Endovascular Intervention for Central Venous Stenosis in Hemodialysis Patients: A Single-center Experience

Introduction

The term “central venous stenosis” (CVS) usually refers to a significant stenosis of a big intrathoracic vein such as the subclavian, brachiocephalic, or the superior vena cava. Stenosis is considered significant if it causes disturbance or obstruction to forward flow and usually occurs if the diameter of the stenotic portion is <50% of the unaffected portion. Central veins are commonly injured as a result of placement of intravascular devices or avascular access in hemodialysis (HD). Nearly, 80% of patients with end-stage renal disease (ESRD) in the United States initiate dialysis using a catheter and consequently central vein injury and subsequent restorative response leading to CVS are common. The effect of the stenosis depends on the flow rate and CVS may be clinically silent until an ipsilateral arteriovenous fistula (AVF) or graft in created for HD purpose. Subsequently, it often becomes a major issue because these patient present with poor fistula flow rates, sometimes making HD difficult, increased venous pressure during HD, and bleeding episodes and in more severe cases swelling of the arm bearing the AVF and AVF/graft thrombosis. The incidence of unsuspected CVS in patients with functioning grafts was reported to be 29%.[1]

Percutaneous transluminal angioplasty (PTA) with or without stent placement has been the recommended preferred approach to CVS. Guideline of the Kidney Disease Outcomes Quality Initiative (K/DOQI) suggests that the percutaneous intervention with transluminal angioplasty is the preferred treatment for CVS.[2] We aimed to look at the outcomes of endovascular intervention for HD patients with diagnosed CVS and poor AV fistula flow rates in our institute over a two year period from April 2017-April 2019.

Materials and Methods

A total of 10 consecutive HD patients with documented CVS with AV fistula malfunction and poor blood flow on HD over the 2-year time period from April 2017-April 2019 underwent venous angiography and percutaneous angioplasty and self-expanding metallic stent insertions under sedation. The procedure was performed by the interventional cardiologist in the institute. All patients received antiplatelet therapy with aspirin 150 mg and clopidogrel 75 mg for thrombosis prophylaxis and this was continued post procedure for at least 4 weeks.

Results

Out of the 10 patients with CVS who underwent PTA or PTA with stenting, 7 were males with 3 females with mean age of 58.4 years ± 10.35 yrs. A total of 50% of patients had diabetes and the average time on dialysis was 15.87 ± 7.1 months [Table 1]. All patients had previous HD catheter insertions, while two (20%) of patients had documented history of catheter related infection. One patient underwent only PTA, whereas nine (90%) had OTA with primary stent insertion. Primary patency was 90% at 3 months, 80% at 6 months, whereas at 12 months, it was 70% and remained at 70% at 24 months. Figure 1 depicts the results in a patient who was intervened successfully. In one patient, though there was good postprocedural flow, the ipsilateral AVF could not be used due to distal vessel narrowing. We did not find any association between age, gender, diabetic status, dialysis vintage, or previous catheter infection with procedural patency. Two patients had a restenosis, one after 15 months and underwent repeat dilatation with adequate flow rates on dialysis, whereas the other could not be salvaged and a tunneled catheter had to be inserted on the opposite side for continuation of dialysis.

Close

Figure 1

Pre and post veno-dilatation central vein angiography of a patient

Export to PPT

Central venous stenosis leading to ipsilateral AV fistula dysfunction and HD inefficiency can be treated successfully with percutaneous angioplasty and stenting. It is, however, important to consider the limitations of our understanding of endovascular intervention and the modulation of the endovascular response. Despite advances, prevention of CVS should be the primary approach.

Discussion

Central venous stenosis is a major and not so infrequently encountered problem in the dialysis population. It was more frequent when subclavian vein catheter were inserted for HD procedures in the 1980s.[3] In a study, out of 69 consecutive patients subjected to percutaneous placement of a tunneled right internal jugular vein (IJV) catheter who underwent prior venography, 29 (42%) were found to have unexpected stenosis or angulation.[4] In a more recent study[5] comprising 2811 HD patients in a single center from January 2000 to December 2013, CVS was diagnosed in 120 (4.3%) at a median dialysis vintage of 2.9 years (interquartile range1.8–4.6 years).

Early attention focused on HD catheters as etiologic agents responsible for this problem. However, insertion of a variety of indwelling devices including central venous lines, pacemaker wires also leads to the development of central venous stenosis.[67] Central venous stenosis has also been reported in patients, however, without any previous central vein catheter placement or procedures causing thrombosis.[8] It was recognized early that dialysis catheters are associated with late vascular complications that may adversely affect the outcome of permanent vascular accesses. In a prospective study of 50 patients with subclavian catheters and 50 patients with IJV catheters, venography revealed CVS in 42% of patients in the subclavian group compared to 10% in the internal jugular group.[9] Other studies have also shown that rates of CVS at around 40%.[10] Longer time on HD and a history of previous HD catheter insertion were factors associated with stenosis. A tendency toward an increased incidence of stenosis was also observed following cannulation of the left compared to the right IJV.[11] This is thought that angulations between the left IJV, brachiocephalic vein, and superior vena cava as well as the greater length that the catheter must traverse are responsible for this increased occurrence.[12]

A majority of the affected patients remain asymptomatic. In a study of 202 patients who underwent permanent pacemaker implantation with indwelling cardiac wires inserted transvenously, 129 (64%) developed varying degrees of central venous stenosis. However, only 12 (9.3%) patients were symptomatic.[6] CVS, however, becomes clinically significant in the presence of an ipsilateral arteriovenous dialysis access that drains into the affected central veins. High venous pressure and blood flow due to the fistula overwhelms the collateral venous and lymphatic drainage, resulting in the development of tortuous collateral veins over the ipsilateral arm, neck, and upper chest. Venous hypertension may lead to severe and disabling arm edema and discomfort. Reduced access blood flow and compromised dialysis delivery occur due to access recirculation.

Conclusion

Central venous stenosis leading to ipsilateral AV fistula dysfunction and HD inefficiency can be treated successfully with percutaneous angioplasty and stenting. It is, however, important to consider the limitations of our understanding of endovascular intervention and the modulation of the endovascular response. Despite advances, prevention of CVS should be the primary approach.