Editor's Choice - Interventions to Achieve Functionality in Newly Created Arteriovenous Fistulas in the Shunt Simulation Study Cohort.

OBJECTIVE: Although observational cohort studies report that interventions to achieve functionality are clinically successful in 85% of patients, the proportion of newly created autologous arteriovenous fistulas that result in functional vascular access typically is only 70 - 80%. To address this discrepancy, the selection and outcomes of interventions to achieve functionality in a multicentre prospective cohort study were analysed. METHODS: The Shunt Simulation Study enrolled 222 patients who needed a first arteriovenous fistula in nine dialysis units in The Netherlands from 2015 to 2018 and followed these patients until one year after access creation. In this observational study, the technical and clinical success rates of interventions to achieve functionality based on lesion and intervention characteristics were analysed and the clinical outcomes of arteriovenous fistulas with assisted and unassisted functionality were compared. RESULTS: For patients who were on dialysis treatment at the end of the study, unassisted fistula functionality was 54% and overall fistula functionality was 78%. Thirty-four per cent of arteriovenous fistulas required an intervention to achieve functionality, 68% of which eventually became functional. Seventy-five per cent of these interventions were percutaneous balloon angioplasties of vascular access stenoses. Patients with clinically successful interventions to achieve functionality had larger pre-operative vein diameters (2.8 ± 1.0 mm vs. 2.3 ± 0.6 mm, p = .036) and less often presented with thrombosed fistulas than patients with unsuccessful interventions (7% vs. 43%, p = .006). Arteriovenous fistulas with assisted functionality had similar secondary patency as fistulas with unassisted functionality (100% and 98% at six months, p = .44), although they required more interventions to maintain function (2.6 vs. 1.7 per year; rate ratio 1.52, 95% CI 1.04 - 2.18, p = .032). CONCLUSION: Interventions to achieve functionality were needed in about a third of newly created arteriovenous fistulas. Most thrombosed fistulas were abandoned, and when selected for thrombectomy rarely reached clinical success. On the other hand, interventions to achieve functionality of patent fistulas had high clinical success rates and therefore can be done repeatedly until the fistula has become functional.

Access flow volume (Qa) and survival in a haemodialysis population: an analysis of 5208 Qa measurements over a 9-year period.

BACKGROUND: The aim of this study was to determine associations between characteristics of arteriovenous access (AVA) flow volume (Qa; mL/min) and 4-year freedom from cardiovascular mortality (CVM) in haemodialysis (HD) patients. METHODS: HD patients who received a primary AVA between January 2010 and December 2017 in one centre were analysed. Initial Qa was defined as the first Qa value obtained in a well-functioning AVA by a two-needle dilution technique. Actual Qa was defined as access flow at a random point in time. Changes in actual Qa were expressed per 3-month period. CVM was assessed according to the European Renal Association-European Dialysis and Transplant Association classification. The optimal cut-off point for initial Qa was identified by a receiver operating characteristics curve. A joint modelling statistical technique determined longitudinal associations between Qa characteristics and 4-year CVM. RESULTS: A total of 5208 Qa measurements (165 patients; 103 male, age 70 ± 12 years, autologous AVA n = 146, graft n = 19) were analysed. During follow-up (December 2010-January 2018, median 36 months), 79 patients (48%) died. An initial Qa <900 mL/min was associated with an increased 4-y CVM risk {hazard ratio [HR] 4.05 [95% confidence interval (CI) 1.94-8.43], P < 0.001}. After 4 years, freedom from CVM was 34% lower in patients with a Qa <900 mL/min (53 ± 7%) versus a Qa ≥900 mL/min (87 ± 4%; P < 0.001). An association between increases in actual Qa per 3-month period and mortality was found [HR 4.48/100 mL/min (95% CI 1.44-13.97), P = 0.010], indicating that patients demonstrating increasing Qa were more likely to die. In contrast, actual Qa per se was not related to survival. CONCLUSIONS: Studying novel AVA Qa characteristics may contribute to understanding excess CVM in HD patients.

Revision using distal inflow for high flow hemodialysis access alters arterial flow characteristics in the dialysis arm.

