Those are excellent points Dr. Webb. I have to say that I agree that a stent-graft is a better in the central veins. Unfortunately, in this day of declining reimbursements, bare metal stents have to be used. I have had mixed results and am quick to re-line the BMS with a Viabahn stent when needed. I have had fairly poor results with the Fluency Plus stent in the central veins. I find that they act similarly to bare metal stents. The only place that I think a different approach may be needed is at the costoclavicular junction due to vascular thoracic outlet syndrome. Dr. Illig's outcomes have been quite good. I still find that this region needs to be stented after a thoracic release procedure is performed. As far as access to the central veins, I have placed many 12F sheaths in fistulae without issues. The cannulation site needs to be sutured with adequate compression after removal of the sheath. Placing a 12F is a bit of problem in a graft. If the axillary vein can be accessed, then that can be used. Obviously, a femoral vein approach can be used if needed. I have found no issues with "jailing" off other veins. This has been more theoretic than actual. I have had one patient develop a symptomatic right internal jugular vein DVT after placing a right CCJ stent (after surgical decompression). The clot was removed, and the patient did well. My only concern about "jailing" off the veins is a concern about where a catheter can be placed in the future (if needed for access failure). I am not certain that this concern is more important than maintaining an access. With adequate surveillance and monitoring, most dialysis accesses should be able to be maintained to avoid catheters. I have found that it is almost impossible to "bareback" a Viabahn stent.

For central stent-graft placement, I usually deploy from the arm site. The largest diameter device I regularly use is a 13mm Viabahn. For a very small access, a femoral or IJ approach could be entertained – but the arm approach is usually easiest. Going without a sheath could be considered – but there is significant value to having a sheath and being able to inject contrast while you line-up your stent-graft.

A 13 mm Viabahn will actually go through an 11F Terumo Pinnacle sheath - but this is specific to Terumo. Ironically, not all 11F sheaths have the same inner diameter.

Placing the sheath through a larger caliber fistula is no problem – just do your conventional sheath site suture when you pull the sheath out (I use a z-line stich, and have the suture removed after 3-5 days). If you’ve never used a larger sheath like this, the size can be intimidating at first – but it’s really not much different than what you’re used to.

I usually will approach the lesions from an AV graft in the same manner. The 11F sheath is a bit less than 4mm in diameter, and can fit into conventional 6mm grafts. The key is not to advance the sheath too far into the graft – you just need a few centimeters worth of sheath inside the graft to pass your device. If you pass too much sheath into the graft, you’ll functionally occlude the flow with your sheath.

If stable graft pseudoaneurysm areas are present (which often do develop in the setting of a high grade proximal lesions) – take advantage of greater diameter for your sheath site placement. However; avoid thin-walled pseudoaneurysms for your sheath site to minimize complications. You want good skin thickness for a suture here.

Also, avoid relatively immature grafts or grafts that have not epithelized well – you’ll need this epithelial layer of cells encompassing the graft to seal your entry site.

When you inject from the sheath for your imaging, the relative occlusion of the graft will cause the contrast to flow slowly and reduce the access thrill. While this may induce some initial operator panic, this is not a problem - as long as you work relatively smoothly and recognize that your sheath is causing this. Also, you’ll need a relatively higher volume of contrast for your imaging runs, to pass through the longer, larger sheath.

Again, you can just deploy your device and suture as usual here. You’ll feel a good thrill restored when you take your sheath out.

If you’ve never done this, it’s not a bad idea to have your local rep in on your first case – they can be an invaluable source of information and tips.

Sorry to hog the mic here – just wanted to share some tips from the field.

Those are great pearls. It is amazing to me when I have mentioned placing a 12F in a dialysis access how many surgeons were baffled that the entry site could be closed with compression and suture. I also agree that working with a large sheath in a graft is unnerving. It can be done. Also, if an operator thromboses the graft, a Fogarty catheter can be used to quickly clear the fresh clot. I also agree that having a sheath in place is useful to check placement with contrast prior to deploying the stent-graft.

If by "bareback" you mean inserting the stent over the wire without a sheath (I am not familiar with the term), the Viabahn is rough on the outer surface and not stable. Just putting the stent assembly in and then removing it out the sheath to change your wire or whatnot, will show you that the assembly is unstable. I would not insert a Viabahn without a sheath.

We have been using femoral access with large french sheaths to access and deploy covered stents in the central veins mainly because of the unfortunate high incidence of graft access in the population we care for.

I have talked to different operators who some put large sheaths in the 6 mm AVG if there is no large accessible outflow vein to use or introduced the stent with no sheath. I wanted to see what the operators in our society thought.

The comments made are useful and helpful so thanks to Marc , Wesley and Daniel.

