Prognostic value of adiposity indices for hypertension

The association of obesity with cardiovascular risk has been widely described.1, 2 What is not known so well is the role of obesity as a predictor of incident hypertension. In the manuscript by Park et al3 it is shown that body fat percentage (BF %), as a marker of obesity, is associated with the increased risk of hypertension in a cohort of Korean non‐hypertensive subjects. This association occurs even in non‐obese individuals, with low body mass index (BMI), waist circumference (WC), and waist to hip ratio (WHR).

It has been investigated about the different mechanisms that could explain the relationship between obesity and hypertension, such as the adipocyte dysfunction, insulin resistance, the sympathetic nervous system and the renin‐angiotensin system dysfunction, or the increased intravascular volume and elevated cardiac output.4 It has even been suggested that visceral fat is in the origin of excess adiposity underlying the development of adiposity‐related hypertension, according to observational community studies.5 Lee et al suggested that both quantity and quality of the adipose tissue constitutes a major contributor for the development or prevention of cardiovascular risk factor profiles.6

Several studies have explored the relationship of hypertension with obesity in a cross‐sectional way. Thus, different markers of adiposity have been associated with hypertension in the general population of different geographical areas, both in adults and children, whether male or female.7, 8, 9, 10, 11, 12, 13, 14

One of the first key questions that may arise when dealing with this issue is the role of the different markers of obesity, mainly as predictors. Dozens of papers on obesity use different markers to assess obesity, but it is not clear which one is the most reliable. So far, and in most of the guidelines in force, the BMI and perhaps the WC have been pointed out as the most useful. However, in most cross‐sectional studies comparing different visceral and body fat indices and anthropometric measures as related to hypertension, there is an agreement in that BMI has a lower performance with respect to others 5, 7, 8, 14, 15 and only very few of these studies suggest that other adiposity markers are not better than BMI, such as Taing et al in a study on near 6000 adults from India.12 In this way, an observational study on Singaporean, Malays, and Indians, suggested that, according to values of BF%, the currently recommended cut‐off values of BMI for overweight and obesity should be lowered.16

In longitudinal studies, there is also controversial information as regards the best anthropometric measure or adiposity index to predict incident hypertension. In black South‐African women, George et al17 showed that measures of body composition as assessed by dual‐energy X‐ray absorptiometry (DXA) do not add to hypertension prediction beyond anthropometry. Moliner‐Urdiales18 did not find superiority for any of the explored adiposity or anthropometric measurements to predict incident hypertension. Moreover, in a cohort of Chinese children,19 free fat mass percentage was associated with a higher risk of hypertension. And again in Chinese children, it was shown that BMI was a better predictor than adiposity indices.20 On the contrary, in the Dallas Heart Study where 903 normotensive participants were enrolled in followed for a median of 7 years for the development of hypertension, increased visceral adiposity was robustly associated with incident hypertension after including in the model different markers, one of them BMI.21 Other studies found that body fat percentage is the best predictor of incident hypertension.6, 22

One of the main points of the study by Park et al3 is the long follow‐up period, ie, 10 years, but also the finding that high BF% was significantly associated with the increased risk of incident hypertension even in individuals with low BMI, WC, and WHR. This underlies the importance of globally assess adiposity in general population, independently of “apparent” normal weight. Certainly, this study is based in Korean individuals, and it is obliged to confirm these associations also in populations from other countries and ethnicities.

Other relevant point is the finding of different BF% cut‐offs for men and women. Some other markers of obesity such as WC establish different values for both sexes, but others, such as BMI, do not. The cut‐off values found here, 22.5% and 32.5% for men and women, respectively, are quite similar to those recently reported by Jia et al in Chinese adult population,9 ie, 24.9% and 32.5%. But it is of note that methods to calculate BF% are different in all those studies. Jia et al9 measured BF% using a foot‐to‐foot bioelectrical impedance analysis. Park et al3 used a tetrapolar bioelectrical impedance analysis. Although most investigators report bioelectrical impedance analyses to estimate BF%,5, 17, 19, 20, 23 other authors determined BF% by means of other methods such as a chemical four‐compartment model16 or the skinfold thickness measurement.11

The study by Park et al3 validates previously reported information on the importance of measuring BF% even in non‐obese individuals to detect those who have an increased risk of developing hypertension, in a Korean population. The confirmation of this association should be a point of interest for considering its implementation in routine clinical practice and to be taken into account in future guidelines. However, there is a need to be validated in different subgroups of population and to confirm the different cut‐off values found for men and women, and also ideally by means of the same device. Importantly, it should also encourage researchers to study in depth the mechanisms for linking adiposity with hypertension, as this more close association of hypertension with BF% compared to anthropometric measures can give some insight into the physiopathological mechanisms that increase the cardiovascular risk.

DISCLOSURES

None.