Dialysis Fistulas and Grafts - HD

Introduction

Good afternoon, I'm gonna be talking about dialysis fistula and graphs, and I don't have any relevant financial relationships, but I am an interventional radiologist doing IR around 90% of my time. And the remaining 10% is vascular ultrasound, it's gonna be somewhat of a clinically oriented talk.

Background

Briefly about the background. End stage renal disease, meaning people who need dialysis in order to survive. There's around 660,000 people in the US. The only cure is renal transplantation, and around 29% of all people get a renal transplant. But due to the shortage of organs and because of comorbidities which prevent someone from being a candidate for transplantation, the remainder needs some kind of dialysis, either as a bridge to transplantation or just to survive. And around 6% people get peritoneal dialysis, and the remaining 64% get hemodialysis either using a central venous catheter, arterial venous fistula, or a graft, which is what we're gonna talk about.

Fistulas and Grafts

Fistulas versus grafts. A fistula is just a communication between a radial artery to between the artery and the vein, which is artificially created. And the most common, and probably the best possible fistula is the radial artery to cephalic vein fistula, also known as a brassa simino or BC fistula. Then you can also have the brachial artery to cephalic vein. And in individuals who don't have a good satic vein, they will consider creating a transposed basi vein in individuals who cannot have a fistula for some reason, the graft is the next best choice, but it's really not been close to a fistula in terms of how useful or effective it is. And you can either have a straight or a loop graft, and it's usually between the brachial artery to the basilic, brachial or axillary vein. And there are certain novel kind of grafts. There's this core hybrid vascular graft where the, they just, it's like a stent graft on the venous side, and the surgeons just anastomosis the arterial side to the artery itself. And then there's this hero graft, which is really almost just out of clinical trials. It's not really mainstream yet.

Fistula First Initiative

The fistula first is a nationwide quality improvement initiative, which is essentially prodding all physicians to consider placing a fistula first. And the reason for that is fistulas last longer. They have fewer complications, specifically thrombotic complications. And currently around 60% of all people having hemodialysis have a fistula. The only slight advantage of a graft is that the grafts mature sooner, which is around two to three weeks, as opposed to two to three months for a fistula. That being said, when someone has chronic kidney disease stage three or higher, the nephrologists are constantly monitoring the patient's creatinine. And if they think someone's gonna need dialysis within the next few months, they will schedule the surgery in advance. That's what the fistula first breakthrough initiative is.

Role of Ultrasound

Coming to ultrasound, what is the role of ultrasound? And the answer is everywhere. Before creating a fistula, we want to do preoperative mapping. Then once the fistula is created, if it doesn't mature and it's not usable, then there's a role for ultrasound. If after maturing and they're able to use it for a while and it stops being that effective, then again, there's a role for ultrasound. And finally, there's complications of fistula and graft to all this. I'm going to add the fact that a lot of our interventional procedures are done with ultrasound. And I, myself, during an interventional procedure, use ultrasound a lot because it really reduces the radiation dose to myself as well as the patient.

General Principles

The general principles are the same. You want to do it in a supine position, use lots of warm if possible, ultrasound gel, five to 12 megahertz linear transducer. And it's important to remember to consciously avoid applying excessive pressure because these are very superficial structures. You can just press it very slightly and inadvertently and distort the structure and make things look like there's a stenosis or mismeasure the size of the vessel.

Preoperative Mapping

Coming to preoperative mapping for regard to preparation, if the patient's already on dialysis, we want to wait 24 hours after the last dialysis. Generally a tunic is a good thing to put in, but not everybody does it, which is just fine. What are we looking for? It's the arterial inflow and venous outflow with regard to the arterial inflow. If typically surgeons want to first place a fistula or a graft in the non-dominant hand or non-dominant arm, we are looking at the brachial and radial arteries, and if we are looking for wall calcification depth and diameter of the artery, and whether there's any major stenosis in certain centers, they also get in time medial thickness. And they do functional studies and look for reactive hyperemia, which is essentially just inflate a blood pressure cuff for a minute or two above systolic pressure and then deflate to see what happens to the resistive index within the artery. And they say or believe that if the RI really goes down, then this artery is capable of dynamically increasing its flow once the fistula graft is created. We don't routinely do it at my institution, but I just wanted to mention it for the sake of completion.

