Rick De Graaf, MD, PhD is an interventional radiologist in the Department of Radiology at Maastricht University Medical Centre, Maastricht, The Netherlands. At the 2015 VEITH Symposium, he presented the results of the first comparative study of 4 dedicated stents for the deep venous system. Dr. De Graaf spoke with Vascular Disease Management about the state of venous therapy and results of his study.
VDM: Please tell us about your work.
De Graaf: I am an interventional radiologist and I work in two large referral centers in The Netherlands. I treat mostly venous obstructions that are located in the pelvic region: iliac vein occlusions, in many cases with contribution of the common femoral vein or the inferior vena cava. In these cases we usually perform endophlebectomy with arteriovenous (AV) fistula to enhance inflow and increase short-term stent patency. In the last 4 years the number of deep venous recanalizations increased more than 5-fold in our center. We are very much engaged with both clinical and fundamental research at our center. In addition to the growing attention on stent design we should definitely not forget to put more emphasis on studying the fundamentals of deep venous disease.
VDM: For interventional radiologists or interventional cardiologists looking to start treating venous disease, what are some tips you would give them to get started?
De Graaf: First, get yourself educated in the treatment of deep venous disease. Visit dedicated venous conferences like the American Venous Forum. Get familiar with the aspects of venous disease, not only deep venous obstructions, but the whole patient. You’re treating the venous patient, not just a deep venous obstruction. It’s very important to recognize that venous disease is completely different from arterial disease. Differences in anatomy and physiology play a significant role.
Once this concept has been accepted, I would suggest starting getting experience treating May-Thurner syndrome, i.e. compression of the right common iliac artery on the left common iliac vein. The vein in itself is usually normal without significant scarring, so difficulties recanalizing an obstructed vein do not exist. The procedure can be done with just one stent, a relatively short stent of about 6 cm and usually 16 mm in diameter, sometimes 18 mm in diameter. It’s not a difficult or complicated procedure to do, and the long-term patency rates are close to 100%. When the patency report is lower than 100%, it usually is because the indication was not accurate. Some post-thrombotic scarring of the iliac or femoral veins might have been overlooked because of inferior imaging. When you start a venous practice there are some essential requirements one of which is imaging. Cross-sectional imaging, for instance magnetic resonance (MR) venography and computed tomography (CT) venography can also be used. My personal opinion is that MR venography is superior, however I do realize that this modality is not always readily available and the image quality is not necessarily optimal in all centers. Without a doubt, intravascular ultrasound has been most used in studies and practice worldwide. It is essential in the treatment of deep venous obstructions and should be incorporated in your venous practice.
However, an up-to-date angiographic lab or hybrid operating room (OR) with excellent radiographic imaging should not be underestimated. Still, a number of physicians work with an out-of date C-arm in a general OR. Aside from the poor image quality, high radiation dose might unnecessarily endanger the interventionalist.
Interventional material is quite personal, however one should at least have the availability of a wide range of guide wires, catheters, and balloons, with diameters from 12 mm to 24 mm. Most stents I use to treat deep venous obstructions go through a 10 Fr sheath. Regular noncompliant balloons with a rated burst pressure of 12 atm to 14 atm suffice. In special cases with high resistant fibrosis preventing the balloon to inflate completely, high-pressure balloon are needed and should therefore be part of your interventional arsenal.
When you feel comfortable treating May-Thurner syndrome, you might begin to treat other obstructions, at which point recanalization becomes more difficult. It’s completely different than arterial disease and you need practice and experience. It’s not uncommon to work for up to 6 hours to recanalize a venous lesion. The lesions are usually long, and it takes time to wiggle your way through them. You cannot use force, either manual or with traumatic material because you risk going retroperitoneal, after which re-entry is quite difficult. It’s very important in treating post-thrombotic obstructions to be careful, have patience, and try different approaches, such as ipsilateral, contralateral, or through the jugular vein on the right side. But you need to take it slow and try different approaches without going too fast or too forcefully. Then you’ll have very high and consistent success rates.
A very thorough predilatation is important, even when the vein is chronically occluded. You always need to predilate with at least 16 mm balloons in the common iliac vein. I now use a 16 mm balloon all the way to the inguinal ligament. You should always overdilate compared to the stent that you’re going to use. After breaking up the tight fibrosis, which can be seen by the severe waist in the balloon, to 8 atm to 10 atm, the vein diameter will not stay at the diameter of the balloon. Elasticity of the obstructed vein and perivascular tissue make the lumen collapse significantly. The modern stents usually open up to about 80% of the predilated diameter. The stents are definitely not strong enough to open a vein that is not predilated. After thorough predilation you need to stent the entire predilated vein. A common mistake is to leave a decent appearing vein on venography untreated after predilation. Intravascular ultrasound (IVUS) has been shown to identify trabeculations left obscured on venography. If untreated, these can obstruct flow and hamper patency.
