Hyperlipidemia (HLP), a common complication, is very prevalent in children with primary nephrotic syndrome (PNS). HLP not only significantly increases the cardiovascular risk in adulthood, but also accelerates the progression of renal disease. Proteinuria as the most important pathophysiological change can reduce serum colloid osmotic pressure, which leads to an increase in the synthesis of serum proteins including lipoproteins in the liver for export to the serum. Thus, the severity of lipid abnormalities may correlate with the degree of proteinuria. A total of 378 children with PNS were divided into three groups according to their urinary protein excretion (UPE), group A (50 mg/kg/d ≤ proteinuria <100 mg/kg/d, 125 cases), group B (100 mg/kg/d ≤ proteinuria <200 mg/kg/d, 132 cases) and group C (proteinuria ≥200 mg/kg/d, 121cases). In addition, 200 healthy volunteers with neither allergic nor renal disease between 3 and 14 years of age were recruited as the control group. Fasting serum levels of lipoprotein (a) [Lp(a)], total cholesterol (TC), triglyceride (TG), high density lipoprotein cholesterol (HDL-C), apolipoprotein A1 (apoA1), apoB, and albumin (Alb) were measured. Serum low density lipoprotein cholesterol (LDL-C) was calculated by the Friedewald formula. As expected, when all patients were compared with healthy children in this study, UPE and the serum concentrations of Lp(a), TC, TG, HDL-C, LDL-C, and apoB were higher in the PNS than in the control group (p<0.01), whereas for apoA1/B ratio the opposite was observed (p<0.01). Furthermore, patients in group C exhibited significantly higher Lp(a), TC, TG, LDL-C, and apoB concentrations than those in group A or B (p<0.01), whereas for apoA1/B ratio the opposite was found (p<0.01). The increase in serum lipids was accompanied by a significant augmented UPE in all patients (p<0.05). More specifically, positive correlations were observed between serum levels of TC (r=+0.80, p<0.01), HDL (r=+0.49, p<0.01), LDL (r=+0.79, p<0.01), ApoB (r=+0.62, p<0.01) and log proteinuria in group B; additionally, a negative correlation was observed between apoA1/B ratio and log proteinuria in group B (r=−0.38, p<0.01). However, no correlation of serum lipid profiles with UPE was determined in group A and C, respectively (p>0.05). Serum Alb was negatively correlated with Lp(a) (r=−0.96, p<0.01), TC (r=−0.78, p<0.01), TG (r=−0.78, p<0.01), LDL-C (r=−0.88, p<0.01), apoA1 (r=−0.26, p<0.01), and apoB (r=−0.71, p<0.01), while positively correlated with apoA1/B (r=+0.27, p<0.01) in all nephrotic children. Furthermore, no correlation existed between serum lipid profiles and Alb in group A, B and C, respectively (p>0.05). In Conclusion, secondary dyslipidemia in children with PNS is in parallel with the degree of UPE. There are diverse characteristics of lipid metabolism under different UPE. As for the patients with medium-UPE, positive correlation between serum lipids and proteinuria is presented.