Venous Outflow

With regard to the venous outflow, typically they want to use either the cephalic or basilic vein, and we are looking for the depth, diameter and quality of the vein quality. As in, is there any region of focal wall thickening which might represent sequela of prior superficial thrombosis, or simply the fact that people have multiple venipunctures over a course of a life, you can have the vein getting a little damaged in that location. And then finally look for deepness thrombosis in the brachial auxiliary and subclavian veins.

Criteria

These are certain criteria which generally work for me. These are not consensus criteria, but these are the best criteria I can really mention in a tox such as this, which is artery wise, you want the artery to be more than two millimeters in diameter. It's never a problem within the brachial artery, but sometimes you will see radial arteries, which are smaller than that. We don't wanna see any major, moderate or severe stenosis, and we want to see a peak systolic velocity of 50 centimeters per second or above. If you're doing reactive hyperemia, we want the RI to be less than 0.7 regarding the veins. I think most people will accept that more than 2.5 to three millimeters in diameter is important. Then we look for the quality of vein, vein thickening, whether there's any calcification of collaterals and vein branches. I'm just going to add that these, the reports which I personally give out for preoperative mapping, I simply mention facts for what they are and don't really make any specific course of recommendation for the surgeon as to which arm or vein they should use, just because that annoys a lot of people.

Failure to Mature

Coming to failure to mature, the important thing is that this is a clinical diagnosis and it's set to have a 30 to 60% incidence. And the causes are, it could be because of the inflow artery because of a small caliber of stenosis, anastomotic stenosis, or venous outflow as in venous outflow stenosis, or the vein could simply be too deep. And if it's too deep, the flow might be great through the vein, they may not be simply able to palpate the vein in order to cannulate it. And finally, you can have a competing accessory vein or a collection.

Ultrasound Criteria

There are other ultrasound criteria available, especially in the radiology literature, but I personally highly recommend using this criteria, which is the K dokey, which stands for the Kidney Dialysis Outcomes Quality Initiative. They call it the Rule of Sixes. And this is from the National Kidney Foundation, which is the group which essentially sets standards for dialysis and nephrology countrywide. And their recommendation is more than or equal to six millimeter diameter of vein. The vein should be less than or equal to six millimeters in depth, and the flow volume should be more than 600 ml per minute. And part of the reason why I'm recommending this is that this is what the nephrologist truly believe in, and it's just simpler to go along with what they believe in. Unless you have specifically done a study at your institution where you, your criteria seem to be working better for your specific patient population, I simply recommend following this criteria. There are certain additional helpful criteria, which is you want to see at least a 10 centimeter straight length of the fistula vein because that's how much the dialysis cannulas really need to fit in, and you don't wanna see any major competing accessory vein.

Measuring Flow Volumes

I'm just going to briefly mention the technique for measuring flow volumes because there's a lot of confusion regarding this with regard to how people do it. The flow volume is the mean velocity times area, which is pi times diameter square by four, and we are multiplying it by 60 because we want to give it in ml per minute. And when I say mean velocity, we mean the time average mean velocity, which most modern machines with a vascular package, it should be built into the machine regardless of the manufacturer. Certain tips regarding the techniques, you want to do the study on a non-dialysis day. And unlike say, when you're measuring an aneurysm here, you want inheimer to inheimer distance because you want the lumen distance, the lumen diameter and not the vessel diameter. And we usually try to get velocity from around 50 to 70% of the lumen diameter. If it's AV fistula, you want to measure at least two to four centimeters away from the anastomosis in a relatively straight segment of the vein, because even in a normal person at the AV fistula, you will see some turbulent flow, which kind of messes up our numbers a bit. And I, at least at our institution, we routinely get a mean of three measurements. And the reason for that is more to serve as an internal quality check, because initially when they start doing it, you'll see there's such a wide variation of two or three times between one measurement to the other, and you're really not sure as to what exactly is going on. So when you get three measurements and those three more or less are in the same ballpark range, I think that just gives you a lot more confidence with regard to your own measurements. There is an alternative for AV grafts, which is, they say that the flow volume in the brachial artery is more than 90%. The flow of volume in the brachial artery alone constitutes more than 90% of the flow, is 90% of the flow goes through the AV graft and only 10% actually reaches the hand and fingers. So there are a lot of people who simply, if there's an AV graft, simply use the brachial artery measurements as their standard and just follow up using that.

