Dark and Light Side of Obesity: Mortality of Metabolically Healthy Obese People

Lev M Berstein


Expert Rev Endocrinol Metab. 2012;7(6):629-632. 

In This Article

Methods, Participants & Main Results

Participants for the analysis were selected from Health Survey for England and Scottish Health Survey, and were linked prospectively to National Health Service mortality data; thus, the analyses were based on a prospective cohort design. During the first household visit, interviewers used computer-assisted interviewing modules. Various self-reported information was collected including smoking, participation in leisure-time physical activity, socioeconomic status and diabetes Types 1 and 2. Trained nurses took samples of nonfasting blood, measured resting blood pressure, body weight and waist circumference, and collected information on prescribed medication; only participants without a history of cardiovascular disease (CVD) at baseline were included. Data on underlying cause of death were collected from the death certificate, and diagnoses were recorded using International Classification of Diseases 9th and 10th revisions.[101,102]

The final analytic sample consisted of 22,203 participants (aged 54.1 years [SD: 12.7 years]; 45.2% men) with BMI >18 kg/m2. Nonobese participants were defined as those with a BMI of 18–29.9 kg/m2 and obesity as a BMI of 30 kg/m2 or greater. Metabolic risk was based on the inclusion of two or more factors such as large waist (>102 cm in men and >88 cm in women), hypertension (BP >130/85 mmHg or use of antihypertensive medication), diabetes diagnosed by a doctor, low-grade inflammation (CRP >3 mg/l), and HDL-cholesterol <1.03 mmol/l in men and <1.30 mmol/l in women. All participants were categorized into the following four groups: metabolically healthy nonobese, metabolically unhealthy nonobese, metabolically healthy obese and metabolically unhealthy obese. Cox proportional hazard models were used to compute hazard ratios of CVD or all-cause death.

Twenty four percent of participants were defined as obese, and within the obese sample, 22% were categorized as metabolically healthy. The latter individuals were comparable with their healthy nonobese counterparts in age, physical activity, HDL-cholesterol, CRP and a low prevalence of diabetes and hypertension, while they displayed higher waist circumference. Regarding end point data, there were 604 CVD and 1868 all-cause deaths, respectively. Compared with the metabolically healthy nonobese subjects, their metabolically healthy obese vis-à-vis were not at elevated risk of CVD mortality, although both the nonobese and obese participants with two or more metabolic abnormalities were at elevated risk; similar results were seen for all-cause mortality without any specific difference between men and women. The results remained largely unchanged when a sensitivity analysis (in participants with a BMI less than 25 kg/m2 and zero metabolic factors) was performed. Also, when an additional set of analyses was conducted using waist circumference instead of BMI, the results were largely replicated and principal conclusions remained the same: metabolically healthy obese were not at risk of CVD and all-cause mortality over 7-year period of follow-up in contrast to their metabolically unhealthy obese counterparts, and thus the presence of described metabolic factors was in this regard more adverse than obesity itself.