Prior epidemiologic studies have shown that increasing body mass index (BMI) is associated with higher total cholesterol and low-density lipoprotein cholesterol (LDL). However, these studies were limited by underrepresentation of obese subjects. The aim of this study was to determine whether there is an association between BMI and lipid profiles in a population of patients with a broad spectrum of BMI values. A case-control study was performed involving patients seen at the Cleveland Clinic Florida. Cases (BMI >30 kg/m(2)) were obtained from the obesity surgery database between August 31, 2000, and April 4, 2002. Controls (BMI ≤ 30 kg/m(2)) were obtained from a database of primary care physicians between May 1, 2004, and November 18, 2004. Pearson correlation coefficients were used to assess the relationship between BMI and lipid fractions. Multiple linear regression was performed to assess the independent effect of BMI on lipid levels while adjusting for potential confounders and propensity scores. Six hundred thirty-seven patients were analyzed (females, n = 362, 57%). There was no association between higher BMI and LDL (r = 0.19 p = 0.07), a negative association with high-density lipoprotein cholesterol (HDL; r = 0.45, p < 0.001), and a positive association with the log transformation of triglycerides (r = 0.32, p = 0.005).Higher BMI was inversely associated with HDL and directly associated with TG. BMI showed no significant association with LDL. Although the association between BMI and both HDL and TG may be explained by insulin resistance, the lack of a significant association between BMI and LDL remains an unexpected finding that requires further investigation.