Example of Normal Measurement

This is just an example of a normal person where we got some measurements. You can see that the doppler angle is well corrected. We have measured the distance from here to here, that comes out as dis here, and then that's the time average mean velocity. So that's the peak systolic velocity, and the machine automatically calculates somewhere within the spectral envelope what the mean velocity is, and that's the time average mean velocity. So it multiplies it using that formula and comes up with a number such as this, just 6 74 ccs per minute.

Competing Accessory Veins

Coming to competing accessory veins. These are more common with forearm fistulas. And generally speaking, these are considered significant when either it's a large vein, it's within 10 centimeters from the anastomosis, or if there's a sudden caliber in the outflow vein at this level. So if you have two large, I mean like a large competing vein coming out and say the fistula outflow vein changes from eight millimeters to six millimeters, that would be highly concerning that the competing accessory vein is really stealing blood away from the fistula and preventing the fistula from maturing. And another useful criteria is if the flow volume is decreased central to the competing accessory vein. Now, the one caveat to this is when you see a vein coming off the fistula, you must ask yourself whether it's, is it a truly competing vein or is it that the vein came up because there's a stenosis centrally? Because if there's a stenosis central or centrally meaning central or more closer to the heart within the outflow vein, then the blood simply has nowhere else to go and starts growing through a venous channel. And then that becomes a very large vein. So that's just something to consider when calling something a competing accessory vein.

Example of Competing Accessory Vein

This is an example of a 75-year-old man with an AV fistula. I've just not maturing for more than six months. And here you can see that's the fistula. Definitely the velocity within the outflow vein is increased. As you can see, there's some colored ular alia are seeing here, and the peak systolic velocity is more than 400 centimeters per second. And so this is the fistula somewhere here, and we are looking at the outflow vein here. And on color doppler, again, there's the turbulence within that fistula. And this is the outflow vein, which looks wide open, but there is this large vein which is arising from the outflow vein and going peripherally as you can see here. And there's a lot of flow within this peripheral vein. So this was one of my IR cases. And the orientation here is this is towards the shoulder and these are the fingers. And what we did is we punctured the outflow vein here, put in a sheath, and then came out with a small catheter all the way up into the brachial artery, and we inject the brachial artery. And you can see this is the brachial artery here, and then that's the fistula. There is a stenosis, which is not so well seen just because of all these collateral veins, but you can see that this is the outflow artery. So that's the radial artery starting off here. And this is the large competing vein we are seeing. So just to document this, we got our catheter into the competing vein and injected a lot of contrast here, and you can see there's a whole lot of blood going this way and then turning back and going again towards the heart. So this fistula is simply not going to mature until we both treat the stenosis as well as treat the competing vein. So in this is what we did. So we did a angioplasty and treated the stenosis, and in the past the treatment of choice used to be ligate the competing vein using surgery, as in we would do an ultrasound, put a mark on the skin, and the surgeons would just go ahead and ligate the vein. But at least for the last 10 years or so, that's really fallen out of favor. What we nowadays do is simply go in and put in a lot of embolization coils and coil off this competing vein. And you can see that once we inject here, there's nothing going in through the competing vein. Everything is coming out through fistula. So this is an example of a small vein, which is really not competing. Usually we leave these alone unless the fistula is simply not maturing further, in which case we'll consider embolizing this one as well.

Management

Failure to mature with regard to the management against osis or very small outflow vein, it's balloon angioplasty plus or minus stent, like I said, competing accessory vein. The treatment would be co embolization. And if the fistula vein is too deep, then that's surgical revision. So around 10% of cases no clear cause is found in my experience, it's usually people who are more than 75, 80 years of age or someone who has cardiac issues, no matter how long they have the fistula. And no matter how good everything is, the flow within the vein simply doesn't increase to a point that they're able to use it for dialysis. So those people will have to consider the fact of whether is it slowly maturing or not maturing at all. But most of these people will essentially be catheter dependent.

