As part of a cohort study of 10,000 woman for the prevention of breast cancer in Porto Alegre, Southern Brazil, the present study has found alarming results, with 39.3% of obesity and 37.2% of overweight among our study population, which was recruited consecutively from the main cohort. In addition, we encountered a significant percentage of women with type 2 diabetes mellitus, hypertension, dyslipidemia and high WHR. One possible explanation for the high prevalence of overweight in our sample may be the low level of education, as 60% of women had completed just elementary education . The relationship between education and obesity has been described in other studies, as the one carried out with the Danish population, which concluded that education was consistently associated with BMI and obesity . Another explanation for the high prevalence of overweight and obesity could be the lower rate of physical activity, since only 35% of the participants were performing exercises on a regular basis.
We found a positive correlation between serum levels of PAI-1 and serum triglycerides. PAI-1 levels correlate significantly with a variety of adiposity measures (BMI, waist circumference, WHR, total fat, visceral adipose tissue), and also with markers of the metabolic syndrome (inflammatory markers, insulin, glucose, triglycerides and high density lipoprotein) . The activity of PAI-1 is associated with insulin resistance, regardless of serum triglycerides and other potential confounding factors [24–27]. Thus, our findings reinforce the relationship of this adipokine with obesity, especially with abdominal obesity [26–29].
There was also a positive correlation between serum levels of PAI-1 and WHR. It was demonstrated that the increase of visceral adipose tissue (VAT) is directly associated with levels of PAI-1, even after adjusting for BMI . Excess of adipose tissue increases the production of PAI-1, leading to an impairment of the fibrinolytic system . PAI-1 concentration varies according to race, ethnicity and gender, although differences in body composition and in the distribution of adipose tissue may be responsible for much of this variability .
We showed an inverse correlation between BMI and serum levels of VCAM-1. Although obesity is associated with an increase rather than a decrease in cell adhesion molecules expression, including VCAM-1 , the mechanisms that explain the pathophysiological changes which make obesity a risk factor for atherosclerotic diseases, and therefore increase the adhesion molecules, are not yet fully understood. The results of the studies are in fact controversial [33–35].
A study found that elevated levels of VCAM-1 in non-obese patients with polycystic ovary syndrome was associated with insulin resistance, independent of the BMI . We did not measure insulin levels and neither evaluated homeostasis model assessment of insulin resistance index (HOMA IR), which could be helpful in establishing the presence of metabolic syndrome, what could explain variations in VCAM-1 levels.
Adhesion molecules are associated to breast cancer and was shown to be reduced after breast cancer therapy , leading to the point that obesity and metabolic syndrome-type characteristics may possibly increase the likelihood of cancer risk. Therefore, there may be significant differences in ICAM and VCAM in patients with and without breast cancer (regardless of obesity and metabolic syndrome) and this should be investigated in future studies.
Since most women in this study were overweight and had increased WHR we would expect to find a correlation of BMI and WHR with ICAM-1. However, this correlation was not found. The serum levels of ICAM-1 are positively correlated with obesity, in particular, to visceral adipose tissue . On the other hand, it was demonstrated that serum levels of ICAM-1 decreases after weight reduction . Since no data was collected on weight modification, we could not evaluate this association.
Contrary to our expectations, there was no correlation between serum levels of adiponectin and the variables associated to overweight and obesity. According to the literature, serum levels of adiponectin are negatively correlated with BMI in male and female individuals as well as with visceral fat . Patients with type 2 diabetes mellitus, hypertension, and breast cancer, for example, also show reduced peripheral levels of adiponectins [40–42]. Biological variations of adiponectin serum concentrations seem to be associated with differences in gender, age and race . Women have 35% higher plasma adiponectin than male individuals . A possible explanation for this profile is that androgens may reduce the concentrations of this biomarker in men [43, 44].
It seems that the ratio adiponectin:leptin may be more important than the measure of the absolute level of these biomarkers. Recent studies have shown that the adiponectin:leptin ratio is reduced in women with breast cancer, and that BMI presents a negative and a positive correlation, respectively, with serum levels of adiponectin and leptin [45, 46].