Although obesity is a well-established cardiovascular risk factor, some controversy has arisen with regard to its effect on hospital mortality in patients admitted for acute coronary syndrome.
In addition to being a crucial underlying circumstance in the primary cardiovascular risk factors (CVRFs), obesity is an independent risk factor for cardiovascular illness and mortality [
Increased body mass index (BMI) alters the behavior of adipose tissue, which provides insulin resistance as well as resistance to type-2 diabetes, arterial hypertension, dyslipidemia, and proinflammatory and prothrombotic states, thereby favoring the onset of ischemic cardiopathy [
It would therefore be logical to expect obesity to have a lethal effect on patients who have suffered a coronary event. However, some studies have reported better short- and medium-term prognoses in overweight coronary patients [
In light of this controversy, the goal of this study is to determine the relationship between BMI and intrahospital mortality in patients admitted consecutively for ACS.
All patients admitted consecutively between 2009 and 2010 for ACS were included in the RENACI database of the Working Group on Ischemic Heart Disease and Coronary Care Units of the Spanish Cardiology Society. Data were obtained of the clinical medical reports. A total of 853 patients were initially admitted during the period studied with a discharge diagnosis of unstable angina or myocardial infarction. No information on weight or size was available for 29 of the subjects. Therefore, the final sample consisted of 824 patients.
The demographic characteristics and base anthropometry were recorded for all the patients studied. BMI was calculated as weight in kilograms divided by height in meters squared. BMI was studied as a quantitative and a categorical variable. The patients were divided into three groups: normal (BMI < 25 kg/m2), overweight (BMI = 25–30 kg/m2), and obese (BMI > 30 kg/m2).
We analyzed various clinical variables including the classic CVRFs (smoking, diabetes, hypertension, and dyslipidemia), cardiovascular background and treatment prior to the acute event, hemodynamic variables at admission, risk scores according to the TIMI and GRACE scales, electrocardiographic data, analytical parameters and noninvasive cardiologic examination parameters, treatment administered at admission, coronarography examination, and results, as well as percutaneous and surgical revascularization.
Complications occurring during admission and mortality during hospital stays were also evaluated.
All data were analyzed with the one-sample Kolmogorov-Smirnov test to evaluate the normality of their distribution. Continuous variables were expressed as the average and standard deviation and were compared using an unpaired Student’s
A total of 824 patients were studied. The average age of the sample was
BMI had an inverse relation to mortality, that is, to say, higher BMIs were associated with a reduced mortality rate. In the multivariate logistic regression analysis (Table
Multivariate logistic regression.
Variable |
|
Odds ratio | IC 95% |
---|---|---|---|
Age | 0.003 | 1.122 | 1.041–1.208 |
Diabetes | 0.047 | 3.813 | 1.019–14.275 |
BMI | 0.006 | 0.739 | 0.597–0.916 |
GRACE score | 0.002 | 1.037 | 1.013–1.061 |
Killip IV admission | <0.0001 | 88.520 | 8.950–875.517 |
Table
Patients characteristics.
Normal weight | Overweight | Obese |
| |
---|---|---|---|---|
Age |
|
|
|
0.59 |
Heart rate |
|
|
|
0.15 |
Blood pressure |
|
|
|
0.005 |
KILLIP 3-4 | 12 (5.3%) | 30 (6.2%) | 14 (7.8%) | 0.001 |
TIMI score |
|
|
|
0.044 |
GRACE score |
|
|
|
0.34 |
EKG ST depression | 43 (18.9%) | 83 (20%) | 38 (21.1%) | 0.89 |
Elevated troponins | 198 (86.8%) | 329 (79.1%) | 151 (83.9%) | 0.1 |
Creatinine |
|
|
|
0.68 |
Cholesterol |
|
|
|
0.024 |
Blood glucose levels |
|
|
|
0.002 |
EF severely depressed | 16 (7%) | 26 (6.3%) | 16 (8.9%) | 0.66 |
Cardiac catheterization | 174 (76%) | 316 (76%) | 141 (78.3%) | 0.89 |
Two-vessel CD | 34 (14.9%) | 85 (20.4%) | 42 (23.3%) | 0.09 |
Three-vessel CD | 18 (7.9%) | 54 (13%) | 26 (14.4%) | 0.06 |
Coronary angioplasty | 124 (54.4%) | 210 (50.5%) | 89 (49.4%) | 0.85 |
Postinfarction angina pectoris | 23 (10.1%) | 37 (8.9%) | 12 (6.7%) | 0.47 |
Reinfarction | 12 (5.3%) | 18 (4.3%) | 8 (4.4%) | 0.85 |
Cerebrovascular accident | 5 (2.2%) | 1 (0.2%) | 1 (0.6%) | 0.03 |
Worst Killip during admission | 171 (75%) | 341 (82%) | 137 (76.1%) | 0.03 |
Hemorrhage | 4 (1%) | 9 (2.1%) | 1 (0.5%) | 0.24 |
Hospitalization length (days) |
|
|
|
0.24 |
Hospital mortality | 14 (6.1%) | 13 (3.1%) | 8 (4.1%) | 0.19 |
Combined endpoint | 24 (11%) | 37 (8.9%) | 17 (9.4%) | 0.69 |
No difference in antiplatelet, anticoagulant, inotrope, beta blocker, and diuretic treatment at admission. EF: ejection fraction, CD: coronary disease. Combined endpoint: mortality or reinfarction or bleeding or cerebrovascular accident.