Complications

These are some of the complications we are going to cover. From a clinical standpoint, every time they do dialysis on somebody, they are doing surveillance as in clinical surveillance before and after dialysis. What are some of the things they're doing? Those include the dialysis nurses are feeling the thrill before they puncture the vein when they puncture the vein with a needle or they're seeing if there's any clots which are developing within the needle or within the dialysis tubing, which is going into the dialysis machine and back. The machine does calculate this, which is increased recirculation time and whether there's high dynamic out out venous pressures of more than 200 millimeters of mercury. And they also have a way of calculating something called KT and v KT by v. That exact technical details don't matter, but this just suggests that the UIA clearance is not adequate if the KIB and BI v is out of bounds. So usually all these symptoms are associated with outflow vein stenosis. Now, if another group of symptoms is people who have prolonged bleeding after the axis needle removal, again, the causes are outflow vein stenosis, someone being on anticoagulation or there being a small pseu aneurysm. And finally you can have arm swelling, which is almost always associated with central venous stenosis. That's really how it presents itself.

Stenosis

Coming to stenosis, we usually assume that more than 50% diameter reduction is significant. And there is this technical term of what is a juxta anastomotic stenosis, which is within two to five centimeters from the anastomosis. That's very helpful to tell the referring physician as to if a stenosis is juxta anastomotic because that really changes their approach as to how they would go and treat it, as we'll see very shortly. But again, 90% of the osis occur either at the arterial venous anastomosis or in the venous outflow. And the reason for that is that the veins are simply not designed to experience the high pressures which flow within the arteries. And when you create this artificial circuit between the artery and the vein, those veins wall hypertrophies or the vein valves hypertrophy. And that's really what causes a lot of these stenosis. The important thing about stenosis is in addition to inadequate dialysis, they will eventually lead to thrombosis of the graft and 90% of thrombosis, a arteriovenous fistulas or grafts will have some person with stenosis. Now, it's always important to mention whenever you see a stenosis as to the distance from the arteriovenous anastomosis to the site of stenosis, or if it's very centrally, as in let's say someone has a radial artery to cephalic vein fistula, and you see a stenosis within the mid arm, then it's a good idea to mention the distance from say, a bony alarm mac, like the medial ca medial, con medial epicondial of the humerus to the point of stenosis. Because again, like I said, it helps change the interventional approach as to whether to access it from an antegrade approach or a retrograde approach.

Stenosis Criteria

Coming to stenosis criteria, there's no consensus criteria, but in gray scale, if you have a narrowing with the lumen less than three millimeters, that's concerning. Likewise, peak systolic velocity, more than 400 or PSV ratio of more than two. And just like the K dokey guidelines, if the estimated flow is less than 600 ml per minute within the outflow vein, that would be concerning. Or if there's a very high grade stenosis, the velocities can reduce throughout the outflow vein and cause a peak systolic velocity reduction of less than a hundred. And finally, more than 25% flow reduction of the volume from the baseline.

Examples of Stenosis

This is the patient with the brachial artery to basilic vein fistula, and they had difficulty cannulation. And as you can see here, that lumen appears very narrowed and the peak systolic velocity is more than 500 centimeters per second with color doppler liasing as well as spectral broadening. And in the mid humerus, the velocities are more reasonable. But again, this is a high grade stenosis and this is again my case and we just accessed it retrograde. So we are going towards the fistula, that's towards the fingers. And on injecting contrast, you can see there's hardly any contrast getting across because of the very high grade stenosis here. And then this is one of these play examples where I feel ultrasound is very helpful, reducing my own fluoroscope, my own radiation dose. You can see that we just, we do a lot of our balloon angioplasty just using ultrasound guidance. So here you can see that the balloon is getting partially inflated and there's a waste. And on complete inflation it's completely opened up. And again, this is the same thing with fluoroscopy just for comparison. There's another patient with a brachial artery to baic vein AV fistula with increased pressures. As you can see here in the mid humerus level, the pressure and the velocities are normal. And here obviously there's a narrowing within the vein. And color doppler really is seeing spectral broadening, again, high grade stenosis. And as we go more centrally, the velocities have reduced to less than a hundred centimeters per second, which again makes sense. This is the patient's angiogram. You can see that high grade focal stenosis, they, this was one of my colleagues' cases and they first treated with a six millimeter followed by a nine millimeter balloon. And you have a very good end result.

