Randomized trials in adults have shown reduced all-cause and cardiovascular mortality on hemodiafiltration (HDF) compared to high-flux hemodialysis (HD), but the mechanisms leading to improved outcomes are not clear. We studied biomarkers of inflammation, oxidative stress, anti-oxidant capacity and endothelial dysfunction in 22 children (13 female, age 8-15 years). All children received HD for at least 3 months, and were then switched to HDF, keeping all dialysis related parameters and dialysis time constant. All the biomarkers of inflammation (f32-microglobulin, IL-6, IL-10, high sensitive C-reactive protein [hsCRP]), oxidative stress (nitrotyrosine, advanced glycation end-products [AGEs], oxidized low density lipoprotein [ox-LDL] and anti-oxidant capacity) and endothelial dysfunction (asymmetric dimethyl arginine [ADMA], symmetric dimethyl arginine [SDMA]), were comparable between incident and prevalent patients on HD, suggesting that even a short dialysis vintage of 3 months on HD increases inflammation and endothelial stress. After 3 months of HDF therapy there was a significant reduction in f32-microglobulin (p<0.001), hCRP, ADMA, SDMA, AGEs, ox-LDL (p<0.01 for all) and an increase in total antioxidant capacity (p<0.001) compared to HD. All children were maintained on the same dialyser, dialysis water quality, dialysis time and blood flow speeds suggesting that improved clearances on HDF led to an improved biomarker profile. Even in children with residual renal function there was a significant reduction in 132 microglobulin, hsCRP, SDMA, ox-LDL and AGEs on HDF compared to HD. Children with a lower blood flow had higher inflammatory status (higher IL-6/IL-10 ratio; p = 0.04, r = -0.43). Children who achieved a higher convective volume (>= median 12.8L/m(2)) had lower ox-LDL (p = 0.02). In conclusion, we have shown that a significant improvement in inflammation, antioxidant capacity and endothelial risk profile is achieved even within a short time (3 months) on HDF compared to HD treatment.