OBJECTIVE: Revision using distal inflow (RUDI) is currently proposed in patients on hemodialysis having a high flow access (HFA; >2 L/min) or hemodialysis access-induced distal ischemia (HAIDI). However, a recurrence of high flow or hand ischemia is not unusual in the years after RUDI. The aim of the present study was to describe changes in flow characteristics and arterial diameters in the dialysis arm after RUDI for HFA. METHODS: Volume flow, diameter, peak systolic velocity and end diastolic velocity of the brachial artery (BA) were studied 2 and 12 months after RUDI using duplex imaging. In a portion of patients, these characteristics were also assessed at proximal and distal portions of radial and ulnar arteries (proximal forearm radial artery, distal radial artery, ulnar artery, and distal ulnar artery), and in the greater saphenous venous interponate. HFA patients were grouped according to presence of concomitant hand ischemia (HFA-HAIDI) or absence (HFA). RESULTS: Fifteen patients (54 ± 16 year old; 10 males; HFA-HAIDI, n = 6; HFA, n = 9) with a BA HFA (flow volume, 2740 ± 322 mL/min) undergoing RUDI were studied between March 2011 and October 2016 in two Dutch hospitals. After 2 months, flow volume had decreased (1180 ± 189 mL/min), but again increased at 12 months (1520 ± 217 mL/min; P < .001). BA diameters did not change (7.4 ± 0.5 mm), but proximal forearm radial diameters doubled (overall 2.6 ± 0.2 mm to 5.4 ±1.0 mm; P < .001), albeit less prominent in HFA-HAIDI (+80%) than in HFA (+130%; P = .019). During follow-up, the distal ulnar artery peak systolic velocity in HFA-HAIDI (83 ± 10 cm/s) was higher compared with the HFA group (54 ± 5 cm/s; P < .01). Dilatation was not present in the greater saphenous venous interponate. CONCLUSIONS: RUDI for HFA reduction does not reverse BA dilatation, suggesting irreversible structural arterial wall damage possibly contributing to recurrent high flow. Radial artery remodeling is attenuated in HFA patients previously reporting concurrent hand ischemia diminishing the likelihood of high flow recurrence in this subgroup.

The Effect of Geometric Graft Modification on Arteriovenous Graft Patency in Haemodialysis Patients: A Systematic Review and Meta-Analysis.

OBJECTIVE: Arteriovenous grafts (AVGs) are the second best option for haemodialysis access when native arteriovenous fistulae placement is not possible, because they have a lower patency owing to neointimal hyperplasia at the venous anastomosis. This review aimed to evaluate the effect of geometric graft modification to the graft-vein interface on AVG patency. DATA SOURCES: The MEDLINE and Embase (OvidSP) databases were systematically searched for relevant studies analysing the effect of geometrically modified AVGs on graft patency and stenosis formation (last search July 2019). REVIEW METHODS: Data regarding AVG type, patency, and graft outlet stenosis was extracted for further evaluation. Data were pooled in a random effects model to estimate the relative risk of graft occlusion within one year. Follow up, number of patients, and relevant patient characteristics were extracted for the quality assessment of the included studies using Newcastle-Ottawa Scale and Cochrane Risk of Bias Tool. The quality of the evidence was determined according to the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) system. RESULTS: Search strategies produced 2772 hits, of which eight articles met predetermined inclusion criteria. Overall, the included articles had low to moderate risk of bias. In total, 414 expanded polytetrafluoroethylene AVGs (232 geometrically modified and 182 standard) were analysed, comprising two modified AVG types: a prosthetic cuff design (Venaflo®) and grafts with a Tyrell vein patch. Overall, modified grafts did not show a statistically significantly higher one year primary (relative risk [RR] 0.86, 95% confidence interval [CI] 95% 0.64-1.16; GRADE: "low to very low") or secondary patency (RR 0.57, 95% CI 0.32-1.02; GRADE: "low to very low") when compared with standard AVGs. Analysis of prosthetic cuffed grafts (112 patients) separately demonstrated a statistically significantly higher one year primary (RR 0.75, 95% CI 0.61-0.91) and one year secondary patency (RR 0.47, 95% CI 0.30-0.75) compared with standard grafts (92 patients). The results on stenosis formation were inconclusive and inadmissible to quantitative analyses. CONCLUSION: The meta-analysis showed that a prosthetic cuff design significantly improves AVG patency, while a venous cuff does not. Although the heterogeneity and low number of available studies limit the strength of the results, this review shows the potential of grafts with geometric modification to the graft-vein anastomosis and should stimulate further clinical and fundamental research on improving graft geometry to improve graft patency.

Loss of Kidney Function after Endovascular Treatment of Peripheral Arterial Disease.