Risk factors and relevant history.
Normal weight | Overweight | Obese |
| |
---|---|---|---|---|
Risk factors | ||||
Tobacco | 87 (37.4%) | 136 (32.9%) | 47 (26.4%) | 0.009 |
Diabetes | 61 (26.8%) | 143 (34.4%) | 75 (41.7%) | 0.016 |
Hypertension | 133 (58.3%) | 280 (67.3%) | 129 (71.7%) | 0.063 |
Dyslipemia | 120 (52.6%) | 247 (59.4%) | 112 (62.2%) | 0.24 |
| ||||
CV history | 144 (63.2%) | 275 (66.1%) | 137 (76.1%) | 0.015 |
Infarction | 44 (19.3%) | 101 (24.3%) | 51 (28.3%) | 0.098 |
Heart failure | 13 (5.7%) | 11 (2.6%) | 10 (5.6%) | 0.097 |
Peripheral vascular disease | 24 (10.5%) | 36 (8.7%) | 15 (8.3%) | 0.67 |
Cerebrovascular accident | 16 (7%) | 37 (8.9%) | 9 (5%) | 0.24 |
Renal failure | 23 (10.1%) | 46 (11.1%) | 16 (8.9%) | 0.72 |
| ||||
COPD | 22 (9.6%) | 60 (14.4%) | 34 (18.9%) | 0.027 |
| ||||
Prior PCI | 29 (12.7%) | 46 (11.1%) | 32 (17.8%) | 0.2 |
| ||||
Prior cardiac surgery | 11 (4.8%) | 24 (5.8%) | 7 (3.9%) | 0.89 |
No difference in treatment prior to admittance except greater use of ARA2 and diuretics in overweight and obese subjects than in normal subjects, CV: cardiovascular, COPD: chronic obstructive pulmonary disease, PCI: percutaneous coronary intervention.
Our study shows that BMI is an independent predictor of hospital mortality in that a higher BMI is associated with a lower mortality rate. This finding is correlated with a lower mortality rate in obese and overweight patients compared to patients in the normal range although this result is not statistically significant.
Unlike other authors [
Although BMI is an inverse predictor of mortality in our series, indicating that overweight patients with ACS have better survival rates than normal weight subjects, when subjects are classified into BMI subgroups (normal weight, overweight, and obese), we were unable to demonstrate with statistical significance that patients classified as obese have lower hospital mortality although this tendency does exist. Among other reasons, this may be due to the sample size. Our data is consistent with other data published [
Other authors have not corroborated the protective effect of obesity in ACS [
Several hypotheses have been proposed to explain the inverse relationship between obesity and the prognosis of patients with ischemic cardiopathy. The higher mortality rate in the normal weight group may be due to a higher mortality rate in underweight patients. Some studies [
Another aspect that is often not given appropriate consideration is the role of pharmacological treatment, especially the adverse effects of multiple medicines used in ACS (fibrinolytic agents, antiplatelet drugs, anticoagulants, inotropes, antiarrhythmic agents, diuretics, nitrates, beta blockers, etc.), which could explain a higher mortality rate. Given the difficulty of anthropometric measurement in the first hours of cardiologic emergency treatment, errors in medicinal dosage may arise, primarily in those patients with lower body weight. Nevertheless, unlike other authors, in our series we found that BMI group had no effect on either treatments administered or hemorrhages [
Since BMI cannot differentiate between muscle and fat mass, overweight and obese subjects with coronary disease may have more muscle mass. When BMI is very high and better reflects body adiposity, the obesity paradox disappears [
The main approaches used to measure obesity are BMI, waist-hip ratio, and waist circumference. There is considerable disagreement on which of them is best [
Although the sample consisted of more than 800 patients, classifying the patients into three groups may prevent hospital mortality from being appropriately assessed because its incidence is relatively low. The clinical appraisal of obesity has many limitations, and an alternative might be the combined appraisal of other anthropometric measurements such as waist circumference, which was an unknown variable in our sample. Better anthropometric appraisal using data on abdominal obesity and an understanding of the physical condition of the patients so that they can be more accurately classified into groups could play a critical role in clarifying this paradox.
The relationship between higher BMI and a greater incidence of coronary disease in the general population is well documented, and excess weight should clearly be avoided. However, once coronary heart disease arises, the association between BMI and the prognosis becomes more complex, even paradoxical according to some authors. So the controversy over the predictive value of BMI in patients with ACS remains latent. Our study confirms that BMI is an independent predictor of hospital mortality; a higher BMI is associated with a lower mortality rate. Our overweight and obese patients had a higher incidence of diabetes, cardiovascular history, chronic obstructive pulmonary disease, and lower Killip classes and TIMI scores. However, they did not show an increased mortality rate, which is apparently paradoxical.