Discussion

This case illustrates the multifactorial nature of AVF maturation failure in a patient whose serial fistulograms consistently demonstrated patent central veins without discrete stenosis. Multiple factors likely contributed to impaired maturation, including reduced cardiac output from ischemic cardiomyopathy (LVEF 40-45%), accessory venous branches diverting flow from the main outflow tract, vascular geometry at the anastomotic site, and the uremic milieu of ESRD, all of which are well-established contributors to AVF dysfunction [1,2]. However, the clinical course raises the possibility that the ipsilateral transvenous ICD lead may have served as a potentially important and modifiable contributor, not through overt mechanical obstruction or angiographic stenosis, but through subclinical hemodynamic perturbation that impaired the flow dynamics necessary for fistula maturation. The improvement in upper extremity swelling and subsequent clearance of the fistula for hemodialysis use following lead extraction is consistent with, but does not establish, this interpretation.

Proposed role of the transvenous lead in functional hemodynamic disturbance

The central paradox of this case--a non-maturing AVF with patent central veins--may be explained by the concept of functional obstruction. Transvenous leads do not need to produce angiographic stenosis to impair venous hemodynamics. Computational analyses have demonstrated that leads create focal pockets of low velocity and shear stress between the vein wall and the lead, upregulating inflammatory markers and promoting a prothrombotic milieu even in the absence of visible narrowing [4]. Early after lead insertion, endothelial damage and flow perturbation trigger an inflammatory cascade mediated by neutrophils, macrophages, and foreign body giant cells, with circulating procoagulant factors increasing soon after placement [4]. Over time, fibrin deposition on the lead surface incorporates into the intima, with smooth muscle cell infiltration and eventual collagen encapsulation that further reduces the functional venous diameter without necessarily producing a discrete stenotic lesion [4].

This distinction between functional and mechanical obstruction is clinically important. Standard two-dimensional (2D) fistulography evaluates luminal patency but cannot detect the hemodynamic disturbances created by a lead occupying the venous lumen. A transvenous lead may reduce the effective cross-sectional area available for flow and contribute to disturbed flow at the lead-wall interface, creating conditions analogous to the disturbed flow patterns that drive neointimal hyperplasia at swing points in AVFs [3]. Although the traditional rule of sixes uses a flow of at least 600 mL/min and a vein diameter of at least 6 mm as reference thresholds, the 2019 KDOQI guidance emphasizes individualized clinical usability rather than fixed ultrasound criteria alone. Even modest hemodynamic disturbance may interfere with the outward remodeling required for maturation [1].

The epidemiologic data are consistent with this proposed mechanism. Symptomatic central vein stenosis was reported in up to 71% in selected cohorts of hemodialysis patients when the AVF and transvenous lead were positioned ipsilaterally, at a mean of 12.6 months after AVF creation [3]. In a large venographic study of over 3,000 CIED carriers, proper AVF function was uncertain in nearly 50% of patients ipsilaterally versus only 2% contralaterally [5]. The time to first intervention is notably shorter with ipsilateral access compared to contralateral placement [3,6]. These data suggest that the lead's hemodynamic footprint--its cumulative effect on flow dynamics, endothelial activation, and venous remodeling--may be substantial even when conventional imaging appears reassuring.

Compounding factors

While the transvenous lead may have represented a potentially important contributing factor, several non-modifiable factors likely compounded the hemodynamic insult. Pre-existing systolic dysfunction is among the most powerful predictors of AVF maturation failure; patients with systolic dysfunction are more than five times less likely to achieve maturation by one year [7]. LVEF is independently associated with unassisted maturation, and this patient's LVEF of 40-45%, combined with extensive multivessel coronary artery disease, placed him in a high-risk category based on cardiac output alone [7,8]. Reduced cardiac output limits the hemodynamic drive--the sustained high-flow, high-shear environment--necessary for the compensatory vascular remodeling that underlies successful maturation [1,8].

Accessory venous branches, identified and embolized with coils during the clinical course, diverted flow from the main outflow tract. While these branches do not cause stenosis of the main channel, they reduce the effective flow volume through the maturing vein, further diminishing the shear-stress stimulus for outward remodeling [1]. The vascular geometry at the anastomotic site--appreciated clinically as kinking--is consistent with a swing-point mechanism, where sharp angulation creates nonlaminar flow and decreased wall shear stress that can trigger neointimal hyperplasia [3]. Computational fluid dynamics studies have consistently linked disturbed flow patterns, including low wall shear stress and high oscillatory shear index, to regions prone to neointimal hyperplasia and stenosis [3,9]. Importantly, these hemodynamic disturbances are invisible on standard 2D fistulography.

The convergence of these factors--a proposed lead-related hemodynamic disturbance superimposed on reduced cardiac output, accessory branch steal, unfavorable vascular geometry, and the uremic milieu--may explain why this fistula failed to mature despite apparently patent central veins. Each factor alone may have been insufficient to prevent maturation, but their combination created a hemodynamic environment inhospitable to the outward remodeling process.

Clinical implications: the case for s-ICD in dialysis patients

The clinical improvement and subsequent clearance of the AVF for hemodialysis use after transvenous lead extraction and S-ICD placement may have important implications for providers managing the intersection of CIED therapy and hemodialysis access. When a transvenous ICD is identified as a potential contributor to impaired AVF maturation, the clinical decision is not simply whether to relocate the AVF to a contralateral site, which may not be feasible given limited vascular options, but whether to transition the ICD platform itself to one that leaves the venous system entirely untouched.

