ischemia are well-known. Broadly speaking, these include (1) precordial chest discomfort, pain, heaviness, or fullness, possibly radiating to the arm, shoulder, back, neck, jaw, epigastrium, or other location; (2) symptoms exacerbated by exertion or stress; (3) symptoms that may be relieved by rest or use of nitroglycerin; and (4) symptoms associated with shortness of breath, diaphoresis, weakness, nausea or vomiting, and light-headedness.
“Atypical” chest pain or discomfort has been described as not severe; not prolonged; not classic in presentation; not exactly like prior cardiac symptoms; a burning, sharp, pleuritic, positional pain or discomfort that is reproducible on palpation of the chest wall and localizable by 1 finger; or pain or discomfort in areas of the upper body other than the chest, such as the arms, epigastrium, shoulder, and neck. However, the location of one's chest pain has not been predictive in discriminating which patients will have confirmed ACS.31,32
Herrick,33 in his historical treatise on MI in 1912, reported that MI may occur in the absence of chest pain or discomfort. Symptoms occurring in the ACS setting without chest pain or discomfort have been described34,35 and are frequently labeled as “atypical.” These include unexplained shortness of breath, especially in those who present with ACS and left ventricular dysfunction36; pain or discomfort in other body locations, such as that localized to the arm(s), shoulder, middle back, jaw, or epigastrium; indigestion; nausea or vomiting; diaphoresis; faintness or dizziness; fatigue; generalized weakness; palpitations; cardiac arrhythmias; syncope; cardiac arrest; and central nervous system manifestations with strokelike symptoms including numbness or unexplained confusion. Each of these symptoms may occur alone or in combination and may be experienced by both men and women in the presence or absence of chest discomfort.
Several studies provide data on possible sex differences in ACS presentation. Studies from large cohorts (Table 1)7,8,37- 43 and those based on small reports or personal interviews (Table 2)6,9,27,46- 57,59 suggest that approximately one-third of patients in the large cohort studies and one-quarter of patients in the smaller reports and interviews presented without chest pain or discomfort. The absence of chest pain or discomfort with ACS was noted more commonly in women than in men, both in the cumulative summary of large cohort studies (37% vs 27%) (Table 1) and the single-center and small reports or interviews (30% vs 17%) (Table 2).
Given the large number of patients in the National Registry of Myocardial Infarction report,8 which may potentially dominate the overall figure from the large cohort studies, we repeated the analysis excluding this study, and still almost one-quarter of all women with ACS did not have chest pain or discomfort. In reanalyzing only those studies that included both women and men, the cumulative summary from large cohort studies did not change, but the sex differences noted in single-center and small reports or interviews were attenuated (24% [women] vs 20% [men]).
There are important limitations in interpreting the literature on ACS symptom presentation according to sex. Many studies based the ACS definition on the presence of chest pain or discomfort, thereby excluding patients with ACS without chest pain or discomfort. Others grouped patients with chest pain or discomfort together with those who had pain or discomfort localized to other areas of the upper body in the absence of chest pain or discomfort. Overall, these studies, which are frequently quoted in the literature, reported either small or no sex differences in clinical presentation in the setting of ACS.60- 71 These studies are not included in our overall estimates but are presented in Table 3.
The frequency of other associated symptoms with ACS differs between women and men. Generally, women are more likely to experience middle or upper back pain,6,31,40,46,53,55,57,62 neck pain,31,40,62,72 jaw pain,40,62 shortness of breath,46,53,55,71,72 paroxysmal nocturnal dyspnea,57 nausea or vomiting,40,53,62,65,67 indigestion,55 loss of appetite,46,57 weakness or fatigue,46,53 cough,53,62 dizziness,53 and palpitations55 compared with men. Differences in the frequency of diaphoresis have been inconsistent between men and women.6,40,67 Women appear to have a greater number of associated symptoms as part of their ACS presentation compared with men (average of 2.6 symptoms in women vs 1.8 in men).55
It is well recognized that MI may occur without any symptoms (silent) or that MI may occur without “typical” chest pain or discomfort and thus go unrecognized by patients or their physicians. On average, more than one-quarter of patients with MI in the cohort studies reviewed had silent or unrecognized MI (Table 4),73- 83 and overall, they did not appear to significantly differ by sex (men, 27%, vs women, 26%). In these studies, silent or unrecognized MI was usually determined by the presence or absence of symptoms and electrocardiographic criteria and were generally not validated by use of cardiac biomarkers or other imaging technologies. Among women with unrecognized MI in both the Framingham74 and Reykjavík Cohort Studies,77 approximately one-half lacked identifiable symptoms, whereas the other half had symptoms without chest discomfort that were not recognized by either the patient or their physician.
Advanced age is arguably a more important predictor of ACS presentation in the absence of chest pain or discomfort than sex.8,38,41,43- 45,58,84 Although women are generally older than men at the time of hospitalization for MI, only a limited number of studies have adjusted for age in examining possible sex differences in ACS clinical presentation. In a prospective study of more than 434 000 patients with confirmed MI in the National Registry of Myocardial Infarction 2, increasing age (10-year interval) (odds ratio [OR], 1.28; 95% confidence interval [CI], 1.26-1.28) had a stronger association with MI and the absence of chest discomfort than did female sex (OR, 1.06; 95% CI, 1.04-1.08).8 Among Medicare beneficiaries with unstable angina in the state of Alabama in this primarily elderly cohort (mean age, 72 years), increasing age was the most important predictor of ACS presentation in the absence of chest discomfort, while female sex was not.38
Similarly, in 2 separate studies from the population-based Worcester Heart Attack Study, the overall MI prevalence without chest pain was 20%40 and 33%,41 and occurred with greater frequency in women than in men. However, after multivariate analysis, sex was no longer an important predictor of MI without chest pain40 and was only a marginal predictor in women 75 years and older.41 In the Global Registry of Acute Coronary Events,37 a large multinational observational study involving 14 countries, approximately 1 in 12 patients with ACS presented without chest pain or discomfort, one-quarter of whom were not initially recognized as having ACS. In this study, women and patients older than 75 years were considerably more likely to present without chest pain or discomfort.
