Native arteriovenous fistulas (AVFs) are considered the vascular access of choice for hemodialysis patients for their superior durability and lower rate of infectious complications than central venous catheters and grafts.1 However, AVF dysfunction remains a significant clinical challenge, with juxta-anastomotic venous stenosis emerging as one of the leading causes of this issue.2

Juxta-anastomotic venous stenosis is typically located within the first 5cm of the arteriovenous anastomosis.3 This lesion accounts for a considerable proportion of early AVF dysfunction,4 affecting both maturation and long-term patency. The etiology of these stenoses is multifactorial, involving hemodynamic factors that induce vascular wall stress and promote intimal hyperplasia.2,5,6 The presence of juxta-anastomotic venous stenosis can lead to reduced blood flow through the AVF, resulting in ineffective hemodialysis sessions and an increased risk of access thrombosis.7 Additionally, these complications increase morbidity and healthcare costs related to repeated interventions and increase of hospitalizations.8,9

The management of juxta-anastomotic venous stenosis has been a subject of debate, with treatment options including surgical interventions and endovascular therapies. Traditionally, surgical correction by creating a new proximal anastomosis has been the preferred approach.10–12 However, in recent decades, endovascular interventions, such as percutaneous transluminal angioplasty (PTA), have gained prominence for their minimally invasive nature and the ability to be repeated as needed.13–15

Despite the multiple therapeutic options available, the optimal treatment for juxta-anastomotic venous stenosis remains controversial. This study aimed to describe short-term assisted primary patency and functional outcomes at six months following endovascular treatment of juxta-anastomotic venous stenosis in a real-world clinical setting. Particular attention was given to the temporal presentation of stenosis and to outcomes in fistulas undergoing intervention for delayed maturation, as these scenarios are highly relevant in advanced chronic kidney disease (ACKD) programs aiming to avoid catheter-dependent dialysis initiation.

Juxta-anastomotic venous stenosis in AVFs is a common cause of vascular access dysfunction in hemodialysis patients.2,16 In this retrospective single-center cohort, endovascular treatment achieved favorable short-term outcomes, with assisted primary patency rate of 90% at 90 days and 84% at 6-months, together with a functional success rate of 90%. These findings are consistent with previous studies supporting PTA as an effective intervention to preserve AVF functionality and maintain access patency.15,17

The median time from AVF creation to stenosis detection was 112.5 days (IQR 60.5–511), underscoring the marked temporal heterogeneity of juxta-anastomotic lesions. As expected, stenosis was identified significantly earlier in AVFs undergoing intervention for delayed maturation compared with established accesses. As shown in Table 2, apart from this marked temporal difference, baseline characteristics and short-term outcomes were largely comparable between groups, although differences were observed in patient status (HD vs. ACKD) and time from AVF creation to stenosis detection. These findings suggest that early flow-limiting lesions during the maturation phase may reflect a different temporal presentation pattern, while short-term patency and functional success after intervention remain similar across maturation status.

The observed 6-month recurrence rate of 8% suggests that endovascular management of juxta-anastomotic lesions provides reasonable durability in this anatomical segment. Although most procedures in our cohort were performed using conventional balloons, adjunctive use of cutting balloons and DCB was reserved for selected cases based on lesion characteristics. Recent meta-analysis has suggested that DCB angioplasty may improve patency compared with conventional balloons in AVF stenosis.18,19 However, their use in our cohort was limited, preventing a direct comparison within our study context.

All observed failures occurred in patients older than 65 years and among those receiving antiplatelet therapy. However, due to the limited number of events (n=8) and the presence of complete separation, conventional Cox proportional hazards models yielded non-estimable hazard ratios. To address this limitation, we performed penalized logistic regression (Firth method) as an exploratory analysis. Although antiplatelet therapy showed a possible association with 6-month failure (OR 18.1; 95% CI 0.97–337.32), the wide confidence intervals indicate substantial statistical uncertainty. Age >65 years was not independently associated with failure in penalized models. Therefore, these findings should be interpreted cautiously and regarded as exploratory.