BACKGROUND: Administration of radiocontrast during endovascular procedures for peripheral arterial disease (PAD) may cause acute kidney injury, which generally recovers with supportive treatment. Long-term effects of endovascular procedures on renal function remain to be investigated. METHOD: This retrospective observational cohort study includes all patients who newly presented to the vascular surgery outpatient clinic with Rutherford class II or III PAD and who were treated with either supervised exercise therapy or endovascular interventions. Changes in estimated glomerular filtration rates (eGFR) after 1 year were compared between the 2 treatment groups. Multivariate regression analysis and propensity score-matched paired analysis were done to correct for potential confounders. RESULTS: One year after treatment, eGFR was reduced by 8.6 mL/min (95% confidence interval [CI], 7.3-9.9, P < 0.001) after endovascular intervention (n = 284) and by 1.7 mL/min (95% CI, 0.9-25, P < 0.001) after supervised exercise therapy (n = 299). After correction for potential confounders, endovascular interventions were associated with 9.2 mL/min (95% CI, 5.9-12.4, P < 0.001) more renal decline compared to exercise therapy. Similar results were found in the propensity score-matched paired analysis. CONCLUSIONS: Endovascular procedures for PAD are associated with clinically relevant and long-term loss of kidney function.

Loss of kidney function in patients with critical limb ischemia treated endovascularly or surgically.

BACKGROUND: Acute kidney injury after the administration of contrast material during endovascular procedures for peripheral arterial disease generally recovers with supportive treatment. However, long-term effects of endovascular procedures for critical limb ischemia on renal function remain to be investigated. METHODS: This retrospective observational cohort study included all patients who newly presented to the vascular surgery outpatient clinic with Rutherford class 4 to class 6 peripheral arterial disease and who were treated with either endovascular or surgical interventions. Changes in estimated glomerular filtration rate (eGFR) after 1 year were compared between the two types of intervention. Multivariate linear regression analysis was done to correct for potential confounders. RESULTS: One year after treatment, eGFR was reduced by 15.0 mL/min (95% confidence interval [CI], 13.1-17.0; P < .001) after endovascular interventions (n = 209) and by 7.6 mL/min (95% CI, 5.1-10.0; P < .001) after surgical therapy (n = 81). Although eGFR rates decreased in both groups, loss of renal function was significantly greater in patients after endovascular interventions (P < .001). Furthermore, 77% of patients receiving endovascular interventions experienced fast renal decline (defined as >4 mL/min loss of eGFR within 1 year) compared with 54% of patients treated surgically (P < .001). After correction for potential confounders, endovascular intervention was associated with 7.4 mL/min (95% CI, 5.4-9.3; P < .001) greater loss of renal function compared with patients treated surgically. CONCLUSIONS: Endovascular procedures for critical limb ischemia are associated with clinically relevant permanent long-term loss of kidney function. This loss of renal function is greater than in comparable patients who were treated with open surgery.

Radiation Exposure during Percutaneous Transluminal Angioplasty for Symptomatic Peripheral Arterial Disease.

BACKGROUND: Endovascular interventions often are the preferred treatment in peripheral arterial disease for both intermittent claudication and chronic critical limb ischemia. Ever more advanced and anatomically complex procedures are being performed, which may increase the radiation exposure of patients and medical staff. We sought to determine the radiation exposure sustained in contemporary interventional radiology practice. METHODS: A single-center observational cohort study of patients undergoing endovascular interventions for symptomatic peripheral arterial disease between January 1st, 2011 and July 31st, 2013 was performed. Radiation dose area product (DAP), peak skin dose (PSD), fluoroscopy time, and volume of contrast media were either measured or calculated and stratified according to patient demographics and anatomic location. RESULTS: In 951 patients, who were treated with endovascular interventions, mean DAP was 108 ± 88 Gy × cm(2) and mean fluoroscopy time was 13 ± 12 min. DAP was significantly higher for aortoiliac procedures (142 ± 103 Gy × cm(2)) compared with femoral (58 ± 73 Gy × cm(2)), below-knee (45 ± 51 Gy × cm(2)) and multilevel (83 ± 90 Gy × cm(2)) procedures (P < 0.001 for all comparisons). PSD was significantly higher for aortoiliac procedures (0.99 ± 0.54 Gy) compared with femoral (0.55 ± 0.55 Gy), below-knee (0.48 ± 0.27 Gy), and multilevel (0.68 ± 0.47 Gy) procedures (P < 0.001 for all comparisons). In contrast, mean fluoroscopy time in aortoiliac interventions was significantly shorter (10 ± 15 min) compared with femoral (15 ± 12 min), below-knee (19 ± 10 min), and multilevel (23 ± 14 min) procedures (P < 0.001 for all comparisons). CONCLUSIONS: Endovascular interventions in patients with symptomatic peripheral arterial disease involve a substantial radiation dose. Although fluoroscopy time is shorter, aortoiliac interventions result in greater radiation exposure than femoropopliteal or infrapopliteal interventions.

