For a long time, several studies have indicated a correlation of waist circumference (WC) with abdominal (subcutaneous and intra-abdominal) fat mass1^,2 and accumulation of visceral adipose tissue.3^,4 Subsequent studies have confirmed that WC, a simple measurement, is the best surrogate marker of visceral adiposity.5^,6 In fact, a panel promoted by the Association for Weight Management and Obesity Prevention (NAASO), the Obesity Society, the American Society for Nutrition, and the American Diabetes Association has encouraged the use of WC in clinical practice.7

It has become evident that WC is more linked to cardiovascular risk factors than body mass index (BMI). A survey assessing 168,000 primary care patients across 63 countries8 has demonstrated that BMI, and particularly WC, are strongly associated with cardiovascular disease and diabetes mellitus. Similarly, a recent population-based Australian survey with 4,487 women aged 20-69 years has concluded that central obesity measures (mainly WC) are more strongly correlated with cardiovascular risk when compared with measures of general obesity like BMI.9

In addition to WC, blood pressure, fasting glucose, HDL-cholesterol, and triglyceride levels have become important allies in clinical practice due to the increasing prevalence of metabolic disorders, such as overweight, obesity, and hypertension. All these risk factors are present in metabolic syndrome (MS), which has become a concern for many health organizations.

Following the World Health Organization (WHO)'s recommendations, the National Cholesterol Education Program Adult Treatment Panel III (NCEP-ATPIII) and the International Diabetes Federation (IDF) have emphasized the importance of fat in the etiology of MS, and WC is a crucial component of MS definition.

In addition, it is interesting to note that the cut-off values for WC proposed by the IDF and NCEP-ATPIII are categorized by gender but are not age-specific, although changes in WC with age have been reported for many years.10

To address this issue, this study aimed to investigate the relationship between WC and individual components of cardiometabolic diseases and to assess the ability of WC to predict cardiometabolic risk factors across age groups in adult women.

This observacional, cross-sectional study was approved by the Research Ethics Commitee of the Federal University of Goiás with protocol number 784.446/2014. The cohort comprised 164 women aged 20-78 years, who responded to an invitation to participate in physical exercise program for women's health, promoted by the university and the municipal administration of the city of Santo Antônio de Goiás (Goiás, Brazil).

A sample size of 164 participants was calculated considering a number of 850 women in this age range (online data published by the Instituto Brasileiro de Geografia e Estatística, IBGE11), with a margin of error of 7% and confidence interval of 95%. When the sample was divided into age groups, the margin of error increased to 15%.

The exclusion criteria comprised women younger than 19 years, pregnant women, and women with a diagnosis of infectious or contagious disease, neurological or cognitive deficit, cardiac disease, or hypertension.

The study included 164 participants with a median age of 44.0 years [IQR13.75]. Median [IQR] of WC, SBP, DBP, glucose, HDL-c and triglycerides in the overall cohort was 94.2 [IQR18.5]cm, 127.0 [IQR 20] mm Hg, 81.0 [IQR 14] mm Hg, 86.0 [IQR 18] mg/dL, 48.0 [IQR 13] mg/dL and 117.5 [IQR 68.5] mg/dL, respectively.

Results of the median or mean ± SD of age, SBP, DBP, glucose, HDL-c and triglycerides values were compared according to WC categories defined by IDF and NCEP-ATP III criteria (Table 1). Compared with normal WC participants, the group with increased WC was older, and showed higher levels of WC, SBP and DBP (all p < 0.05). In contrast, levels of glucose, HDL-cholesterol, and triglycerides showed no difference between the two categories of WC defined by both criteria.

Values of age, SBP, DBP, fasting glucose, HDL-c and triglycerides values in the overall cohort was distributed by age group (Table 2). Significant changes were observed between the three age groups in WC (p = 0.036), SBP (p < 0.001) and DBP (p = 0.011).

Table 3 shows the correlation coefficients of the studied variables with each WC category according to the IDF and NCEP-ATPIII cut-off values. For participants with increased WC, the analysis showed that the WC categorized by IDF criteria had low correlation with DBP (ρ = 0.425, p = 0.000) and a negligible correlation with SBP, HDL-cholesterol (ρ = 0.285, p < 0.01; ρ = -0.270, p < 0.001, respectively). As for the WC defined by the IDF criteria, we observed a low correlation with SBP, DBP (ρ = 0.315, p = 0.001; and ρ = 0.442, p = 0.000, respectively) and a negligible correlation with HDL-cholesterol and triglycerides (ρ = -0.227, p = 0.017; and ρ = 0.225, p = 0.018, respectively), by the NCEP-ATPIII criterion.

Bivariate regression analysis showed that for both criteria, IDF and NCEP-ATPIII, only the variability of SBP, DBP, and HDL-cholesterol levels could be explained by increased WC (Table 4). Among them, DBP showed the highest percentage of variability in the age group > 50 years (27% and 30.5%, respectively). A total of 22.7% and 18.4% in the SBP variability was explained by increased WC in participants aged < 40 years, while 9.8% and 8.5% of the variability in HDL-cholesterol levels was explained by increased WC among participants aged 40-50 years.

