De Schutter, A.; Kachur, S.; Lavie, C. J.; Boddepalli, R. S.; Patel, D. A.; Milani, R. V.
BACKGROUND: Despite the well-known adverse effects of obesity on almost all aspects of coronary heart disease, many studies of coronary heart disease cohorts have demonstrated an inverse relationship between obesity, as defined by body mass index (BMI), and subsequent prognosis: the 'obesity paradox'. The etiology of this and the potential role of inflammation in this process remain unknown.; PATIENTS AND METHODS: We studied 519 patients with coronary heart disease before and after cardiac rehabilitation, dividing them into groups based on C-reactive protein ((CRP) >= 3mg l(-1) and CRPo3mg l(-1) after cardiac rehabilitation). BMI was calculated and body fat was measured using the skin-fold method. Lean mass index (LMI) was calculated as (1 - % body fat) x BMI. The population was divided according to age- and gender-adjusted categories based on LMI and body fat and analyzed by total mortality over > 3-year follow-up by National Death Index in both CRP groups.; RESULTS: During > 3-year follow-up, all-cause mortality was higher in the high inflammation and in the low BMI group. In proportional hazard analysis, even after adjusting for ejection fraction and peak O-2 consumption, higher BMI was associated with lower mortality in the entire population (hazard ratio (HR) 0.38; confidence interval 0.15-0.97) and a trend to lower mortality in both subgroups (HR 0.45 in low CRP, P = 0.24 vs HR 0.32, P = 0.06 in high CRP). High body fat, however, was associated with significantly lower mortality in the high CRP group (HR 0.22; P = 0.03) but not in the low CRP group (HR 0.73; P = 0.64). Conversely, high LMI was associated with markedly lower mortality in the low CRP group (HR 0.04; P = 0.04).; CONCLUSIONS: The obesity paradox has multiple underlying etiologies. Body composition has a different role in different populations with an obesity paradox by BMI. Especially in the subpopulation with persistently high CRP levels, body fat seems protective.