OBJECTIVES: Recent clinical data show that a supplemented very low protein diet (sVLPD) allows to postpone dialysis (HD/PD) of about one year without increasing risk of death or hospitalization in CKD5 elderly patients; here we compared benefits and costs of sVLPD vs dialysis in these patients. METHODS: A cost-benefit analysis technique was applied. Direct medical and non-medical benefits and costs were analysed, from the perspective of the Italian National Health Service (NHS). We quantified resource adsorption for dialysis, keto-amino acid supplements, hospitalizations, pharmacological therapy, blood and instrumental tests, medical visits, reimbursement for transportation of patients in HD and for home treatment of patients in PD. Prices and tariffs applied in 2007 were used. Results are expressed as mean  € per patient-year. RESULTS: Data from a subgroup of patients with GFR of 5-7 ml/min/1.73 m2 BSA, participating in a RCT were analysed. Thirty patients were randomized to dialysis (age=77.2 4.6 years; 23 HD, 7 PD); 27 to sVLPD (age=78.6 6.5 years): during the follow up 19 switched to HD and 2 to PD. Follow-up was, in mean, 3.24 years in dialysis, and 3.27 years in sVLPD group, including both diet and following period on dialysis. Erythropoietin (EPO) dose was on average 7,160 IU/patient-week in sVLPD patients, and 8,104 IU/patient-week in dialysis patients.Patients cost €45,209/patient-year on dialysis; €10,786/patient-year during sVLPD and €40,900/patient-year when shifted to dialysis. Dialysis was the cost driver (€26,000/patient-year, 60% of costs), main benefit to NHS derived from saving this costs in the first year (mean period in sVLPD) of treatment. CONCLUSIONS:to our knowledge this is the first study analysing benefits and costs of initiating sVLPD whereas dialysis in CKD5 patients. sVLPD allows to save 1 year per patient of dialysis and to gain €34,000/patient-year on the NHS's perspective.