Robert J. Goldberg, PhD; Jiang Cui, MD, MS; Barbara Olendzki, RD, MPH; Frederick Spencer, MD; Jorge Yarzebski, MD, MPH; Darleen Lessard, MS; Joel Gore, MD 

Am Heart J.  2006;151(6):1297-1304.  ?2006 Mosby, Inc.

Abstract and Introduction

Abstract

Background: Excess body weight is increasingly being recognized as a major health problem in American men and women. It is unclear, however, whether body weight is associated with the demographic and clinical profile, treatment of, and hospital prognosis after acute myocardial infarction (AMI).
Methods: Nonconcurrent prospective epidemiologic investigation of Worcester (Massachusetts) metropolitan residents hospitalized at all 11 greater Worcester medical centers with validated AMI in 1997, 1999, 2001, and 2003.
Results: A total of 2008 men and 1505 women were hospitalized with confirmed AMI during the 4 study periods. Approximately 41% of men and 29% of women were classified as being overweight (body mass index [BMI] of 25-29.9), whereas 27% of men and 26% of women hospitalized with AMI were considered to be obese (BMI ≥30). Obese men and women were significantly younger than individuals of normal weight. Effective cardiac treatment regimens were less often used in men and women of normal body weight, compared with patients who were overweight or obese. After controlling for several potentially confounding prognostic factors, there were no significant differences in the risk of dying during hospitalization for either overweight or obese men and women, compared with patients of normal body weight. Obese men and women were, however, at greater risk for developing heart failure during the acute hospitalization than men and women of normal weight.
Conclusions: The results of this community study suggest an association between BMI and use of different treatment approaches in patients with AMI. Further examination of the impact of excess body weight on hospital outcomes associated with AMI remains warranted.

Introduction

Obesity is a major public health concern in the United States. The prevalence of excess body weight has increased markedly in the United States over the past 40 years in men and women of all ages, with these undesirable trends expected to continue.^[1-3] Numerous longitudinal studies have demonstrated that excessive body weight is associated with increased morbidity and all-cause mortality, as well as with the development of several major risk factors for coronary artery disease.^[4-10] The results of several large prospective studies have shown that obesity is an important independent risk factor for coronary disease morbidity and mortality in women and in men.^[11,12] Based on these and additional findings, the American Heart Association continues to emphasize the importance of obesity as a major modifiable risk factor for coronary disease and other chronic diseases to practicing clinicians and public health advocates.^[13,14]

Although obesity has been shown to be associated with an increased risk of acute myocardial infarction (AMI), the relation between being overweight and the management, clinical course, and hospital prognosis of men and women with AMI is unclear. The Worcester Heart Attack Study is an ongoing observational study of greater Worcester residents hospitalized with confirmed AMI.^[15-17] Using data from this population-based investigation, we examined the association between excess body weight with the clinical profile, use of different treatment approaches, and hospital outcomes of 3513 men and women hospitalized with AMI in 1997, 1999, 2001, and 2003.

Methods

This study is part of an ongoing investigation that is examining long-term trends in the descriptive epidemiology of coronary heart disease in residents of the Worcester metropolitan area hospitalized with AMI at all 16 greater Worcester hospitals in 1975, 1978, 1981, 1984, 1986, 1988, 1990, 1991, 1993, 1995, 1997, 1999, 2001, and 2003.^[15-17] Fewer hospitals (n = 11) were included in more recent study years because of hospital closures, mergers, or conversion to chronic care or rehabilitation facilities. These periods of investigation were selected based on funding availability and for purposes of examining trends in our principal study outcomes on an approximate alternating yearly basis. The present study was limited to greater Worcester residents hospitalized at all 11 area medical centers in 1997, 1999, 2001, and 2003. This was because information about height and weight was not collected from the review of medical records until this time.

The details of this population-based investigation have been described previously.^[15-17] In brief, the medical records of residents of the Worcester metropolitan area (2000 census estimate = 478000) hospitalized for possible AMI were individually reviewed, and the diagnosis of AMI was validated according to predefined criteria.^[15-17]

Sociodemographic and clinical data were abstracted from the hospital medical records of geographically eligible patients with confirmed AMI by trained study physicians and nurses. Information was collected about patients' age, sex, comorbidities, clinical complications, AMI order (initial vs previous) and type (Q-wave vs non?Q-wave, ST- vs non?ST-segment elevation), and hospital survival status. Data about body height and weight were obtained through the review of information contained in the hospital chart. Among the men, 413 (17.1%) did not have information available to calculate a body mass index (BMI), whereas 335 women (18.2%) were missing data on BMI. Initial heights and weights during hospitalization were recorded. Body mass index was calculated in a standard manner, defined as weight in kilograms divided by the square of height in meters.