My opinion differs from some in that I believe you do not have to stent with large stents of 16 mm to 18 mm in the common iliac vein, 14 mm in the external iliac vein, and 12 mm in the common femoral vein (CFV). In an occluded iliac vein, no additional flow is entering the stents between the CFV and the inferior vena cava. The inflow from the leg is maximized to a 12 mm lumen, therefore placing 16 mm to 18 mm stents downstream would slow flow down, which increases thrombogenicity and the risk of blockage. In formerly chronically occluded iliac veins, I start with a 14 mm stent at the iliac confluence, usually with a length up to 150 mm. I use the 14 mm stent because of the unpreventable residual compression at the level of the overriding common iliac artery. The surface area will be similar to the fully expanded 12 mm stent placed distally. This 12 mm stent is placed into the CFV where the vein is healthy again or into the surgically desobstructed vein.
Here we reach another very important point: you need to be sure that at the location where you’re going to place the stent, there is a clean segment of vein. There cannot be large trabeculations and synechiae at that location because the inflow, which is most important in that moment, will be obstructed too much. Without good inflow, the stent is going to block sooner or later. The femoral vein and the profunda vein are providing inflow for the recanalized iliac veins. If those two veins are good and the confluence of the femoral and profunda veins at the level of the groin or a little bit more distally in the groin is free of disease, you have a very good chance that the stents will stay open.
If the common femoral vein and both inflow veins are severely scarred, I would suggest not doing a purely endovascular procedure. Our solution is to do a surgical desobstruction of the CFV and include as many profunda branches as possible into the inflow of the stent. That’s why I repeatedly say that treating deep venous disease should be done within a dedicated vein clinic and with a team of venous experts, amongst these with endovascular and surgical skills.
The most extensive disease is where the vena cava is involved. This is usually due to congenital problems or, for example, in very premature infants who had a central line after birth. As with any other occluded vein, the vena cava can be recanalized and predilated with large balloons (up to 25 mm). Then I place a high radial force large stent up to 24 mm with still some flexibility. In bilateral cases where the iliac veins are also involved, I usually use 2 stents to come down to the confluence and then extend it bilaterally. I do not place 2 self-expandable stents inside the vena cava stent, because I have the experience that one stent is always compressing the other. I use 2 specially designed balloon-expandable Andrastents (Andramed; not available in the United States) and then extend with self-expandable stents towards the iliac veins. These are very complex procedures that also take a very long time to recanalize. We encounter them in about 8% of all our cases.
VDM: What are the tools available for venous stenting?
De Graaf: A topic that is very popular today is dedicated venous stents, which is one of the reasons I’m here to talk at this meeting. One of the major concerns at the moment in the United States is that for a long time only the Wallstent (Boston Scientific) was FDA approved for deep venous treatment. The Wallstent was reported in the literature to have done well, with about 75% to 85% primary patency, but there is room for improvement. In Europe we’ve had the availability of multiple dedicated venous stents over the last several years, and this definitely made a difference in deep venous treatment. With this experience we further boosted the knowledge and motivated industry to come up with innovative designs. In the near future several companies will come out with dedicated venous stents. Many innovations and improvements will be there soon to help the physician treating the patient better.
VDM: How do you choose the best stent?
De Graaf: Right now, we have three dedicated venous stents. We have Zilver Vena (Cook Medical), Vici (Veniti), and the sinus-Venous stent (Optimed), which at the moment is only available in Europe. New stents are coming. Bard is soon entering the market with the Venovo stent. Medtronic is going to introduce venous stents for sure and there will probably be numerous smaller and larger companies introducing innovative designs. It is very likely that not just one stent is best for all indications. Generally speaking, the best stent, in my opinion, does two things: it’s optimal at the high pressure points and it’s optimal at the high flex points, and both challenges can be found throughout the deep venous system. With May-Thurner syndrome, a strong stent is required, a stent with high crush resistance and radial force. A stent that is too weak will not be able to push the overlying artery away, which results in residual compression. On the other hand, a strong nitinol stent tends to be more rigid. The more material you need to create a strong stent, the more rigid it becomes. For sure, I do not like a stent to manipulate the anatomical configuration of the iliofemoral veins. What I suggest to simulate this is that you try to form a supposedly dedicated venous stent into an S shape with your hands. If you look at the venous anatomy from multiple projections, you see some sort of S shape in most occasions. The stent should be able to follow that shape effortlessly. If you feel some resistance creating that S shape with the stent, in my opinion, it’s not the optimal stent. These longitudinal forces might result in kinking or fracture of the stent at flex point or cause straightening of the vein with all kinds of negative consequences like perforation or the curtain phenomenon, in which the vein is pulled up by the stent and collapses over the stent, causing flow obstruction.