The importance of age-specific differences in men and women with ACS must also be considered. In the National Registry of Myocardial Infarction 2,28 women were more likely to present without chest discomfort as the initial MI presentation than men (Table 5). However, after stratifying by age group, there was a progressive sex-specific decline in MI presentation without chest discomfort with advancing age.28 Similarly, women younger than 75 years in the Framingham Heart Study were more likely to have unrecognized MI than men (Table 5); however, this difference was almost 2-fold higher in the younger age groups (<65 years).74
In general, publications comparing sex differences in ACS presentation have several important limitations. First, there is a lack of standardization in data collection and reporting on women's principal or associated ACS symptoms. Thus, a formal meta-analysis was not performed given the potential for biased results in the face of marked heterogeneity of the studies on symptom presentation. Second, a number of studies have only included patients with ACS who presented with chest pain or discomfort and do not provide data on patients with ACS without chest discomfort. Third, chest pain or discomfort is often lumped together with pain or discomfort localized to other areas of the upper body in the absence of chest symptoms. Fourth, hospital records are notoriously imprecise in characterizing the presence and nature of chest discomfort as well as associated symptoms. Furthermore, the patient's history of present illness in hospital records is undoubtedly influenced by the admitting physician's suspicion of ACS. If the physician suspects that the pretest probability for heart disease is generally lower in women than in men, a more “atypical” history may be recorded.85 Fifth, studies that involve interviews are conducted retrospectively and subject to patient recall bias. Sixth, in studies limited to patients with ACS diagnosis, the sensitivity of a particular symptom may be ascertained, but the specificity of a symptom may not be examined. Seventh, the potential association of comorbid conditions, such as diabetes, with symptom presentation has not been discussed given limitations of currently available data, though likely important.
The published literature lacks standardization in characterizing ACS presentation, data collection, and reporting on women's symptoms, which make it difficult to provide definitive conclusions or recommendations. Nevertheless, this review may help provide a framework for public and professional communication about the presenting symptoms of ACS in women as well as men and provide the impetus for clinical trials and epidemiological studies in ACS to create more standardized symptom definitions and data collection. We report that approximately one-third of patients in the large cohort studies and one-quarter of patients in the smaller reports and interviews presented without chest pain or discomfort, with the absence of chest pain or discomfort being noted more commonly in women than in men. However, these potential differences are not likely to be significant enough at this time to warrant a separate or different message for awareness of ACS symptoms in women compared with men.
Our review suggests that age may be an important contributor to any sex-based differences in ACS presentation. Older age is associated with less frequent reporting of chest pain or discomfort as an ACS presenting characteristic. This is important because women with ACS are older than men. However, this point needs to be confirmed in well-designed studies. From a public health perspective, women, by virtue of being older when they present with ACS, may in fact have a different presentation than men. Until more detailed data are available, it is our opinion that the present public health message on symptom presentation should not be altered for women or men.
The range of symptoms highlighted in this review is currently encompassed in the existing MI symptom message promoted by the National Heart, Lung, and Blood Institute (NHLBI) and the American Heart Association (AHA) in the Act in Time to Heart Attack Signs (http://www.nhlbi.nih.gov/actintime. campaign. The campaign, launched in cooperation with the AHA and other partners,86 aims at educating Americans about warning signs of ACS-MI and the importance of calling 911 immediately. The current MI presentation message is given below:
“Chest discomfort. Most heart attacks involve discomfort in the center of the chest that lasts for more than a few minutes, or goes away and comes back. The discomfort can feel like uncomfortable pressure, squeezing, fullness, or pain.”
“Discomfort in other areas of the upper body. Can include pain or discomfort in one or both arms, the back, neck, jaw, or stomach.”
“Shortness of breath. Often comes along with chest discomfort. But it also can occur before chest discomfort.”
“Other symptoms. May include breaking out in a cold sweat, nausea, or light-headedness.”
Further research from well-designed clinical and epidemiological studies must systematically investigate differences in the ACS-MI clinical presentation according to sex, age, and race, including separation of the chief complaint from more ancillary symptoms. This must include standardized data collection efforts and better understanding of the ACS symptom profile and possible sex differences therein. Based on the available evidence, the current AHA and NHLBI MI symptom message, which targets women and men equally, should remain unchanged.
Correspondence: John G. Canto, MD, MSPH, Watson Clinic, 1600 Lakeland Hill Blvd, Lakeland, FL 33805 (firstname.lastname@example.org).
Accepted for Publication: July 7, 2007.
Author Contributions:Study concept and design: Canto, Goldberg, Hand, Bonow, Sopko, and Long. Acquisition of data: Canto and Hand. Analysis and interpretation of data: Canto, Goldberg, Bonow, and Pepine. Drafting of the manuscript: Canto, Bonow, and Pepine. Critical revision of the manuscript for important intellectual content: Canto, Goldberg, Bonow, Sopko, Pepine, and Long. Statistical analysis: Canto and Goldberg. Administrative, technical, and material support: Hand, Bonow, and Long. Study supervision: Canto, Hand, Bonow, and Long.
Financial Disclosure: None reported.