Skip to main content
. Author manuscript; available in PMC: 2011 May 1.
Published in final edited form as: Obesity (Silver Spring). 2009 Oct 29;18(5):1039–1045. doi: 10.1038/oby.2009.343

TABLE 3.

Multivariable-adjusted Backwards Elimination Linear Regression Models for Relations Between Inflammatory Markers & Fata.

Model: 1 2 3 4

Clinical Covariates Onlyb Clinical Covariatesb + Markers P + BMI & WC P + BMI & WC + Abdominal Visceral Fat P + BMI & WC + Additional Covariatesc P
Pericardial Fat
CRP --- 18.0 (13.7-22.2) <0.0001 6.8 (2.6-10.9) <0.01 --- --- 6.1 (1.9-10.3) <0.01
Interleukin-6 --- 4.3 (0.8-7.7) 0.02 --- --- --- --- --- ---
MCP-1 --- 4.4 (1.1-7.6) <0.01 3.4 (0.3-6.4) 0.03 --- --- 3.1 (0.02-6.1) <0.05

Model R2 0.046 0.244 0.338 0.496 0.351

Intrathoracic Fat
CRP --- 25.1 (20.5-29.6) <0.0001 12.3 (7.9-16.6) <0.0001 6.8 (3.3-10.3) 0.0001 11.9 (7.4-16.3) <0.0001
Interleukin-6 --- 4.0 (0.3-7.6) 0.03 --- --- --- --- --- ---
MCP-1 --- 5.0 (1.6-8.5) <0.01 3.9 (0.7-7.1) 0.02 --- --- 3.8 (0.6-7.0) 0.02

Model R2 0.034 0.425 0.503 0.692 0.509
a

Expressed as change in fat volume (cm3 (95% CI)) per inter-quartile increase in log inflammatory marker concentration.

b

Adjusted for age, sex, current smoking, heavy alcohol use, aspirin use, menopausal status and hormone replacement therapy.

c

Adjusted for blood pressure, lipid treatment, total cholesterol/HDL, triglycerides, diabetes mellitus, and cardiovascular disease.