OBJECTIVE: To evaluate the relationship between plasma lipid profiles and lipoprotein(a) [Lp(a)] concentrations in diabetic patients, taking into account the Lp(a) phenotype. RESEARCH DESIGN AND METHODS: We included 191 consecutive diabetic outpatients (69 type 1 and 122 type 2 diabetic patients) in a cross-sectional study Serum Lp(a) was determined by enzyme-linked immunosorbent assay, and Lp(a) phenotypes were assessed by SDS-PAGE followed by immunoblotting. The statistical methods included a stepwise multiple regression analysis using the Lp(a) serum concentration as the dependent variable. The lipid profile consisted of total cholesterol, HDL cholesterol, LDL cholesterol, corrected LDL cholesterol, triglycerides, and apolipoproteins AI and B. RESULTS: In the multiple regression analysis, LDL cholesterol (positively) and triglycerides (negatively) were independently related to the Lp(a) concentration, and they explained the 6.6 and 7.8% of the Lp(a) variation, respectively. After correcting LDL cholesterol, the two variables explained 3.8 and 6.4% of the Lp(a) variation, respectively. In addition, we observed that serum Lp(a) concentrations were significantly lower in patients with type IV hyperlipidemia (mean 1.0 mg/dl [range 0.5-17], n = 16) than in normolipidemic patients (6.5 mg/dl [0.5-33.5], n = 117) and in type II hyperlipidemic patients (IIa 15.5 mg/dl [3.5-75], n = 13; IIb 9 mg/dl [1-80], n = 45); P < 0.001 by analysis of variance. CONCLUSIONS: Lp(a) concentrations were directly correlated with LDL cholesterol and negatively correlated with triglyceride levels in diabetic patients. Therefore, our results suggest that the treatment of diabetic dyslipemia may indirectly affect Lp(a) concentrations.
OBJECTIVE: To evaluate the relationship between plasma lipid profiles and lipoprotein(a) [Lp(a)] concentrations in diabeticpatients, taking into account the Lp(a) phenotype. RESEARCH DESIGN AND METHODS: We included 191 consecutive diabetic outpatients (69 type 1 and 122 type 2 diabeticpatients) in a cross-sectional study Serum Lp(a) was determined by enzyme-linked immunosorbent assay, and Lp(a) phenotypes were assessed by SDS-PAGE followed by immunoblotting. The statistical methods included a stepwise multiple regression analysis using the Lp(a) serum concentration as the dependent variable. The lipid profile consisted of total cholesterol, HDL cholesterol, LDL cholesterol, corrected LDL cholesterol, triglycerides, and apolipoproteins AI and B. RESULTS: In the multiple regression analysis, LDL cholesterol (positively) and triglycerides (negatively) were independently related to the Lp(a) concentration, and they explained the 6.6 and 7.8% of the Lp(a) variation, respectively. After correcting LDL cholesterol, the two variables explained 3.8 and 6.4% of the Lp(a) variation, respectively. In addition, we observed that serum Lp(a) concentrations were significantly lower in patients with type IV hyperlipidemia (mean 1.0 mg/dl [range 0.5-17], n = 16) than in normolipidemic patients (6.5 mg/dl [0.5-33.5], n = 117) and in type II hyperlipidemicpatients (IIa 15.5 mg/dl [3.5-75], n = 13; IIb 9 mg/dl [1-80], n = 45); P < 0.001 by analysis of variance. CONCLUSIONS:Lp(a) concentrations were directly correlated with LDL cholesterol and negatively correlated with triglyceride levels in diabeticpatients. Therefore, our results suggest that the treatment of diabetic dyslipemia may indirectly affect Lp(a) concentrations.