Abstract Background and Aims Vascular access (VA) type used at hemodialysis (HD) initiation varies greatly across countries and dialysis providers, and central venous catheters (CVC) usage is associated with poor survival. Since middle 2019, the Romanian Renal Registry (RRR) imposed compulsory reporting for VA. Therefore, we aimed to evaluate at a national level the VA at HD initiation and its impact on survival. Method We examined the outcome on September 30, 2022, in incident patients starting HD between January 1, 2020 and December 31, 2021. RRR records data on patient age, sex, primary kidney disease (PKD), history of kidney replacement therapy (KRT) and VA, date and cause of death. Results A total of 6153 patients (median age 65 (IQR 54, 72) years, 60% male) started HD in the study period. Glomerulonephritis, diabetic kidney disease (DKD) and tubulointerstitial disease were the main PKD (12, 11 and 9%); the diagnosis was unknown in almost half of the patients (47%). Only 15% of the patients started HD on arterio-venous fistula (AVF); CVC was used in most of the patients: 63% temporary and 22% tunnelled. As in the first 90 days of the study 1674 patients died (74%), recovered the kidney function (17%), received a kidney transplant (2%) or were lost to follow up (7%), the final cohort for survival analysis (i.e. patients that survived more than 90 days) included 4479 patients. The proportion of VA type at HD initiation was similar in this cohort: only 20% AVF, 58% temporary and 23% long-life CVC. A total of 664 (15%) patients died after the first 90 days. They were older, had more often DKD and more frequently started HD on CVC. Cardiovascular disease (56%) was the main cause of death, followed by infectious (14%), cancer (6%), neurological (5%) and gastroenterological (2%) causes. There were no differences in the causes of death between patients who started HD on AVF or CVC. The mean survival time for the entire cohort was 25.3 (95%CI, 25.0-25.6) months. Survivals at 6, 12 and 24 months were 93%, 88% and 80%. In univariate analysis, patients who started HD on AVF had a significantly better mean survival time than those starting on temporary or on tunnelled CVC: 26.6 (95%CI, 26.2-27.1) vs. 25.1 (95%CI, 24.8-25.5) and 24.6 (95%CI, 24.0-25.2) months. In the multivariate Cox proportional hazard model, higher age at HD initiation, DKD, and CVC as initial VA were significantly associated with a poorer survival (Table 1). We reexamined the VA at 6 months after HD initiation - on December 31, 2021, or at the last treatment recorded – in 5654 patients. At six months from HD initiation, 33% were using an AVF, 39% a tunnelled, and 28% a temporary CVC. Conclusion In this nationwide, registry-based study we observed a high rate of CVC usage (80%) among incident HD patients. Also, there was a low conversion from CVC to AVF, with only 33% of patients having a patent AVF six months after HD initiation. HD initiation on a CVC was independently associated with a higher mortality. Furthermore, close to a half of the studied patients had unknown PKD. Taken together, these data point to late nephrology referral of patients in stage 4 CKD, and deficient predialysis nephrology care. A national strategy should be implemented to improve CKD care, aiming to increase both the rate of AVF usage at HD, and the conversion from CVC to AVF in the six months.