Introduction In an attempt to reduce unnecessary central venous catheters (CVC) dependence in prevalent population, a "CVC removal" programme was initiated in our unit. The objective was to diminish the number of CVC and to analyze the causes of their insertion and maintenance. On 09/01/07, 38 out of 173 prevalent patients on hemodialysis. Methods were CVC dependant (21.9%): 16 incident patients (42,1%) and 22 (57,8%) prevalent patients. All of them were re-evaluated for permanent vascular access utilization, included those previously rejected for surgery. Physical, psychical and vascular aspects regarding each patient were taken into consideration. The following variables were age, Body mass index, time on hemodialysis, Charlson analyzed comorbidity index, DM or vascular comorbidities and number of vascular accesses created prior to CVC placement. In those patients in whom AVF creation was feasible, the following were registered after 18 months of 1.Primary failure rate. 2. Related complications. 3. up follow Percentage of usable AVF or AVG. 4. Percentage of removed CVC. 5. Mortality and hospitalization. The decision of no realization of a new vascular access and? Results maintenance of TVC was performed in 21 patients (55.2%). In 9 of them was decision of the nephrologists and the family. In 11 patients was decision of vascular surgeon due to vascular bed exhaustion. A permanent VA was achieved in 17 patients (44.7%), 50% of them was incident patients and 45% prevalent patients. The primary failure rate was 0%. The TVC was withdrawn in 11/17 (64.7%). Only 4 patients remain without TVC after 18 months of follow-up: 5 patients died (4 with a AVF functioning and 1 VA closure due to steal syndrome) and in 8 patients was mandatory to reintroduce the TVC again. Immediate complications after the surgery were: 1 severe haematoma, 2 infections and 2 severe sequestrations. 3 patients required hospital admission following complications directly related to VA. Mean time of free of TVC was 5.2 months (range: 0.7- 14.3 months). The mortality was 29,4% (5/17) in the group of patients with a new VA and 9,5% (2/21) in the other group(RR 3,19; IC 95%: 0,68-13,98; p: 0,11). No significant differences were observed between patients who transitioned to a permanent vascular access and those who remained TVC dependant in regards to age, gender, BMI, time on dialysis, DM, comorbidity or number of previous VA, Transition from CVC use to hospitalizations or mortality. Conclusions permanent vascular access is possible in prevalent patients in dialysis. Improvement was achieved by a modest amount, and after a prolonged process with gravel complications, some of which proved severe. Need for CVC dependence is confirmed in a high percentage of prevalent patients on dialysis. Comorbidity might have influenced the results. The quality standard should be adequate to reality of current population on hemodialysis.