One of the most important issues in deep venous stenting is the fairly young patient group we treat compared to the patients suffering from peripheral arterial disease. The stent has to perform much longer – for up to maybe 60 years. We do not have experience with that. We are used to stents being in the body for 10 years to 20 years because that’s the remaining life span of the patient. With venous disease, the patients need stents that keep their integrity, strength, and flexibility over a much longer period of time. This aspect needs to be taken very seriously when thinking about dedicated venous stent design.
VDM: What is coming in the future for venous therapy?
De Graaf: This is a time of innovation that will bring something new to deep venous therapy, but we may have to wait 5 or 10 years before we get the perfect venous stent. However, as said, probably one size will not fit all, there might be a specific May-Thurner stent, a confluence stent and a general venous stent. Other innovations might be at least as important as new stents. Artificial valves might become available in the future, which will make a huge difference to deep venous therapies. Other solutions might include flow-enhancing techniques that make endophlebectomy obsolete. The whole venous field will evolve rapidly so that therapy is going to be easier, faster, and more effective.
VDM: What is the importance of follow-up?
De Graaf: It’s quite easy to put in a stent, but to treat the whole patient is different and the treatment doesn’t stop after stenting. A thorough follow-up is essential. Before we discharge the patient, naturally we check patency, flow, and stent configuration. But the first office visit has to be planned within the first 2 weeks because in case of early blockage you can still perform thrombolysis with a reasonable chance of success. If you see the patient for the first time after 6 months, you have no idea when the thrombus occurred and if thrombolysis is a valuable option. At least in our experience, thrombolysis doesn’t work well after 2 weeks. In those cases you may need to place a stent-in-stent after balloon angioplasty, which increases the risk of complications. We believe that the cause of early blockage is most likely to be found in suboptimal inflow. Therefore, after successful thrombolysis the inflow is carefully evaluated again and in case of doubt we might perform additional endophlebectomy and AV fistula. Alternatively, a surgical thrombectomy can be performed. When the first 2 weeks pass without complication, we feel comfortable that a sudden blockage is unlikely and we see our patients after 6 weeks, 3 months, 6 months, and after that annually. In these follow-up episodes gradual lumen narrowing is possible. We then look for the most likely cause and try to solve it before the stent blocks completely.
VDM: So you’ve seen improvement in outcomes for patients over the years as technologies improve?
De Graaf: Absolutely. We just published our first results for the sinus-Venous stent. They were very good with a primary patency of 92% at 1 year. Because we did not do a comparative study, we cannot conclude that the specific stent design alone is responsible for this. However, we did not see any stent-related patency loss. This means no residual or recurrent compression, fractures, or other integrity loss. I believe we gained about 5% to 10% of the primary patency only based upon the stent. Nevertheless, some stents still occlude and we need a better understanding of why this happens. Naturally, anticoagulation plays a huge role and we only just started to scratch the surface on hemodynamics. Many variables are involved, of which stent design is one.
A remaining challenge is a severe outflow obstruction of both the femoral and profunda veins. If the major outflow vessels are obstructed, then the blood will try to get out of the leg through small branches in which you cannot place a stent. You might consider placing a stent in the CFV and performing an endophlebectomy to include as many inflow vessels as possible and place an AV fistula. It is then likely that the stent will stay open, but the patient will not undoubtedly benefit, because most blood from the leg might not reach the stent, rather it might find pre-existing collateral pathways. I have no solution for that yet.
In principle, it would be acceptable to tell some patients we can’t help them, but practically, I’m not always able to tell which patient is going to benefit fully and who will not. What we do know for sure however is that if we don’t treat at all, patients are not going to improve. Generally speaking, we give the patient an excellent chance of improving from an unacceptable status quo, while limiting the risk of worsening very close to nil. In the meantime we should seize every technical and medical opportunity to improve outcomes further. If optimal stent design can add just 5% to 10% flow, I’ll be more than happy to take that.
Editor’s note: Rick De Graaf, MD, is an interventional radiologist in the Department of Radiology at Maastricht University Medical Centre, Maastricht, The Netherlands. He reports consultancy to Optimed.