The inverse relationship between obesity and adverse cardiovascular outcomes has been coined the 'obesity-paradox'. We sought to determine the relationship between measures of obesity [body mass index (BMI), body surface area (BSA) and body fat percentage (BF%)] and coronary artery calcification (CAC). We retrospectively analyzed patients who underwent CAC using the Agatston score. Baseline demographics were collected and BMI, BSA and BF% were calculated. A two-stage regression modeling approach was used to evaluate the association between BMI, BSA, BF% and Agatston score. Of the 6661 patients [mean age = 57.1 ± 10.8 years, men = 54.3%, median Agatston score = 14 (0, 163)], 0.1% were underweight, 21.3% had normal BMI, 39.1% were overweight and 39.4% were obese. The mean BMI, BSA and BF% were 29.6 ± 6.1 kg/m(2), 1.97 ± 0.25 m(2) and 37 ± 10 %, respectively. There was an independent association between the presence of CAC and BMI (5 kg/m(2) increments) (OR 1.05, CI 1.00-1.11, P = 0.038) and BF% (OR 2.38, CI 1.05-5.41, P = 0.038). Neither BMI categories nor large BSA independently predicted the presence of CAC. BF% predicted the extent of CAC in men but not in women, and higher BF% was associated with higher category of CAC severity in men only. BMI and BF% were independent predictors of the presence of CAC. BF% was associated with the extent of CAC and higher BF% was associated with higher category of CAC severity in men only. These results suggest that further study is needed to better understand the obesity-paradox.