The S-ICD and transvenous ICD differ fundamentally in their relationship to the vasculature and their therapeutic capabilities. The transvenous ICD places leads through the subclavian vein into the cardiac chambers, enabling bradycardia pacing, antitachycardia pacing (ATP), and cardiac resynchronization therapy (CRT) in addition to defibrillation. The S-ICD, in contrast, is implanted entirely in the extrathoracic subcutaneous space with a lead positioned superficial to the sternum, providing defibrillation without any intravascular components. This avoids risks directly attributable to intravascular leads, including lead-related venous obstruction, endovascular infection, central venous stenosis, and endovascular infection, but at the cost of being unable to provide bradycardia pacing (beyond brief post-shock pacing), ATP, or CRT.

The inability to deliver ATP is the most clinically significant limitation. In the PRAETORIAN trial, the first ATP attempt successfully terminated 46% of monomorphic ventricular tachycardias, but accelerated the arrhythmia in 9.4% of cases [10]. Patients in the S-ICD group were more likely to receive an ICD shock (19.2% vs. 11.5% at 48 months; P=0.02), though the total number of appropriate shocks was not significantly different between groups, and first shock efficacy was comparable (93.8% vs. 91.6%) [10]. The PRAETORIAN-XL extension at 87.5 months of follow-up demonstrated that transvenous ICD patients had significantly more major complications and lead-related complications than S-ICD patients, and the as-treated analysis showed significantly more overall complications with transvenous ICD (hazard ratio (HR) 0.64; P=0.047) [11]. These data led the investigators to conclude that the S-ICD should be considered for all patients without a pacing indication who are evaluated for ICD therapy [11].

For dialysis patients specifically, the rationale for S-ICD is particularly compelling. The 2017 AHA/ACC/HRS guidelines provide a Class I recommendation that patients who meet ICD criteria and have inadequate vascular access or are at high risk for infection, including those with ESRD, should receive an S-ICD when pacing is neither needed nor anticipated [12]. The 2024 AHA Scientific Statement on CIED infections similarly recommends consideration of S-ICD in patients at high risk of infection, explicitly including hemodialysis-dependent kidney disease [13]. The ACC/AHA/HRS 2025 Appropriate Use Criteria note that due to the increased risk for endovascular infection in patients on chronic dialysis, the S-ICD is now often considered as an alternative to transvenous ICD systems [14]. Nationally, the proportion of S-ICDs among dialysis patients increased from 10% in 2012 to 69% in 2018, reflecting rapid adoption of this strategy [15].

Should S-ICD be used empirically in dialysis patients?

Current guidelines do not mandate S-ICD over transvenous ICD in all dialysis patients, but the trajectory of evidence and practice patterns suggests this may be an area of evolving consensus. The existing Class I recommendation is framed around patients with "inadequate vascular access or high risk for infection," and ESRD patients on hemodialysis meet both criteria. The JACC State-of-the-Art Review on CIEDs in chronic kidney disease (CKD) patients recommends that AVF and transvenous leads be positioned contralaterally when possible but acknowledges that limited vascular sites and frequent AVF revisions make this challenging [3]. When contralateral placement is not feasible, as in this patient, whose right-sided venous options were inadequate, the choice becomes between accepting the potential vascular-access consequences of ipsilateral transvenous lead placement or transitioning to a platform that avoids the venous system entirely.

This case raises the question of whether empiric S-ICD placement should be considered at the time of initial ICD implantation in dialysis patients, rather than reserving it for situations where transvenous leads have been implicated in complications. The rationale would be proactive preservation of vascular access--a finite and life-sustaining resource in dialysis-dependent patients--rather than reactive management after maturation failure has occurred. The primary barrier to this approach is the S-ICD's inability to provide ATP or bradycardia pacing, which limits its use in patients with concomitant bradycardia or frequently recurring monomorphic ventricular tachycardia (VT) amenable to ATP. Notably, renal dysfunction is associated with conversion of S-ICD to transvenous ICD (odds ratio (OR) 2.67; P=0.008) due to subsequent need for pacing or CRT [3]. The emerging extravascular ICD (EV-ICD), which places a lead in the substernal space and can provide both ATP and pause-prevention pacing without intravascular components, may eventually bridge this gap, though further data in dialysis populations are needed [16].

For the broader multidisciplinary team--vascular surgeons, nephrologists, interventional radiologists, and electrophysiologists--this case underscores the importance of early, coordinated planning. When a dialysis patient requires ICD therapy, the decision about device platform should incorporate not only the arrhythmia substrate and pacing needs but also the current and anticipated vascular access requirements. If the patient does not require bradycardia pacing, ATP, or CRT, the S-ICD offers a strategy that preserves the venous system for hemodialysis access without compromising defibrillation efficacy.

Limitations

This report has several limitations. As a single observational case, it cannot establish causality between the ipsilateral transvenous ICD lead and delayed AVF maturation, and its findings may not be generalizable to other patients with cardiac implantable electronic devices and hemodialysis access. Although improvement followed lead extraction, delayed natural maturation, the preceding accessory-vein embolization, and other patient-specific factors may also have contributed to the observed course. Serial duplex-derived access-flow measurements, outflow-vein diameter measurements, venous pressure measurements, quantitative assessment of upper-extremity swelling, and direct computational or physiologic evaluation of flow disturbance were unavailable. In addition, although the AVF was cleared for hemodialysis use, subsequent successful cannulation, sustained functional use, dialysis adequacy, and tunneled catheter removal were not confirmed. Accordingly, the proposed mechanism of lead-related functional hemodynamic disturbance should be considered hypothesis-generating and warrants prospective investigation.