Visceral adiposity index and prognosis among patients with ischemic heart failure

ABSTRACT CONTEXT AND OBJECTIVES: The obesity paradox has already been established in relation to heart failure, but it is not known which obesity indicator best reflects this phenomenon. The aim of this study was to evaluate the association between obesity indexes and mortality among patients with heart failure. DESIGN AND SETTING: Cohort study conducted in the Department of Cardiology of Hospital Nossa Senhora da Conceição (Brazil). METHODS: Clinical, demographic, socioeconomic, biochemical and anthropometric data on 116 patients aged 30 to 85 years with a diagnosis of heart failure were evaluated. Arm fat area, body mass index, body surface area, body adiposity index, lipid accumulation product (LAP) and visceral adiposity index (VAI) were calculated. Cox regression was used to perform survival analyses. RESULTS: At baseline, the individuals with ischemic heart failure who remained alive showed higher VAI (3.60 ± 3.71 versus 1.48 ± 1.58; P = 0.04) and a trend towards higher LAP, in comparison with the individuals who died. After an average follow-up of 14.3 months, ischemic heart failure patients who had VAI > 1.21 showed 78% lower risk of death (HR 0.12; 95% CI: 0.02-0.67; P = 0.02) and the Kaplan-Meier survival curves showed better prognosis for these individuals (P = 0.005; log-rank test). CONCLUSION: Our results suggest that VAI is a good predictor of better prognosis among ischemic heart failure patients.


INTRODUCTION
Heart failure is a complex clinical conditional and it is considered to be an important public health problem.Despite significant therapeutic advances, heart failure is highly associated with morbidity and mortality in both developed and developing countries. 13][4] However, studies conducted among ischemic and non-ischemic heart failure patients have suggested that high adiposity indexes may be associated with better prognosis and longer survival than among patients with normal and lower weight. 5,6This association is known as the "obesity paradox" or "reverse epidemiology". 7,8st studies on the obesity paradox use body mass index to assess overweight and detect possible associations with the prognosis. 9,102][13] Studies have been conducted with the aim of evaluating the prognostic value of other adiposity measurements such as waist circumference 14 and body composition assessed through cutaneous skinfolds, 15 dual-energy X-ray absorptiometry (DEXA) or bioelectrical impedance analysis 16,17 among subjects with heart failure.
The lipid accumulation product index 18 and the visceral adiposity index 11 have been proposed as indirect measurements of visceral adipose tissue and have been correlated with cardiometabolic risk, cardiovascular disease and mortality in the general population. 19,20The body adiposity index is an alternative to the body mass index for determining the percentage of body fat, 21 and neck circumference has been characterized as a pathogenic fat depot correlated with visceral adipose tissue and worse metabolic profile. 22However, none of these adiposity measurements has been tested among patients with ischemic or non-ischemic heart failure, within the context of the obesity paradox.
The purpose of this study was to evaluate associations shown by indexes of general and abdominal obesity in relation to mortality, among patients with ischemic and non-ischemic heart failure.

OBJECTIVE
To evaluate associations shown by indexes of general and abdominal obesity in relation to mortality, among patients with ischemic and non-ischemic heart failure.

Setting and study design
A cohort study was conducted between July 2011 and January were collected by trained personnel (physicians, medical students and nutritionists) using a standardized questionnaire.Clinical data were obtained directly from the medical records or after medical evaluation: angina (stable or unstable) was diagnosed by a trained cardiologist in accordance with current guidelines through reviewing the patient's electronic medical records; 23 atrial fibrillation was also defined by a trained cardiologist in accordance with the guidelines and after cardiac auscultation; 24 and blood pressure measurements were made using an aneroid sphygmomanometer with an adequately sized cuff around the arm circumference.Hypertension was defined as a diagnosis of systolic blood pressure ≥ 140 and/or diastolic blood pressure ≥ 90 mmHg, from a previous medical diagnosis or through use of antihypertensive pressure-lowering agents. 24sting blood glucose ≥ 126 mg/dl or glycated hemoglobin ≥ 6.5% or a previous medical diagnosis were used to detect patients with type 2 diabetes mellitus. 25Ejection fraction values, in percentages, were obtained from color Doppler-derived measurements on tissues, by means of transthoracic echocardiography, performed using the GE Vivid 3 system (General Electric, Norway).
Complete blood count, serum creatinine, HDL-cholesterol, serum triglycerides, glycated hemoglobin and fasting blood glucose were obtained from 10 ml venous blood samples from the patient, using standardized techniques at the hospital's certified laboratory.Hematological parameters were detected using a Sysmex XE-5000 analyzer (Sysmex, Kobe, Japan); serum creatinine values were obtained through the kinetic colorimetric Jaffé method; HDL-cholesterol and triglyceride levels were detected by means of the colorimetric enzymatic method; glycated hemoglobin levels were examined through electrochemiluminescence; and fasting blood glucose values were obtained using the modified hexokinase enzymatic method.
After hospital discharge, the patients were contacted every six months by telephone to collect information about their vital status.All patients were contacted at least three times during the follow-up.In cases of death, this was confirmed from the medical records and/or the death certificate, which was brought to the researchers by a member of the family.

