‘Metabolically healthy obese’ persons are still at higher CHD risk, thus not so healthy after all
Separate and combined associations of obesity and metabolic health with coronary heart disease: a pan-European case-cohort analysisLassale C, Tzoulake I, Moons KGM et al., - Eur Heart J, 2018: 39: 397–406
Overall and abdominal obesity are important risk factors for coronary heart disease (CHD), and this risk is thought to be partly mediated by cardiometabolic risk factors such as insulin resistance, atherogenic dyslipidemia and type 2 diabetes (T2DM) . The term ‘metabolically healthy obese’ (MHO) refers to a subgroup of obese people with few or no elevated metabolic risk factors that are included in the definition of the metabolic syndrome (MetS) , and it has been suggested that this group is not at elevated CV risk .
Various definitions have been used to study this group, and conflicting evidence exists on whether MHO are at higher CV risk or T2DM. The ESC guidelines for CVD prevention question the existence of the concept . Methodological limitations have led to an unclear picture. The current study therefore aimed to clarify the association between MHO and CHD risk, by using precisely defined outcomes. Data of incident CHD cases recorded during 12.2 years of follow-up from the European Prospective Investigation into Cancer and Nutrition CV disease (EPIC-CVD) case-cohort study  were used. A random subcohort of the large cohort was used as a reference group, rather than matched controls. The subcohort consisted of 10474 participants and there were 7637 incident CHD cases (394 of whom were also in the subcohort). In total, 17733 participants contributed 117829 person-years at risk. The aim was to disentangle the effects of obesity and metabolic health with CHD.
- 15.8% of subcohort participants were obese, 25.6% had MetS and 45.2% of obese participants were MHO.
- A positive log-linear association between BMI and CHD risk was seen, after adjusting for potential confounders (HR per SD: 1.25, 95%CI: 1.19-1.32, P<0.0001). After a near perfect log-linear association from quintile 1 to 4, the association deviated from log-linearity at the highest quintile, to HR: 1.96 (95%CI: 1.66-2.32, P<0.0001), comparing a mean BMI of 32.7 kg/m² with 21.5 kg/m² in the lowest quintile.
- The association between BMI and CHD dampened after adjustment for waist circumference (WC)(HR: 1.06, 95%CI: 0.97-1.15, P=0.20), suggesting the effect of lean mass and peripheral adipose tissue. The association was also attenuated in a model adjusted for intermediate cardiometabolic risk factors (BP, total and HDL-c, diabetes, HR: 1.05, 95%CI: 1.01-1.10, P=0.03).
- WC showed a positive approximately linear association with CHD, after adjustment for BMI (HR: 1.24, 95%CI: 1.10-1.40, P<0.0001). This association was attenuated after adjustment for cardiometabolic factors.
- In a fully adjusted model, compared with normal weight persons without MetS, MHO were at higher risk of CHD (HR: 1.28, 95%CI: 1.03-1.58, P=0.02). All other phenotypes defined by body size and metabolic status also showed a significantly increased risk, up to HR: 2.54 (95%CI; 2.21-2.92, P<0.0001) in metabolically unhealthy obese. MetS strongly positively associated with CHD risk, as normal weight participants with MetS showed an HR of 2.15 (95%CI: 1.79-2.57, P<0.0001).
- There was no interaction of gender for the association between metabolically defined body size-phenotype and CHD (P-interaction=0.63).
This prospective case-cohort study in participants from 8 European countries showed higher CHD risks associated with both general (BMI) and central (WC) obesity, with adjusted analyses suggesting that the effect of WC obesity is stronger. Metabolically healthy overweight and obese individuals showed a higher risk of CHD compared with those with normal weight and without MetS. Metabolically unhealthy individuals in all tested BMI categories were at higher CHD risk than their metabolically healthy counterparts. Although MetS aggravates the elevated CHD risk, these data show that ‘metabolically healthy obese’ is not a benign condition.
Garcia-Moll  reiterates that obesity is a major epidemic, with up to about half of the populations of western countries being overweight or obese. In this context, he calls the concept of a metabolically healthy obese phenotype a sunbeam in a dark landscape of increased risk of T2DM, CHD and mortality. MHO individuals were initially thought to have prognoses similar to those with normal weight. More recent reports with longer follow-up challenge this finding. The carefully carried out study by Lassale and colleagues in a large study cohort that allowed complex statistical analysis and exhaustive adjustment with potential confounders also challenges the concept of MHO.
As acknowledged by the authors, this study was limited by the use of only the most frequently used definitions of MHO and MetS. Moreover, there is no longitudinal follow-up on evolution of weight and metabolic status. Also, the design of the EPIC-CVD study makes that some centers only recruited women, or a high proportion of vegetarians, which may limit the external validity of these finding. Nevertheless, the conclusions of this study that obesity increases CHD risk regardless of metabolic status are relevant. And metabolic abnormalities confer extra CHD risk to both normal weight and overweight/obese individuals. Considering that obesity also increases the risk for other risk factors such as hypertension, T2DM and cardiac conditions, studies with longer follow-up may find further support of the CV risk associated with obesity.