Data Analysis

Differences in the distribution of demographic, medical history, and clinical characteristics between male and female patients of normal weight (BMI <25) (referent category), those who were overweight (BMI 25-29.9), and those who were classified as obese (BMI ≥30) were examined through the use of χ² tests and analysis of variance for discrete and continuous variables, respectively. We also further classified obese individuals of both sexes into those with mild/moderate obesity (BMI 30-34.9) and those with severe obesity (BMI ≥35). A logistic multivariable regression analysis was performed to examine the association between BMI and use of selected medical regimens, separately in men and women, while controlling for potentially confounding demographic, medical history, and clinical prognostic factors. The referent group for these analyses was patients with a BMI <25. Similarly, a logistic regression analysis was performed to examine the association between BMI and hospital case-fatality rates, separately in men and women, while controlling for several potentially confounding demographic, medical history, laboratory, and clinical prognostic factors (see table footnotes). Given the magnitude and adverse prognostic impact of heart failure and cardiogenic shock,^[18,19] we also examined the relation between BMI and development of these important clinical end points during hospitalization while controlling for various factors (see table footnotes). Because we anticipated differences in the treatment for patients with AMI according to BMI, additional analyses of our principal hospital end points (death and development of heart failure or cardiogenic shock) were performed controlling for the use of aspirin, angiotensin-converting enzyme (ACE) inhibitors, ß-blockers, lipid-lowering agents, thrombolytics, percutaneous coronary intervention (PCI), and coronary artery bypass grafting (CABG), in addition to various demographic and clinical variables. Unless otherwise stated, the referent group for each of our regression analyses was patients with a BMI <25 within each sex.

Results

A total of 3513 patients with validated AMI in whom data about height and weight were available comprised our study population. Among the 2008 men with AMI, 41.4% were overweight and 27.3% were considered to be obese. Among 1505 women with confirmed AMI, 29.1% were overweight and 25.4% were classified as being obese. In our study sample, only 36 men (1.8%) and 105 women (7.0%) had a BMI <18.5; therefore, we were unable to examine the effects of being considerably underweight on the receipt of effective therapies and hospital outcomes after AMI. On the other hand, 373 men (18.6%) and 234 women (15.5%) from our study sample had BMIs between 30 and 34.9, whereas 175 men (8.7%) and 149 women (9.9%) had BMIs >35, allowing us to examine the effects of mild versus severe obesity on our principal study end points.

Characteristics of Men and Women Hospitalized With AMI According to BMI

Overweight and obese men were significantly younger, less likely to have a history of heart failure or stroke, more likely to have prior diabetes, and were significantly more likely to develop a Q-wave AMI than the referent group of normal-weight men ( Table I ). Levels of serum creatinine and blood urea nitrogen declined with increasing BMI in men, whereas hematocrit values were highest in the most obese patients. In examining differences between the mild/moderate obese (BMI 30-34.9) men, as compared with those with more severe obesity (BMI ≥35), men with more severe obesity were slightly younger and less likely to have a history of stroke.

Overweight and obese women were significantly younger and were more likely to have a history of hypertension or diabetes than the referent group of women of normal body weight ( Table I ). In the obese subgroup of patients, the more severely obese women were younger and were more likely to have had diabetes and heart failure diagnosed in the past than women with mild/moderate obesity ( Table I ). Serum levels of blood urea nitrogen were highest in normal-weight and severely obese women.

Receipt of Hospital Therapies and Coronary Interventional Procedures in Men and Women With AMI According to BMI

Men of normal body weight were the least likely to be treated with effective cardiac medications ( Table II ). These included the use of aspirin, ß-blockers, lipid-lowering agents and thrombolytics. Men of normal weight were also significantly less likely to undergo cardiac catheterization or a PCI than men with higher levels of BMI ( Table II ). Among the obese patients, severely obese men were more likely to be treated with thrombolytics and undergo cardiac catheterization than men with mild/moderate obesity. Women of normal body weight were significantly less likely to be treated with ß-blockers, receive lipid-lowering agents, or undergo cardiac catheterization, PCI, or CABG during hospitalization, compared with overweight and obese women. Women with severe obesity were less likely to undergo a PCI and be treated with ß-blockers than the comparison group of mild/moderately obese women ( Table II ).

Because men and women of different BMIs differed with regard to several factors that might affect the receipt of selected therapies and procedures during the index hospitalization, we examined differences in the receipt of cardiac therapies and coronary interventional procedures, separately in men ( Table III ) and in women ( Table IV ), while controlling for several potentially confounding factors.