We abandoned stenting centrally, unless needed for real symptoms. Earlier on about 8-10 yrs ago, I was liberal with placing Bard Luminex stents and Cordis Smart stents in varying configurations Bare Metal Stents are not designed to move against shear forces of respiration in the thoracic cage at 17,280 to 23,040 times a day at 12 to 16 breaths per minute. They are designed for either the hepatobilary tract or for tracheobronchial deployment. Tachypnea from exercise will change that number siginificantly. The design of the stents lends to fractures, that we have seen several times in the right brachiocephalic, left brachiocephalic and the subclavian veins (in that order), Endothelialization - sometimes very rapidly. We have seen complete dislodging of fracture stents, resulting in pulmonary embolization a few times and in dramatic case, lodging in the tricuspid valve (was not easy to find) leading to severe right heart failure, eventually treated with a valve replacement. Agree with Marc, Danny and Dany that Viabahn work well centrally when needed.

Placing a large sheath is not an issue in AV Access. The reps will tell you that so and so places them bareback. It is messy. If the French size is of concern, as regards to flow, I heparinzine the circuit. Do all my imaging, mark the monitor and deploy quickly. For larger sheaths, I use a purse string, instead of a Z stitch to get the best hemostasis.....

Lovely discussion --- Sumit Kumar, Dallas

"We abandoned stenting centrally, unless needed for real symptoms" - I agree with Dr. Kumar - why would you place an expensive foreign object in a reactive vascular pathway without real symptoms? I guess what he means is "ugly is as ugly does", and ugly without symptoms should be left alone.
On a related topic, we have had several patients lose their central patency after access placement over the years, and manage to maintain a well functioning fistula via collaterals without significant arm swelling, elevated pressures, etc. I have gone so far as to deliberately place two fistulas this year ipsilateral to known unrecanalizable central occlusions, and again have ended up with functional accesses with minimal to no arm swelling. Tip Jennings will mention flow reduction in these cases, and we have done such to alleviate swelling in central occlusion while preserving the access.

Responding to Dr. Issa, and echoing an earlier contestant, a twelve french sheath can be placed in a 6 or 7mm graft. I give 2-3000 units of heparin and get the 12F sheath out as soon as the stent is deployed - either puncture the access a separate time and use two sheaths temporarily, or pursestring suture around the 12FR sheath, and downsize as soon as the stent is placed, using the purse string to stop bleeding around the smaller sheath. Either way, I have wasted a lot of time trying to figure out why my flow abruptly goes down after stent placement, only to find that the larger sheath has advanced and occluded an intraaccess stenosis that I missed. Sometimes, just pulling the sheath back a bit does the trick.

When I read your note, I was thinking of Dr. Jennings until you mentioned him. He always remarked on the marginal flow fistula with central obstruction. I think your statement about collaterals as well as the first line about stenting centrally and "real symptoms". If an access works using collaterals, why change it. I have actually re-routed fistulae through collateral veins that emptied centrally when the anatomy required it. I have had the cephalic arch become chronically occluded, but a perforator developed to the axillary vein, so I used this for outflow. I think the same issue is true centrally. If collateral veins allow for central drainage, leave the issue alone. I have had SVC occlusions that drain through the azygous vein without any significant problems, so I leave it alone. The same is true for central stenosis. There are many times that I find a central stenosis without symptoms or problems at the dialysis unit with the access. With these patients, I leave the lesion alone. I think the cardiologists are doing this as well. They will study a patient with >50% stenosis but no significant blood flow issues in the coronary artery, so they do not intervene. To me, it makes sense to not break something that is working. We all know that once you angioplasty a central vein, the lesion will likely recur with worsening stenosis.

Agreed. The big sheaths will likely occlude at places that did not look stenosed on the initial angiogram. Sometimes, it is easy enough to just pull the large sheath out of the access and use the dilator to occlude bleeding from the cannulation site while using another access point to perform the angiogram.

I think central venous stenosis in the setting of dialysis access is under studied and over treated.

I commend the conservative interventionists who only intervene when the central venous stenosis has declared itself as a problem: swollen arm, increased post de-cannulation bleeding with no other explanation etc..

Using the angiographic diameter as a criterion to intervene in isolation of the clinical situation could lead to over treatment.

Along the same lines, I think of the dialysis access as a circuit- so if there is a symptomatic central venous stenosis, considering inflow restriction could be a way to deal with the situation if the flow into the access is high enough and could be restricted to see if it improves central venous stenosis symptoms while maintaining enough flow for adequate dialysis.

Re the issue of inflow - the first clue for me is in the ultrasound I do in the OR in every case prior to cannulation - hopefully I can cannulate in a zone without stenoses, and from which I can treat all relevant lesions. I pivot the sheath in the access to address both directions, and only occasionally require a second sheath.
But to the point, I look at the feeding artery - if the artery is over 7mm, I consider the issue of arterial hypertropy. We have seen feeding arteries up to 1.5 cm in size. They can carry a lot of flow.
Second, I look at the size of the anastomosis - we try to keep the greatest dimension to 6mm or less, but we see anastomoses of 1 cm diameter or more. In the situation of a large artery and a large anastomosis, the fistula becomes a fully pressurized part of the arterial tree - a "firehose fistula", "mega fistula" , whatever. Then we measure flows, and very often we find flows of three to four liters a minute. No amount of central stenting will solve this problem, although we have seen a lot of stents placed in this situation. The real solution is inflow reduction.