Prospective study of a single treatment strategy for local tumescent anesthesia in Muller phlebectomy.

BACKGROUND: Ambulatory Muller phlebectomy for varicose veins can be performed under local anesthesia. However, subcutaneous injection of local tumescent anesthetics may cause discomfort because of acidity of the solution. Addition of sodium bicarbonate lowers the acidity of anesthetic solutions, which might cause less pain. The objective of this study was to study whether alkalinization of the local anesthetic solution with sodium bicarbonate 1.4% decreases perioperative pain during Muller phlebectomies. METHODS: It is a double-blind single-center randomized controlled trial. In all, 101 patients scheduled for ambulatory Muller phlebectomy were randomized to receive either local anesthesia with alkalinized solution (lidocaine 1% and epinephrine in sodium bicarbonate 1.4%) or standard solution (lidocaine 1% and epinephrine in saline 0.9%). Primary outcome was pain during injection of local anesthetics with the use of the Visual Analogue Scale (VAS). Secondary outcomes were peroperative and postoperative pain, use of analgesics, patient satisfaction, return to function, and complications. RESULTS: Patients receiving subcutaneous injection of local anesthetics diluted in sodium bicarbonate 1.4% experienced significantly less pain during injection compared with patients treated with standard anesthetic solution (VAS, 1.75 ± 1.8 vs. 3.55 ± 2.2, P < 0.00). Peroperative and postoperative pain, complication rates, use of analgesics, patient satisfaction, return to function, and complications did not differ between the 2 groups (P > 0.10). CONCLUSIONS: Alkalinization of local anesthetic solution with sodium bicarbonate 1.4% significantly improves patient comfort during injection of local tumescent anesthesia.

Transplantation of kidneys from paediatric DCD donors: a comparison with DBD donors.

BACKGROUND: Although acceptable outcomes have been reported in kidney transplantation from donation after cardiac death (DCD), little is known about kidney transplantation from paediatric DCD. The objective of this study was to compare the outcome of kidney transplantation using paediatric DCD with the outcome of paediatric donation after brain death (DBD). METHODS: Recipients from DCD and DBD donors <18 years of age transplanted in the Netherlands between January 1981 and July 2006 were included in this study. Ninety-one patients were transplanted with kidneys from paediatric DCD donors and 405 patients received grafts from paediatric DBD donors. RESULTS: Grafts from DCD donors were associated with higher percentage of primary non-function (9 versus 2%, P < 0.01) and delayed graft function (48 versus 8%, P < 0.001) compared with DBD donor grafts. Estimated glomerular filtration rate did not differ between groups (57 ± 17 versus 58 ± 21 mL/min at 1 year and 62 ± 14 versus 57 ± 22 mL/min at 5 years, respectively). After correction for confounding variables, the risk of graft failure was higher in the DCD group [hazard ratio 2.440 (95% confidence interval (CI) 1.280-4.650; P = 0.007]. Patient survival, however, was similar between groups [hazard ratio 1.559 (95% CI 0.848-2.867; P = 0.153)]. CONCLUSIONS: Paediatric DCD kidneys represent a valuable source of donor kidneys that has not been fully utilized. Although transplantation of paediatric DCD kidneys is associated with a higher risk of graft failure than transplantation of paediatric DBD kidneys, results are comparable with adult donors. We therefore conclude that paediatric DCD kidneys can be safely added to the donor pool.

Improvements in kidney transplantation from donors after cardiac death.

To reduce the growing waiting list for kidney transplantation, we explored the limits of kidney transplantation from donors after cardiac death by liberally accepting marginal donor kidneys for transplantation. As the percentage of primary non-function (PNF) increased, we evaluated our transplantation program and implemented changes to reduce the high percentage of PNF in 2005, followed by a second evaluation over the period 2006-2009. Recipients of a kidney from a donor after cardiac death between 1998 and 2005 were analyzed, with PNF as outcome measure. During the period 2002-2005, the percentage of PNF increased and crossed the upper control limits of 12% which was considered as unacceptably high. After implementation of changes, this percentage was reduced to 5%, without changing the number of kidney transplantations from donors after cardiac death. Continuous monitoring of the quality of care is essential as the boundaries of organ donation and transplantation are sought. Meticulous donor, preservation, and recipient management make extension of the donor potential possible, with good results for the individual recipient. Liberal use of kidneys from donors after cardiac death may contribute to a reduction in the waiting list for kidney transplantation and dialysis associated mortality.