RESULTS
During the follow-up period (mean duration of 14.   1).Other anthropometric data did not show any relationship with prognosis in survival analyses.

DISCUSSION
To our knowledge, this is the first study on patients with heart failure to have evaluated non-traditional adiposity indexes as prognostic markers.Moreover, our study showed an inverse relationship between the visceral adiposity index as an abdominal obesity index and mortality, among patients with ischemic heart failure, and we were unable to find associations between general adiposity indexes (body adiposity index and body mass index) and survival in our population.
The obesity paradox has been widely studied, but controversy remains with regard to the best adiposity index for making the prognosis among individuals with ischemic and non-ischemic heart failure.A meta-analysis that evaluated nine studies and included 28,209 individuals with heart failure showed that overweight (body mass index 25-29.9kg/m 2 ) and obesity (body mass index ≥ 30 kg/m 2 ) detected by means of body mass index cutoff points were associated with lower cardiovascular and all-causes mortality. 8Among individuals with heart failure and type 2 diabetes mellitus, body mass index 25-30 kg/m 2 was also correlated with lower mortality. 31On the other hand, obesity class III (body mass index ≥ 40 kg/m²) was correlated with higher mortality rates. 32,33though higher body mass index (≥ 25 kg/m 2 ) has been commonly correlated with a favorable prognosis, a meta-analysis including 12 studies and 6,142 individuals with acute decompensated heart failure showed that higher body mass index was associated with lower mortality, particularly among very elderly people (age > 75 years), individuals with reduced ejection fraction (< 50%), people without type 2 diabetes mellitus and cases of newly diagnosed heart failure.This suggested that aging, heart failure severity/chronicity and metabolism might explain the obesity paradox regarding use of the body mass index as an adiposity measure. 34Moreover, among obese individuals with heart failure, symptoms may appear earlier, and therefore the diagnosis and treatment can take place at earlier stages of heart failure.
We were unable to find any association with body mass index and mortality according to heart failure etiology, and other authors have shown conflicting results.Among chronic heart failure patients, after a mean follow up of 30.7 months (annual event rate 6.0%), the ischemic heart failure overweight subgroup had a lower survival rate than the non-ischemic heart failure overweight subgroup; however, normal weight and obese individuals showed no difference regarding mortality and heart failure etiology. 5Among individuals with chronic heart failure classes II and III, the obesity paradox was only observed in patients with nonischemic heart failure after adjustment for age, sex, New York Heart Association functional class, ejection fraction, comorbidities and treatment. 6These studies were conducted among individuals who were in a stable condition and receiving outpatient treatment, whereas our patients were evaluated in a hospital setting and had higher prevalence of New York Heart Association functional classes III and IV.Thus, it can be suggested that heart failure severity strongly contributes towards mortality independently of heart failure etiology and body mass index values.
Using body mass index as a body fat measurement may lead to misclassification of around 40% of patients with heart failure 35 and the usefulness of body mass index for evaluating body composition in these subjects has been questioned. 10,12Body mass index does not detect the total amount of body fat 12,35,36 and its classification does not vary according to sex, age or ethnicity. 36 well as in cases of heart failure, the obesity paradox has been shown in other cardiovascular disorders such as coronary artery disease. 37Grade I obesity has not been associated with mortality among the general population. 38However, body mass index has a good correlation with lean mass among individuals with coronary artery disease 13 and also in the general population. 12Thus, the poor diagnostic performance of body mass index in discriminating fat and lean mass and also in assessing body fat content could explain the obesity paradox.Furthermore, the prognostic value of the body mass index in specific situations, such as athletes and patients with fluid accumulation, is also limited. 12e effects of body fat detected through other assessment measurements such as cutaneous skinfolds, bioelectrical impedance analysis and DEXA on the obesity paradox have also being  studied, given the limitations of the body mass index.Increased body fat detected through DEXA has been correlated with lower mortality among cases of heart failure. 35However, these techniques are not always available and may be expensive, thus restricting their use in clinical practice.The body adiposity index is a simple alternative to the body mass index for detecting percentage body fat; however, we were unable to detect any association between higher body adiposity index and survival among cases of heart failure.It has been suggested that using the body adiposity index as an indicator of overall adiposity is likely to produce biased estimates of percentage body fat, in which the errors may vary according to sex and level of body fatness.Thus, estimates based on the body adiposity index may not be more accurate than those based on the body mass index. 39e prognostic value of waist circumference as an abdominal obesity index has also been studied among patients with heart failure.Interestingly, higher waist circumference values have been shown to have a protective role regarding mortality rates among patients with heart failure. 7,14Increased waist circumference and normal body mass index may reflect lower levels of physical activity among healthy individuals, with consequently higher levels of fat mass and lower lean mass. 40However, waist circumference has been correlated with both lean and fat mass detected by means of bioelectrical impedance analysis among individuals with heart failure, thus suggesting that both compartments are associated with a better prognosis among these patients. 17ist circumference cannot distinguish subcutaneous adipose tissue from visceral adipose tissue, but it seems to be more strongly associated with subcutaneous fat, especially among overweight individuals. 41Higher levels of visceral fat have been correlated with higher levels of inflammatory cytokines such as tumor necrosis factor-α, which have catabolic effects on lean mass and may contribute towards cardiac cachexia. 42,43However, tumor necrosis factor-α receptors are highly expressed in subcutaneous adipose tissue, 44 and heart failure patients with an enlarged waist may be protected from the negative impact of increased levels of tumor necrosis factor-α through production of higher levels of these receptors, compared with patients with normal weight or who are underweight. 7Since weight lost is associated with lower survival in situations of heart failure, 45 patients with greater severity of heart failure who also have excess body fat (including both the subcutaneous and the visceral compartment) may have greater metabolic reserves and be more resistant to the increased catabolic burden. 46ese individuals and patients with ischemic heart disease have higher levels of lipoproteins such as cholesterol and chylomicron, and this contributes towards higher bacterial lipopolysaccharide levels, thereby stimulating release of proinflammatory cytokines. 47,48Anthropometric and serum lipid values are needed in order to calculate both the lipid accumulation product index and visceral adiposity index.However, the formula for the visceral adiposity index includes two biomarkers that relate to lipoproteins (HDL-cholesterol and triglycerides) and two obesity indexes that are associated with lean mass and subcutaneous fat (waist circumference and body mass index), thus suggesting that instead of being a "visceral adipose function" as proposed originally, 11 the visceral adiposity index might reflect an "excess of weight function" among patients with heart failure.This exploratory study may have been influenced by the small sample size, which will have conferred higher variability, and it may have lacked power to detect some associations, such as the lipid accumulation product index and death.Thus, some of our results may have been due to chance.Furthermore, we did not construct an adjusted multivariable model to detect independent associations between anthropometric indexes (especially the visceral adiposity index) and mortality.Our patients were relatively young and were predominantly in New York Heart Association classes III-IV, without a preserved ejection fraction, which may have limited the extrapolation of our results.On the other hand, the data collection was prospective, which considerably improves data quality.In addition, important prognostic factors were considered in our analysis.