Overweight men were significantly more likely to receive aspirin, ß-blockers, and lipid-lowering therapy during hospitalization than men of normal body weight. Overweight men were also significantly more likely to undergo cardiac catheterization and PCI than men of normal weight. Obese men were significantly more likely to be treated with ß-blockers and lipid-lowering agents and undergo a PCI than men of normal weight. In examining differences in the receipt of various medical therapies and interventional procedures between men with varying degrees of obesity, men who were severely obese were more likely to be treated with aspirin and thrombolytics (albeit nonsignificant) than men with mild/moderate obesity ( Table III ). Men with more severe obesity exhibited trends of being more likely to undergo cardiac catheterization, but less likely to undergo PCI or CABG, than men who were mildly/moderately obese.

Overweight women were significantly more likely to be treated with ß-blockers and lipid-lowering agents and were more likely to undergo cardiac catheterization and CABG than the reference group of women of normal weight. Obese women were significantly more likely to receive ß-blockers and lipid-lowering agents and undergo each of the invasive cardiac procedures examined than normal-weight women ( Table IV ). Among the subgroup of obese women, women with severe obesity exhibited trends of being less likely to be treated with aspirin and ß-blockers, but were more likely to be treated with thrombolytics, than women with mild/moderate obesity. Severely obese women were less likely to undergo a PCI than women with mild/moderate obesity. These practice patterns were only slightly attenuated when we further controlled for hospital survival status and development of heart failure and cardiogenic shock during hospitalization in additional regression analyses.

Hospital Outcomes in Men and Women With AMI According to BMI

In men, the risk of developing heart failure, cardiogenic shock, or dying during the acute hospitalization was inversely related to BMI. The incidence rates of heart failure (38.2%, 29.0%, and 29.1%), cardiogenic shock (6.0%, 5.8%, and 3.7%), and death (12.2%, 7.0%, and 5.7%) were consistently higher in normal-weight as compared with overweight and obese men. The rates of heart failure, cardiogenic shock, and hospital death were 29.2% versus 28.6%, 2.7% versus 5.7%, and 5.4% versus 6.9%, respectively, in men with mild/moderate obesity in comparison to those with more severe obesity.

Differences in these outcomes in relation to BMI were less apparent in women. The proportion of normal-weight, overweight, and obese women developing heart failure was 43.7%, 45.2%, and 40.7%; cardiogenic shock, 5.4%, 5.7%, and 4.9%; and dying, 12.6%, 10.8%, and 9.4% during the acute hospitalization was essentially similar. The rates of heart failure (40.6% vs 40.3%), cardiogenic shock (4.3% vs 6.0%), and hospital death (9.8% vs 8.7%) were relatively similar in women of mild/moderate obesity in comparison with those of more severe obesity.

Separate logistic regression analyses were carried out in men and in women to examine the association between desirable or excess body weight and our 3 primary hospital outcomes. After simultaneously controlling for several potentially confounding demographic, medical history, and clinical characteristics, overweight men were not at significantly increased risk for each of the end points examined in comparison with normal-weight men ( Table V ). Severely obese men, however, exhibited statistically nonsignificant trends of being more likely to develop heart failure, develop cardiogenic shock, or die during hospitalization, compared with mild/moderately obese men.

Overweight women exhibited borderline significant trends of being more likely to develop heart failure during the index hospitalization than women of normal weight ( Table VI ). Obese women were significantly more likely to develop heart failure compared with the reference group of normal-weight women. In most of the analyses carried out, the associations between BMI and our 3 hospital outcomes were attenuated after controlling for the receipt of effective cardiac therapies during the index hospitalization.

Discussion

The results of our community-wide investigation demonstrate that overweight and obese men and women differ significantly from their thinner counterparts with regard to several presenting characteristics, most particularly so by their markedly younger age at the time of onset of AMI. Overweight and obese men were more likely to experience an AMI an average of 5 and 10 years earlier, respectively, than their normal-weight counterparts. Increasing BMI was associated with greater use of effective cardiac medications and coronary interventional procedures in both men and women even after controlling for age and other factors that may have affected the receipt of these therapies. After adjusting for several factors, there were no significant differences in the risk of dying during hospitalization in overweight and obese men and women, compared with normal-weight men or women.

Prevalence Rates of Overweight and Obesity Iin Hospitalized Men and Women

Our study of patients hospitalized with AMI demonstrated higher prevalence rates of overweight and obesity than have been observed in several recent national surveys. Using data from the Behavioral Risk Factor Surveillance System, the percentage of overweight persons in the United States has increased from 45% in 1991 to 58% in 2001, and approximately 1 in 5 individuals were classified as being obese during the most recent year of this survey.^[20]

Although we were unable to systematically determine the contributions of an increasing BMI to the development of acute coronary disease, partially because persons experiencing out-of-hospital deaths due to coronary disease were not able to be studied, we were struck by the markedly younger age of men and women with AMI who were overweight or obese. Moreover, we were struck by the relatively large proportion of both men (1 in 11) and women (1 in 10) who were classified as being severely obese at the time of hospitalization for their AMI. Because overweight and obese patients have been shown to have an earlier age of onset of AMI and are at increased risk for recurrence of acute coronary disease,^[21,22] aggressive medical treatment and intervention in these patients remain warranted.