Guideline recommendations on minimal blood vessel diameters and arteriovenous fistula outcomes.

OBJECTIVE: Clinical guidelines provide recommendations on the minimal blood vessel diameters required for arteriovenous fistula creation but the evidence for these recommendations is limited. We compared vascular access outcomes of fistulas created in agreement with the ESVS Clinical Practice Guidelines (i.e. arteries and veins >2 mm for forearm fistulas and >3 mm for upper arm fistulas) with fistulas created outside these recommendations. METHODS: The multicenter Shunt Simulation Study cohort contains 211 hemodialysis patients who received a first radiocephalic, brachiocephalic, or brachiobasilic fistula before publication of the ESVS Clinical Practice Guidelines. All patients had preoperative duplex ultrasound measurements according to a standardized protocol. Outcomes included duplex ultrasound findings at 6 weeks after surgery, vascular access function, and intervention rates until 1 year after surgery. RESULTS: In 55% of patients, fistulas were created in agreement with the ESVS Clinical Practice Guidelines recommendations on minimal blood vessel diameters. Concordance with the guideline recommendations was more frequent for forearm fistulas than for upper arm fistulas (65% vs 46%, p = 0.01). In the entire cohort, agreement with the guideline recommendations was not associated with an increased proportion of functional vascular accesses (70% vs 66% for fistulas created within and outside guideline recommendations, respectively; p = 0.61) or with decreased access-related intervention rates (1.45 vs 1.68 per patient-year, p = 0.20). In forearm fistulas, however, only 52% of arteriovenous fistulas created outside these recommendations developed into a timely functional vascular access. CONCLUSIONS: Whereas upper arm arteriovenous fistulas with preoperative blood vessel diameters <3 mm had similar vascular access function as fistulas created with larger blood vessels, forearm arteriovenous fistulas with preoperative blood vessel diameters <2 mm had poor clinical outcomes. These results support that clinical decision-making should be guided by an individual approach.

Prediction of time of death after withdrawal of life-sustaining treatment in potential donors after cardiac death*.

OBJECTIVE: Organ donation after cardiac death increases the number of donor organs. In controlled donation after cardiac death donors, the period between withdrawal of life-sustaining treatment and cardiac arrest is one of the parameters used to assess whether organs are suitable for transplantation. The objective of this study was to identify donation after cardiac death donor characteristics that affect the interval between withdrawal of life-sustaining treatment and cardiac death. DESIGN: Prospective multicenter study of observational data. PATIENTS: All potential donation after cardiac death donors in The Netherlands between May 2007 and June 2009 were identified. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of the 242 potential donation after cardiac death donors, 211 entered analysis, 76% of them died within 60 mins, and 83% died within 120 mins after withdrawal of life-sustaining treatment. The median time to death was 20 mins (range 1 min to 3.8 days). Controlled mechanical ventilation, use of norepinephrine, absence of reflexes, neurologic deficit as cause of death, and absence of cardiovascular comorbidity were associated with death within 60 and 120 mins. The use of analgesics, sedatives, or extubation did not significantly influence the moment of death. In the multivariable logistic regression analysis, controlled mechanical ventilation remained a risk factor for death within 60 mins, and norepinephrine administration and absence of cardiovascular comorbidity remained risk factors for death within 120 mins. The clinical judgment of the intensivist predicted death within 60 and 120 mins with a sensitivity of 73% and 89%, respectively, and a specificity of 56% and 25%, respectively. CONCLUSION: Despite the identification of risk factors for early death and the additional value of the clinical judgment by the intensivist, it is not possible to reliably identify potential donation after cardiac death donors who will die within 1 or 2 hrs after life-sustaining treatment has been withdrawn. Consequently, a donation procedure should be initiated in every potential donor.

Hemolysis compromises nitric oxide-dependent vasodilatory responses in patients undergoing major cardiovascular surgery.

