Elderly (i.e.>75 yrs) represent about 40% of incident dialysis patients in the French REIN Registry and are growing population in high income countries. Those elderly patients with CKD have a high risk of dying within the first months of dialysis start or could have a good long-term prognosis suggesting kidney transplantation access. Elderly patients with CKD also have a higher risk of dying than reaching ESRD. Therefore, outcomes of elderly patients with advanced CKD need to be better described and understood to be able to give accurate information to the patients and their relatives and help decision making concerning the treatment strategy including several options (i.e preemptive kidney transplantation, in center or home dialysis and conservative care). In France we have information about patients who started dialysis with French dialysis registry (REIN) but not about elderly patient treated with conservative care. Then we don't have information about description of the therapeutic project in this population and their evolution in a prospective cohort design. Finally we need to identify patients with high risk of early mortality to help shared decision making for better care organisation.

The project of the study is the development of a French multicenter prospective cohort including elderly patients more than 75 years old reaching ESRD. Objectives are as follow:

Description of the characteristics of the population such as, clinical and social conditions, medicine treatment, therapeutic project declared by nephrologist and laboratory value at inclusion.

Description of the evolution of therapeuic project, kidney function and the outcomes defined as death or dialysis or kidney transplantation.

Development of a mortality prognosis tools to help decision making in this population.

At inclusion for each patient data were collected on demographic and clinical conditions (comorbidities and disabilities), mobility (walk without help, need assistance and totally dependent for transfers), way of living (home living or institutionalized), biological data (blood and urine), therapeutic projects about the option for dialysis and the medicals reason to start dialysis. As it is an observational study of current practices, no additional tests or additional visits were mandatory either at inclusion or during the follow-up.

Data on dialysis start and outcomes were extracted from the French National ESRD registry REIN. This linkage was approved by the French Advisory Committee on Information Processing in Material Research in the Field of Health (CCTIRS), the National Commission for Information Technology and Privacy (Commission Nationale de l'Informatique et des Libertes) and ethic committee. The starting dialysis condition included patient's clinical characteristics, the use of a central vascular catheter (CVC), the unplanned condition of dialysis start and the first modality of dialysis (peritoneal or hemodialysis). After dialysis start, patient survival and information about dialysis withdrawal before death was collected.

The number of patients to include was fixed to more than 500 to study at least 15 prognosis factors with estimation of more than 30% of death during follow-up.

Inclusion Criteria:

• Patients were age older than 75 > years,
• Chronic Kidney Disease with eGFR less than 20 mL/min/1.73 m2 (estimated by sMDRD formula)

Exclusion Criteria:

• Acute renal failure and late referral defined by dialysis starting without previous nephrologist follow-up

Elderly patients follow by nephrologist with CKD and eGFR less than 20 mL/min/1.73 m2 (estimated by sMDRD formula)