CONCLUSION
We suggest that the visceral adiposity index may be a good predictor of mortality among patients with ischemic heart failure.However, further studies are needed in order to confirm these results.
measurements were obtained.Weight (kg) was measured with the patients wearing light clothes and barefoot, standing on weighing scales with scale divisions of 100 g (Filizola model 31, São Paulo, Brazil), and height was obtained using a stadiometer with scale divisions of 0.1 cm (Tonelli model E120 A; IN Tonelli SA, Santa Catarina, Brazil).Circumferences were measured with a nonelastic measuring tape.Arm circumference was obtained at the midpoint between the acromion and the olecranon, with the arm extended down the side of the body and the palm of the hand facing the thigh.Waist circumference was measured at the larg-
Visceral adiposity index and prognosis among patients with ischemic heart failure Sao Paulo Med J. 18

Table 2 .
There were no significant differences in most of the anthropometric parameters, except for the visceral adiposity index and a trend for lipid accumulation product index values among individuals with ischemic heart failure, which were both higher among survivors than among patients who died (3.60 ± 3.71 versus 1.48 ± 1.58; P = 0.04 for visceral adiposity index; and 70.49 ± 50.73 versus 35.49± 32.62; P = 0.08 for lipid accumulation product index).
3 ± 10.2 months), the general mortality rate was 20.6%.It was 30% (six deaths) among the patients with ischemic heart failure and 18.8% (18 deaths) among those with non-ischemic heart failure.No statistically significant correlation was identified between the heart failure etiology and mortality rate (P = 0.3).tension,angina, atrial fibrillation or type 2 diabetes mellitus, or the number of drugs in use according to survival status (dead or alive).However, there was a trend for serum creatinine levels (P = 0.07).Among the individuals with non-ischemic heart failure, only hemoglobin levels were significantly lower in the group of patients who died (P = 0.01).Visceral adiposity index and prognosis among patients with ischemic heart failure Sao Paulo Med J.

Table 1 .
Demographic and clinical characteristics of the sample [mean ± SD or n (%)]