Cardiovascular morbidity and mortality are increased in children on conventional haemodialysis. Haemodiafiltration (HDF) is a newer type of dialysis which has two main types (post-dilution and predilution HDF). Post-dilution HDF is associated with better vascular health, blood pressure and growth in children. Furthermore, pre-dilution HDF is shown to remove a wider spectrum of uremic toxin compared to post-dilution HDF in adults. The investigators need more data to define the optimum dialysis modality for children.

Hypothesis: In children with end stage kidney disease pre-dilution HDF (pre-HDF) achieves improved clearances across a wide molecular weight range compared to post-dilution HDF (post-HDF).

Plan of Investigation: Prevalent patients on thrice weekly HD or HDF who have a single pool Kt/v>1.2 will be randomized in to either study arm A (pre-HDF, post-HDF) or B (post-HDF, pre-HDF) after a conditioning period on post-HDF. Dialysis prescription will be kept constant during study periods including blood flow, dialysate flow, dialysate content, filter type and size. Pre-dialysis and post-dialysis bloods will be drawn at baseline and at the end of each treatment with either modality. Reduction ratios of small and middle molecular weight toxins and protein bound toxins as well as markers of inflammation and nutrition will be compared between two modalities. Assessment of blood pressure (ambulatory blood pressure monitoring) and patient wellbeing (questionnaire) will be performed at the end of each period.

Children will be recruited from paediatric dialysis units in London, Istanbul Heidelberg and Lyon with extension to other centres to be confirmed.

Outcomes: If the results of this study demonstrate better clearance with pre-dilution HDF compared to post-dilution HDF, this will inform a future long-term outcome study comparing different HDF modalities and will contribute to define optimum dialysis modality for children.

Inclusion Criteria:

1. Patients treated in paediatric dialysis centres between 5-20 years of age
2. Receiving maintenance haemodialysis with HD or HDF for the preceding 3 months
3. In post-dilution HDF for at least 4 weeks before the start of the study
4. Patients with stable vascular access (central line or arteriovenous fistula) and no plan to change access modality for the duration of the study
5. Provision for ultrapure water for HDF (defined as containing <0.1 colony forming unit /ml and <0.03 endotoxin unit/ml) documented in the month prior to study start

Exclusion Criteria:

1. Incident dialysis patients (on HD or HDF for < 3 months)
2. Patients with acute infections in the preceding 2 weeks
3. Patients with underlying chronic inflammatory disorders (Including vasculitis)
4. Single pool Kt/V less than 1.2 in the month prior to the study start.
5. Children who have haemoglobin concentrations lower than 10 g/dL

Children on thrice weekly maintanenece HD or HDF