Receipt of Effective Cardiac Medications and BMI

Our study found that, compared with normal-weight patients, overweight and obese men and women were more likely to receive most effective cardiac medications examined, although the magnitude of these associations differed in men and women. Heavier men and women were also more likely to undergo cardiac catheterization and PCI, whereas heavy women, but not men, were considerably more likely to undergo CABG.

The reasons why overweight and obese patients with AMI are treated more aggressively than individuals of normal weight are unclear. Clearly, their younger age and varying prevalence rates of selected comorbidities may have contributed to some of the practice patterns observed. However, differences in the management of patients according to level of BMI persisted even after controlling for these potential confounders. It is possible that physicians may be more aggressive in their treatment for overweight patients because they perceive these patients to be at considerably greater risk for adverse outcomes than patients of normal weight. It is also possible that overweight and obese patients may be treated more intensely because of their high prevalence of several important comorbidities.

Elevated BMI and Adverse Hospital Outcomes

Previous epidemiologic studies have shown an increased risk for AMI and sudden cardiac death in obese individuals,^[11,23] but the outcomes of acute coronary patients of varying weights are inadequately defined. Increased body mass may be a risk factor for recurrent coronary events after AMI and with an increased risk of acute coronary disease in patients with angiographically established disease.^[22,24] In a large study of patients with AMI, increased BMI was associated with better short-term (<6 months), but not long-term, outcomes despite an increased risk of recurrent coronary events.^[25] The mechanisms responsible for the potentially adverse effects of an elevated BMI on the clinical outcomes of patients with AMI are unclear.^[26] Obese patients, in general, are younger and have less extensive coronary artery disease, in comparison with individuals of lower body weight.^[27]

In a study of BMI and overall mortality in critically ill patients, Galanos et al^[28] developed a predictive model for mortality using age, underlying diagnosis, and BMI. The results of this study demonstrated greatest mortality among the thinnest patients after controlling for preadmission weight loss and other factors. These and other data suggest that although greater body weight may precipitate AMI at a younger age, it may offer resilience and enhanced nutritional reserves that are mobilized in response to the acute stress of myocardial infarction and its subsequent hospital course. There is a sizeable variation in the percentage of lean body mass at all strata of BMI, and a low BMI may indicate the presence of preexisting conditions, cognitive impairment, and poor nutritional intake, especially in older patients.^[28,29]

Consistent with the results of most previous studies, we found that overweight and obese men and women were considerably younger than men and women of normal weight. Older age has been consistently shown to be one of the most important factors associated with a worse early prognosis after AMI. However, after controlling for age and other potentially confounding factors, there were no significant differences in the risk of dying during hospitalization in overweight or obese men and women, in comparison with individuals of normal body weight. These and other data suggest that younger age is not the only contributor to the survival advantage of overweight and obese men after development of AMI. Differential survival to the time of seeking acute medical care and eventual hospitalization, according to extent of body weight, may partially explain our observed findings.

Study Strengths and Limitations

Although there are numerous strengths to our multihospital population-based investigation, including the large number of men and women of all ages with validated AMI studied and careful measurement of other possible contributory factors to the associations under study, there are several limitations that need to be considered in interpreting the present findings. Several clinical and risk factor data, such as patient's smoking status and presence of other comorbidities, were either not available or were not collected in the current study, which may have contributed to some of the associations observed. These and additional prognostic factors may adversely impact the hospital outcomes under study and/or receipt of selected therapies and may differ in men and women according to extent of excess body weight. Body mass index may not be the most accurate method for assessing fat mass; other indices, including measurement of waist-hip circumference and skinfold thicknesses, may be used for better accuracy. Recent weight loss secondary to another underlying condition was not measured in the present study, which may have contributed to a poorer prognosis after AMI in lean patients. Body weight and height are generally self-reported and likely differentially reported by men and women, resulting in possible misclassification of our key exposure variable between the 2 sexes.

Conclusions

The results of this study suggest that a large proportion of men and women hospitalized with AMI in the community are overweight. Given the markedly earlier age of onset of AMI in obese men and women and the likely effect of obesity in accelerating the underlying atherosclerotic process, further urgency is provided about the importance of individual and community level approaches to the prevention of excess body weight in both sexes.