When dependent variables were categorized according to the cut-off value of each criterion, the percentage of variability of the risk factor in the altered category, explained by increased WC, was always greater than the variability of the non-categorized variable (as observed in Table 4). Thus, increased WC (³ 80 cm by the IDF or ?γτ; 88 cm by the NCE-ATPIII) correlated only with three health risk factors - SBP, DBP, and HDL-cholesterol. Increased WC explained 63.8% and 63% of the variability of increased SBP (in women aged < 40 years), 36.1% and 35.9% of that of increased DBP (in women > 50 years), and 17% and 18% of that of altered HDL-c (in women aged 40-50 years), respectively for IDF and NCEP-ATPIII.

The women participating in the present study were grouped according to categories of WC based on cut-off values determined by the IDF and NCEP-ATPIII criteria. The purpose of this distribution was to assess the association of WC categories with the following health risk factors: age and levels of SBP, DBP, HDL-cholesterol, fasting glucose, and triglycerides.

The prevalence of increased WC categorized according to the IDF criteria (≥ 80 cm) was superior to that of increased WC categorized according to the NCEP-ATPIII criteria (> 88 cm): 84% versus 68%, respectively. This was already expected, considering the lower cut-off recommended by the IDF compared with the NCEP-ATPIII. Considering that the WC is a crucial criteria for the diagnosis of MS, the higher prevalence of increased WC found in this population of women is concerning.

We also observed that median SBP, DBP, and WC values increased significantly with age. These results are aligned with data from the literature that show a continuous increase in SBP between the ages of 30 and 84 years and over. Although DBP values have a varying pattern with aging, they also increased until the fifth decade.15 In contrast, increases in WC with increasing age are more difficult to evaluate, since this evaluation require longitudinal studies and repeated measures analysis, which are uncommon in cross-sectional studies.16 A 5-year longitudinal cohort study including Australian adults reported an increase in mean WC of 0.46 cm/year.17 In our study, we observed that, in addition to the increase in mean WC with increasing age, mean WC values were above the normal values in all groups (for both IDF and NCEP-ATPIII criteria), corroborating the high prevalence of increased WC in our study.

In the association analysis, we showed a significant correlation between WC and all risk factors evaluated. The same was also observed with increased WC.

Several studies have reported a relationship between WC and risk factors for cardiovascular disease. A classic study, conducted with 70 women aged 23-50 years, reported that WC correlates moderately with plasma triglycerides and HDL-cholesterol levels and weakly with fasting glucose levels (blood pressure was not evaluated in the study).4 Significant correlations between WC and cardiovascular risk factors (HDL-cholesterol, total cholesterol, SBC, and DBP) have also been reported in subjects aged 20-59 years.3

Some authors have reported correlations of WC with indicators of MS. Shen et al.,18 evaluated 1,010 healthy men and women and found weak but significant correlations between WC and each MS component (SBP, DBP, and serum levels of glucose, TG, HDL-cholesterol, and insulin). Another population-based study published in 2002 by Zhu et al.,19 including 9,019 American men and women found that WC correlated weakly but significantly with SBP, DBP, and levels of serum LDL-cholesterol, HDL-cholesterol, and plasma glucose.

A substantial limitation of all the studies mentioned above was the lack of WC categorization, as done in our study. Only a few authors have assessed individuals grouped into normal and increased WC categories. For example, Elbassuoni20 reported an association of increased WC with cardiovascular risk factors (SBP, DBP, fasting glucose, triglycerides, and HDL-cholesterol) among 68 pre-menopausal women with a mean age of 32 years. Also, using previous data from a large cohort representative of the US population (US Department of Health and Human Services, National Center) and gender-specific WC cut-off points according to the NIH guidelines, Janssen et al.,21 showed that the health risks associated with increased WC (> 102 cm for men and > 88 cm for women) in men were limited to overweight individuals, but in women, the risks affected those with normal weight, overweight, and class 1 obesity. This finding emphasizes the importance of incorporating WC measurements into clinical practice.

Considering that in the present study the cut-off values for WC in both IDF and NCEP-ATPIII criteria showed correlations with SBP, DBP, and HLD-cholesterol levels, WC values ≥ 80 cm (which would also include those above 88 cm), could potentially be included as a regular measurement in clinical practice.

On regression analysis, we found that increased WC could explain the variability of high SBP (> 60%), high DBP (> 35%) and low HDL-cholesterol levels (≥ 17%) among young adult women. To the best of our knowledge, this is the first study to report such finding.

In 2010, Stevens et al.,16 pointed out that the inclusion of separate cut-off values by gender was appropriate and that the same is not adequate for age, since the inclusion of age-specific WC cut-off values in adults would require an examination of disease risk and use of different cut-off values for different age groups, compromising the simplicity of this useful health risk indicator.

Limitations of the present study include the small sample size, which does not allow inference of a causal relationship. Future studies with greater sample size should be conducted to confirm our findings.

The results from our study, if confirmed in a larger cohort, emphasize the importance of categorizing WC values by age groups and indicate that two simple measurements - blood pressure and WC - should be performed in young adult women in preventive health programs.

This study showed an association of WC ³ 80 cm with SBP, DBP, and HDL-cholesterol levels in adult women and is in line with population-based studies aimed at simplifying the identification of health risk factors in daily practice. Additionally, the detection of SBP ≥ 130 mmHg and WC ≥ 80 cm in a young adult women patient should trigger further investigation of health risks factors, particularly cardiometabolic ones.