The clustering of events among patients receiving antiplatelet therapy may reflect confounding by indication. Individuals prescribed antiplatelet agents generally have a higher burden of cardiovascular comorbidity, which itself may predispose to vascular access dysfunction. The literature regarding the role of antiplatelet therapy in AVF patency remains inconclusive.20,21 While some studies suggest a potential benefit in reducing thrombosis, others report no significant improvement in long-term patency and raise concerns regarding bleeding risk. Our study was not designed nor powered to clarify this relationship, and no causal inference can be established. Similarly, although advanced age has been described as a potential risk factor for vascular access dysfunction,17 the limited number of events in our cohort precludes definitive conclusions.

From a clinical perspective, our finding supports endovascular therapy as a widely used first-line approach in current clinical practice. In a population characterized by advanced age and substantial cardiovascular comorbidity, percutaneous transluminal angioplasty provided high short-term patency while preserving native vascular capital and facilitating assisted maturation when required. These results are particularly relevant in advanced CKD programs, where timely access optimization may prevent unplanned dialysis initiation with central venous catheters. The minimally invasive and repeatable nature of endovascular intervention makes it especially suitable in nephrology practice, where the balance between durability, safety, and preservation of future access options is critical.

Several limitations must be acknowledged. First, the retrospective and single-center design limits external validity and may introduce selection bias. Moreover, the absence of a comparator group – either surgical or between different endovascular strategies – precludes any inference regarding comparative effectiveness. Therefore, our findings should be interpreted as descriptive of real-world outcomes rather than as evidence of superiority of endovascular treatment over alternative approaches. Second, the modest sample size and low number of events (n=8) restrict the robustness of multivariable analyses and increase statistical imprecision. Although penalized logistic regression was used to mitigate small-sample bias and separation, the resulting confidence intervals were wide, and any observed associations should be considered exploratory and hypothesis-generating. Third, potential confounding by indication cannot be excluded, particularly in relation to antiplatelet therapy. Patients receiving antiplatelet agents had a higher burden of cardiovascular comorbidity, which may itself influence access outcomes. Finally, follow-up was limited to six months. While short-term patency is clinically relevant – especially in elderly patients with high comorbidity burden – longer-term durability (≥12 months) would be necessary to fully assess sustained access performance and allow comparison with series reporting medium- and long-term outcomes. In the absence of a direct surgical comparator, no conclusions can be drawn regarding the relative effectiveness of endovascular versus surgical revision.18

Despite these limitations, this study provides real-world data in an elderly and highly comorbid population and highlights distinct temporal patterns between maturation-related and established-access stenoses, with high short-term functional salvage in the delayed maturation setting. These aspects may be particularly relevant for advanced CKD programs seeking to reduce catheter dependence.

In conclusion, endovascular treatment of juxta-anastomotic venous stenosis in native AVFs, demonstrated high short-term patency and functional success rates in this cohort. Although exploratory analyses suggested possible association between patient characteristics and failure, these findings require confirmation in larger prospective studies. Within the context of current evidence and routine nephrology practice, PTA remains a valuable and pragmatic therapeutic approach in the management of AVF dysfunction.

I.Z. collected and organized the clinical data. M.P.L., X.F., T.A., N.R.T., C.G.J., J.L.R., and M.G.R., critically reviewed the manuscript and contributed to its refinement. M.A.A., M.J.S., and N.T. conceived the study, supervised the project, and wrote the manuscript. All authors read and approved the final version of the manuscript.

This research was funded by ISCIIII-FEDER and ISCIII RETICS REDinREN, grant number PI21/01292, PI24/01510, ERA PerMed JTC2022 grant number AC22/00029, Río HortegaCM23/00213, Marató TV3421/C/2020, Marató TV3215/C/2021, RICORSRD21/0005/0016, and RD24/0004/0031. Enfermedad Glomerular Compleja del Sistema Nacional de Salud (CSUR), enfermedades glomerulares complejas, PI23/01209 Proyectos de I+D+I en Salud AES2023.

M.J.S. reports personal fees from Novo Nordisk, Jansen, Mundipharma, AstraZeneca, Esteve, Fresenius, Ingelheim Lilly, Vifor, ICU, Pfizer, Bayer, Travere Therapeutics, GE Healthcare, MEDICINE and grants and personal fees from Boehringer Ingelheim, outside the current study. I.Z., M.A.A., M.P.L., X.F., N.R.T., C.G.J., J.L.R., M.G.R., N.T., declare no conflicts of interest.

The datasets generated and analyzed during the current study are not publicly available due to institutional privacy policies but are available from the corresponding author upon reasonable request.