Use of Body Fatness Cutoff Points–Reply–I

Abstract

We appreciate Dr Snitker's interest in our recent editorial1 and his insightful comments regarding body fat (BF) and obesity. Additionally, we are aware of his clinical and research efforts in the area of obesity in children at the University of Maryland School of Medicine. Dr Snitker is correct that currently there is no definitive cutoff for percent BF in defining overweightness or obesity in men or women. In our efforts to simplify the message for readers, we referenced an easily accessible National Institutes of Health (NIH) publication2 that we thought was representative. Generally, we have referenced a major source from the World Health Organization (WHO)3 as opposed to this simple NIH Web site in our research publications from Ochsner Clinic4 and Mayo Clinic.5,6 However, we agree with Snitker that, regardless of the reference, there is no criterion standard for defining overweightness or obesity by the BF method. We previously demonstrated in a cross-sectional design of 13,601 participants (age, 20-80 years; 48% men) from the Third National Health and Nutrition Examination Survey (NHANES III)6 that the mean ± SD body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) in men was 26.6±4.6 and the mean ± SD percent BF was 24.8%±6.0%. Corresponding values in women were 27.6±6.4 and 36.7%±7.4%, respectively. In 6171 participants in NHANES III who had a BMI in the reference range (18.5-24.9), the highest tertile of BF was greater than 23.1% in men and greater than 33.3% in women (labeled as normal weight obesity).7 In this cohort with normal weight obesity defined by elevated BF, the prevalence of metabolic syndrome was 4 times higher than in those with low BF, and these individuals had a higher prevalence of dyslipidemia (men and women) and of hypertension (men) and a 2.2-fold increased risk of cardiovascular (CV) mortality (women) compared with those with low BF. These data suggest that this level of BF is associated with adverse CV risk and prognosis in primary prevention. In secondary prevention, having increased BF (>25% in men and >35% in women) appears to be associated with a protective effect in patients with coronary heart disease (CHD).4 In fact, in patients with CHD4,8 and in those with heart failure,9,10 a higher BF was an independent predictor of event-free survival because of the obesity paradox, which we discussed in our editorial.1 Oreopoulos et al11 in their heart failure study used a Gallager classification of BF based on age, sex, and race and classified patients as underweight, normal, overweight, and obese.12 In preliminary data from our CHD population (n=581) using this Gallager classification, we have found the highest mortality in the underweight and lowest mortality in the overweight, who also had significantly lower mortality than the “normal BF” group during a 3-year follow-up (A.D.S, C.J.L, and R.V.M., unpublished observations, May 1, 2010). The obese group had intermediate mortality, which was significantly lower than the underweight and trended lower than the normal BF group but did not reach statistical significance. Therefore, current research suggests that the obesity cutoff points of BF are in the 23%-25% range in men and 33%-35% range in women (or 30% in women from the NIH,2 as Dr Snitker stated in his letter), which are associated with increased CV risk in primary prevention and reduced risk in patients with established CV disease (obesity paradox). However, we agree that additional research is needed to clearly define optimal BF in patients of both sexes and of various ages, races, and ethnic groups, as well as disease states. Clearly, major organizations such as the WHO, NIH, and major obesity societies should attempt to establish such cutoff points for BF, as was done years ago with BMI.