Cephalic arch stenosis is a well-known location for stenosis among hemodialysis patients with AVAs, especially in patients with brachiocephalic fistulas3,4. The pathophysiology of CAS is still being investigated and various contributing factors have been suggested. Physical factors such as venous valves in the cephalic arch and the traversal of the cephalic arch through the deltopectoral groove may constrict the blood vessel and restrict venous return into the cephalic vein. Fluid dynamic changes as a result of the curvature as the cephalic arch enters the axillary vein has corresponds to low wall sheer stress within the cephalic arch, promoting intimal hyperplasia and hypertrophic remodeling13,14. Since CAS lesions are generally short and focal in nature, they are associated with increased morbidity and may lead to high venous pressures, prolonged bleeding after dialysis, dysfunctional dialysis, increased rate of fistula thrombosis, and fistula failure.
Despite being the most common stenosis in dysfunctional AVAs, accounting for 30% to 55% of all upper arm access stenosis3, CAS is still known to be notoriously difficult to treat and responds poorly treatment by venoplasty alone, only providing primary patency rate of 42% at 6 months15. Treatment of CAS is further complicated by higher rupture rate due to increased pressure required for anatomical success. As venoplasty alone has not demonstrated significant lasting benefits for CAS, recent comparison has demonstrated superiority of SG over other modalities of endovascular treatment, specifically venoplasty, BMS and drug-eluting stent in the treatment of CAS, at maintaining CAS patency7.
Although SGs has demonstrated significant benefits over other modalities of treatment of CAS, common occurrence of edge stenosis has prevented SG from becoming the standard treatment for CAS. One previous study has reported the primary stenosis site was observed mostly at the lateral edge of the SG, accounting for 52% of stenosis episodes. Other stenosis site included bilateral edge stenosis, medial edge stenosis and finally in-stent stenosis only occurred in 29%, 17%, and 2% respectively of all SG related stenosis10.
We also documented rapid onset of lateral edge stenosis in our patients after SG placement, especially in patients with SG diameters similar or greater than the adjacent cephalic vein. We noticed among patients with SG smaller than the adjacent cephalic vein, reintervention for stent failure and edge stenosis was less frequent. As a result, our study retrospectively recruited all patients with SG placement at the cephalic arch to compare the patency rate of SG of the patients whose SG are apposed to the cephalic vein with those with SG undersized relative to the cephalic vein.
The two groups were comparable in clinical characteristics (Table 1). Furthermore, the prevalence of cephalic arch stenosis among those with brachiocephalic AVF and radiocephliac AVF was similar to a previously reported study5. Among the patients in the undersized SG group, smaller stents were used resulting in significantly smaller S/V ratios. Outcomes showed a significant advantage for using undersized SG over apposed SG: patients with undersized SG required a decreased amount of post-SG interventions needed per access-year for AVA dysfunction, decreased percentage of lateral edge stenosis, and a higher primary stent patency rate and primary access patency rate.
While SGs has been used to great effect in the treatment of arterial related diseases such as aortic aneurysms, peripheral arterial diseases, and pseudoaneurysms, research trials and guidelines suggest oversized SG to reduce SG migration. However, the nature and structure of veins are known to be vastly different from arteries, and SGs specifically designed for use in arteries may not be ideal for treatment of venous lesions such as refractory stenosis or venous vessel rupture16. To the best of our knowledge, there have been no previous studies discussing proper SG sizing for use in AVAs, whereas undersized SG has been previously shown in Gore REVISE Clinical Study to improve primary stent patency rate17. Without adequate research, the guidelines of oversizing SG may be suboptimal when treating cephalic arch stenosis or other venous stenosis16.
While the majority of SG failure in the cephalic arch is currently attributed to neointimal growth causing edge stenosis at the junction of SG and the apposed vessel wall8,18,19, the exact cause of neointimal growth has not been adequately investigated. Previous fluid dynamic studies have shown that apposition of SG to adjacent vessel wall may lead to higher shear stresses at the outflow vein, and can lead to development of neointimal hyperplasia, resulting in edge stenosis20,21. Undersized SGs minimize the contact between the medial and lateral edges of the SG to adjacent venous wall and may help reduce neointimal hyperplasia at the edge of the SG and possibly prevent future edge stenosis.
