Mediators of the association between adiposity and coronary heart disease: Decomposing direct and indirect effects

Thursday, 21 August 2014: 11:30 AM
Ballroom C (Dena'ina Center)
Yuan Lu, MS , Harvard School of Public Health, Boston, MA
Kaveh Hajifathalian, MD , Harvard School of Public Health, Boston, MA
Majid Ezzati, PhD , Imperial College London, London, United Kingdom
Eric B Rimm, PhD , Harvard School of Public Health, Boston, MA
Goodarz Danaei, MD , Harvard School of Public Health, Boston, MA
INTRODUCTION:  

Body mass index (BMI) is rising globally and is a major risk factor for coronary heart disease (CHD). It is not clear to what extent the effect of BMI on CHD is mediated through metabolic risk factors (blood pressure, cholesterol, glucose) versus coagulation and inflammation. We applied new analytical methods for mediation analysis to quantify how much of the effects of overweight (BMI ≥ 25 to < 30 kg/m2) and obesity (BMI ≥ 30 kg/m2) on CHD, as compared to normal weight (BMI ≥ 20 to < 25 kg/m2),  are mediated through metabolic risks as well as through coagulation and inflammatory markers.

METHODS:  

We analyzed data from 9 prospective cohorts in the United States with 54,359 participants and 8,978 CHD events for metabolic risk factors, and 4 cohorts with 2,951 events for fibrinogen and inflammatory markers. We decomposed the total effects of overweight and obesity on CHD into natural direct and indirect effects, allowing for an interaction between BMI and its mediators. We then pooled the direct and indirect hazard ratios (HRs) using random-effects models and calculated proportion of excess risk mediated through different mediator combinations.

RESULTS:  

For both overweight and obesity, blood pressure was the most important mediator. The individual mediated proportions were 21% for overweight and 25% for obesity for blood pressure, 10% and 1% for cholesterol and 12% and 16% for blood glucose. When the three mediators were combined the proportion mediated was 40% (95% confidence interval 27 - 57) for overweight and 38% (20 - 53) for obesity. Fibrinogen explained another 6-8% of the excess relative risk and C - reactive protein (CRP) explained 4-8% in addition to three metabolic risk factors. CRP played a more important role for obese individuals (17% (7 - 31)) than overweight (3% (0 - 24)) although this difference was not significant.

CONCLUSIONS:  

Interventions that reduce metabolic risk factors may help dampen the adverse effect of adiposity on cardiovascular diseases. The role of coagulation and inflammatory markers is relatively smaller.