The aim of this thesis was to evaluate different diagnostic and interventional radiological techniques to prevent and to treat the most common problems in hemodialysis arteriovenous fistulas (AVF) and arteriovenous grafts (AVG). Most important problems for AVF and AVG are non-maturation after surgical access creation almost exclusively for AVF, with a primary failure rate ranging between 19% and 43% and dysfunction because of stenosis or thrombosis for both AVF and AVG. In chapter 3 and 4 the role of carbon dioxide (CO2), a non-nephrotoxic and non-allergic gas used as an alternative intravascular contrast medium to iodinated contrast material was studied in preoperative venous mapping. In chapter 3 CO2 has been compared to iodine in upper limb venography. CO2 venography had good interobserver agreement (=0.90), comparable to interobserver agreement for conventional venography (=0.96) and an accuracy, sensitivity and specificity of 95%, 97% and 85% respectively. These results showed that venography with CO2 is an acceptable alternative to conventional venography with iodine, the latter harboring a potential risk of further impairment of renal function when used in patients with chronic kidney disease (CKD). In chapter 4 the impact of CO2 venography on the creation and outcome of AVF was examined in patients without suitable veins on clinical examination. CO2 venography allowed identification of veins amenable for AVF creation. AVF (21 forearm AVF, 80 proximal forearm or elbow AVF) was created in 77% of the patients. Vascular access surgery corresponded with the findings of CO2 venography in 90% of cases. In 8 cases access creation was attempted with a vein considered not suitable on CO2 venography; maturation rate of these cases was significantly lower compared to the maturation rate of the AVF created with suitable veins on CO2. Overall maturation rate of the AVF created was 83.5% with patency rates comparable to other studies on vascular mapping such as duplex ultrasound (US).The diagnostic accuracy of multidetector computed tomography (MDCT) angiography to detect a stenosis ≥ 50% or occlusion in dysfunctional AVF compared to digital subtraction angiography (DSA) was evaluated in chapter 5. Interobserver agreement was excellent for both DSA (=0.86) and MDCT angiography (=0.82). Accuracy, sensitivity, specificity, positive predictive value and negative predictive value for MDCT angiography to detect a stenosis ≥ 50% or occlusion was high. Image quality of MDCT angiography was graded as good or excellent with adequate information in 97.2% of cases. No significant difference in image quality was seen between MDCT angiography and DSA or between MDCT angiography with the patients arm stretched overhead or alongside the body. These results demonstrated that MDCT angiography is a reproducible and reliable technique to detect significant stenosis/occlusion in dysfunctional AVF, with good image quality.In chapter 6, the aim of the study was to determine variables that were predictors of outcome (technical success, dysfunction recurrence and patency) after percutaneous transluminal balloon angioplasty (PTA) in de novo AVF. Technical success was significantly higher in radiocephalic AVF compared to proximal forearm AVF and was negatively correlated with initial stenosis (the higher the initial stenosis grade, the smaller the probability of technical success). Dysfunction recurrence occured in 52.7% and was negatively correlated with technical success and AVF age. Thus technical success of PTA in virgin AVF is affected by AVF type and initial stenosis grade and has an effect on dysfunction recurrence, but not on AVF longevity (no difference in secondary patency). Early dysfunction (dysfunction within 6 months) was negatively correlated with AVF age and positively correlated with diabetes mellitus. Higher AVF age resulted in higher primary and secondary patency rates. Treatment of a thrombosed access (AVF and AVG) by pharmacomechanical means using a brush-catheter and a thrombolytic agent was evaluated in chapter 7. The mean procedure time was 99.2 minutes, and was significantly higher in AVF than in AVG. Anatomic success rate was 100%, clinical success rate was 96.2%. No major complications were seen, while minor complications occurred in 8% of the procedures. Patency rates after one year were acceptable, especially for AVF. In chapter 8 the operator radiation dose during percutaneous interventions on AVF and AVG was examined. Correlation between DAP and radiation dose measured at the hands, legs and eye lens was weak to moderate. Radiation exposure to the hands was significantly higher compared to the legs. In recanalization procedures radiation dose were significantly higher at the hands and at the left leg. Radiation dose was higher to the left hand and leg in interventions on a right-sided access and vice versa. Radiation dose to the eye lens may be higher in systems quipped with a flat-panel detector system versus image intensifier system. In general, operator radiation exposure to the hands, legs and eye lens is relatively low. Recanalization procedures result in higher dose to the hands and left leg. The position of the hands tot the X-ray tube is the main determinant for the dose.Diagnostic and interventional radiology has an important role in the prevention of hemodialysis access dysfunction by providing reliable vascular mapping techniques such as CO2 venography, but also in demonstrating underlying stenosis or thrombosis in a dysfunctional access. (Interventional) Radiology is mandatory in access maintenance and in identifying variables that may influence outcome of the interventional procedure. The potential hazards of ionizing radiation during the interventional procedures are relatively low, yet should be taken into account.