Stent migration is a common complication associated with undersized stent placement, and is typically remedied by selection of oversized stents. Despite placing smaller SG in our undersized patient group, there was no evidence of stent migration at the last angiographic follow up for each of the 22 patients, nor were there any symptoms or treatments related to stent migration. We believe that due to the tortuous anatomy of the cephalic arch, SG are able to maintain their position despite being relatively undersized compared to both the cephalic and subclavian vein. Furthermore, venous stenosis at the cephalic arch allows the SG to be fixed in place via apposition of the central portion of the stent to the stricture location. Among the 12 patients with ruptured vessels due to balloon dilation, adequate extravasation control was noted in all patients, despite undersizing SG in 7 of the 12 patients. Control of extravasation may be maintained with undersized SG due to two reasons. Firstly, SG were chosen such that the diameter is equal or greater than diameter of the balloon that caused vessel rupture during prior balloon assisted angioplasty. Secondly, although the SG is undersized relative to the medial and distal vessel, it maintains apposition to the stenosis site, and allows for adequate extravasation control.
Another concern of undersized stent placement is the increased flow resistance and wall stress at the inflow area caused by tiny gaps between the stent and the wall in arteries22. However, we detected no evidence that under sizing SG caused significant flow disturbances in our cohort resulting in venous hypertension, poor hemodialysis function, or other clinical symptoms. It is possible that the relatively slow flow rate in the vein decreases the impact of the flow turbulence on stent stenosis. Among the 22 patients, no events of elevated venous pressure were directly caused by the undersized stent. Rather, edge or in-stent stenosis were the main culprit in patients presenting with high venous pressure after SG placement.
Viabahn Endoprosthesis was used in all cases in this study and was selected for its flexibility and ability to conform to the arch of the cephalic vein. When placed in the tortuous vessels such as the cephalic arch, the natural anatomy is maintained and avoids “tenting” of the cephalic arch23. In addition, full covering of the expanded polytetrafluoroethylene liner over the external metal nitinol structure reduces rate of in-stent and edge stenosis24. The unique mechanical properties of the Viabahn Endoprosthesis appear to prolong the patency time in the treatment of CAS23.
Optimal deployment of SG is another important factor in maximizing the patency rate after treatment of CAS. At our hospital, SGs are deployed across the entire length of any diseased cephalic vein and the cephalic arch with the medial end of the SG protruding into the subclavian vein. Covering this length reduces the likelihood of stenosis in segments of the cephalic arch known to have increased stenosis risk25. However, accurate SG placement at the cephalic arch is especially difficult due to the angle of entry and the anteroposterior orientation of the confluence into the axillary vein. By protruding the SG into the subclavian, we not only ensure that the SG covers the entire cephalic arch, but also lowers the difficulty of positioning of the SG when using angiography alone. Another possible advantage of extending the stent into the subclavian vein is the reduction of turbulent flow at the junction of the cephalic arch and the subclavian vein through alignment of the returning flow vectors23. Operators should remain aware that SG protrusion into the subclavian vein may increase risk of central vein occlusion10, even though, we observed no symptoms of central vein occlusion, including high venous pressure or arm swelling. A secondary benefit of selecting relatively smaller SGs may be reduced obstruction of the returning axillary venous flow, thus mitigating risk of stenosis and occlusion.
To the best of our knowledge, there has been no previous comparative study on the optimal SG diameter for treatment of CAS. Although oversized SG is recommended for treatment of arterial disease to prevent SG migration, oversizing SG may be suboptimal for treatment of CAS or other venous diseases. There should be more considerations when using arterial devices for treatment of venous pathologies.
A number of limitations should be noted in this study, including the reliance of previously collected data for this retrospective study. Second, there is still a relatively low number of patients and may reduce the power and increases the margin of error of this study. Further research with increased patient size and randomized control groups may be necessary to strengthen our study findings. Lastly, symptomatic follow-up through venoplasty and venography potentially underestimates patency rates in subclinical stenosis.
Undersized SG in patients with CAS showed significant higher primary stent and access patency rates, lower number of post-SG interventions per access-year, and lower number of lateral edge SG stenosis compared to patients with apposed SG. The present study emphasizes the need for more deliberations when sizing SG for treatment within AVAs, and future prospective trials comparing the placement of undersized SG verses apposed SG should be done to determine which improves patency.