BACKGROUND: The hemolytic products cell-free oxyhemoglobin (FHb) and arginase-1 reduce nitric oxide (NO) bioavailability by scavenging NO and by degrading the NO precursor arginine to ornithine, respectively. In this study we evaluated the relevance of hemolysis to NO-dependent blood flow in patients undergoing cardiovascular surgery. METHODS: Plasma FHb, arginase-1, and amino acid concentrations were measured perioperatively. Forearm blood flow (FBF) responses to the intra-arterial administered NO-donor sodium nitroprusside (SNP) and the endothelium-dependent vasodilator acetylcholine (ACh) were measured by venous occlusion plethysmography. RESULTS: When peak values plasma FHb and arginase-1 were found, vascular dilatation to SNP, but not ACh, was significantly reduced compared with 1 day postoperatively, when FHb had returned to baseline levels (p < 0.05). Interestingly, plasma FHb concentration was inversely correlated to FBF responses to SNP (r -0.93, p < 0.001). In contrast, the increase in arginase-1 was not biologically relevant as the ratio of plasma arginine to ornithine remained constant. CONCLUSION: We conclude that hemolysis with concomitant release of FHb during cardiovascular surgery is associated with impaired NO-dependent forearm blood flow.

Optimising Access Surgery in Senior Haemodialysis Patients (OASIS): study protocol for a multicentre randomised controlled trial.

INTRODUCTION: Current evidence on vascular access strategies for haemodialysis patients is based on observational studies that are at high risk of selection bias. For elderly patients, autologous arteriovenous fistulas that are typically created in usual care may not be the best option because a significant proportion of fistulas either fail to mature or remain unused. In addition, long-term complications associated with arteriovenous grafts and central venous catheters may be less relevant when considering the limited life expectancy of these patients. Therefore, we designed the Optimising Access Surgery in Senior Haemodialysis Patients (OASIS) trial to determine the best strategy for vascular access creation in elderly haemodialysis patients. METHODS AND ANALYSIS: OASIS is a multicentre randomised controlled trial with an equal participant allocation in three treatment arms. Patients aged 70 years or older who are expected to initiate haemodialysis treatment in the next 6 months or who have started haemodialysis urgently with a catheter will be enrolled. To detect and exclude patients with an unusually long life expectancy, we will use a previously published mortality prediction model after external validation. Participants allocated to the usual care arm will be treated according to current guidelines on vascular access creation and will undergo fistula creation. Participants allocated to one of the two intervention arms will undergo graft placement or catheter insertion. The primary outcome is the number of access-related interventions required for each patient-year of haemodialysis treatment. We will enrol 195 patients to have sufficient statistical power to detect an absolute decrease of 0.80 interventions per year. ETHICS AND DISSEMINATION: Because of clinical equipoise, we believe it is justified to randomly allocate elderly patients to the different vascular access strategies. The study was approved by an accredited medical ethics review committee. The results will be disseminated through peer-reviewed publications and will be implemented in clinical practice guidelines. TRIAL REGISTRATION NUMBER: NL7933. PROTOCOL VERSION AND DATE: V.5, 25 February 2021.

Predicting Kidney Failure, Cardiovascular Disease and Death in Advanced CKD Patients.

Introduction: Predicting the timing and occurrence of kidney replacement therapy (KRT), cardiovascular events, and death among patients with advanced chronic kidney disease (CKD) is clinically useful and relevant. We aimed to externally validate a recently developed CKD G4+ risk calculator for these outcomes and to assess its potential clinical impact in guiding vascular access placement. Methods: We included 1517 patients from the European Quality (EQUAL) study, a European multicentre prospective cohort study of nephrology-referred advanced CKD patients aged ≥65 years. Model performance was assessed based on discrimination and calibration. Potential clinical utility for timing of referral for vascular access placement was studied with diagnostic measures and decision curve analysis (DCA). Results: The model showed a good discrimination for KRT and "death after KRT," with 2-year concordance (C) statistics of 0.74 and 0.76, respectively. Discrimination for cardiovascular events (2-year C-statistic: 0.70) and overall death (2-year C-statistic: 0.61) was poorer. Calibration was fairly accurate. Decision curves illustrated that using the model to guide vascular access referral would generally lead to less unused arteriovenous fistulas (AVFs) than following estimated glomerular filtration rate (eGFR) thresholds. Conclusion: This study shows moderate to good predictive performance of the model in an older cohort of nephrology-referred patients with advanced CKD. Using the model to guide referral for vascular access placement has potential in combating unnecessary vascular surgeries.