Thrombosis

So now we just see a lot of images of other complications. So this is a patient who had a loss of thrill and as you can see here in the first image itself, in the left brachial artery, in the pre anastomotic segment, there's very high resistance wave forms with reversal of flow. So right off the bat, you know, either the graft is completely occluded or there's a very high grade stenosis somewhere, because normally you would expect to see lower resistance flow right here. And as you can see here, there's obvious thrombosis within the graft and there's just some more images showing this echogenic material within the bypass graft.

Arterial Steal

I just wanna talk about arterial steel. I think this is a very important topic. It occurs around five to 8% of patients. They can have pain, numbness, tingling, and ulceration. And many people will give us the history that the pain and numbness increases when they do dialysis. And but at normal times they're fine. And some of the risk factors are diabetes, women, advanced age and peripheral arterial disease. So just like steel anywhere else in the body, you can see either complete reversal of flow or bidirectional flow within the post anastomotic segment of the artery. And one more very useful test is to get physiological information using digital systolic blood pressures. And if it's less than 60 millimeters of mercury, that would be concerning. And we can try some provocate maneuvers as in essentially just compress the outflow vein with our fingers and see what's happening to the pressures and to the flow within the on color doppler because that will give us a very good sense of if the fistula were to be ligated or if drill surgery were to be performed, whether it's actually going to improve the patient's condition or not, or whether there's some other cause. Which comes to what exactly is drill which. So this stands for distill revascularization and interval ligation. And so this is a patient with arterial steel head end hand end and you have flow going this way and this way what they do in the drill procedure is they ligate the artery here and they create an end-to-end anastomosis or sorry site to end anastomosis between the bra. Usually it's the brachial artery to the artery distill to the place where you're ligating. The important thing is they need at least seven centimeters distance between this point to this point. And what happens is when you have that distance, preferably some amount of flow goes like this into the hand and still a lot of flow goes in through the fistula and the fistula is still usable. So that's a very good salvage procedure instead of completely ligating the fistula and basically having someone lose access. And then the ligation here is done immediately still to the fistula.

Examples of Arterial Steal

There's a 58-year-old man with pain and swelling and you can see that's the pre anastomotic segment of the artery. Very good flow post anastomotic. There's clearly reversal of flow 'cause it's blue and blue here is reversal. And again, post anastomotic radial artery, there's reversal of flow. And this is the fistula vein as well as the outflow vein. It's a different patient. I just want to show examples of what it looks like on an angiogram. So here we have gone from a retrograde axis and injected the artery from the pre anastomotic segment and you can see all the blood is going like this and going out through the fistula and really no blood is going across the post anastomotic segment. So everything here is getting fed through via the NAR artery. And what we did is we just advanced the, pulled the catheter back and advanced it here. And when we inject this, you can see still that there is some flow going forward thanks to the speed or pressure of our injection, but still there's a lot of flow going backwards into the thing which should simply not be happening through fistula. So this is a 71-year-old man with a brachial artery to basilic pain, AV fistula and had pain. You can see what looks like almost the presti type of waveforms that we see distilled to the AV F. And this is a very good example of how this test is useful. So this is a plethysmography and the right side is a normal side for comparison. And this is the left side. And here what you're seeing is very nice waveforms within bilaterally in the upper arm at the level of the forearm, you're barely getting any pressures. This is almost looks like artifact as opposed to the other side. And under at the finger level you're simply not getting any flow, what I mean any detectable flow whatsoever as opposed to the right side. So we can also do individual digital pressures at each of the digits. And again, you can see left right side looks very normal with lot of nice pressures in the one fifties and one forties, whereas on the left side there's no detectable flow whatsoever. And we did the provocative manual, which is essentially just manually compress the fistula with our fingers and then just repeat this test. And you can see that when you compress this fistula manually, the waveforms here on the left side are maybe not as good as the right side, but definitely they're markedly improved. So this is a person who's actually going to benefit from a drill surgery and we clearly know that this is the cause and by fixing the fistula it's actually going to improve the patient's condition.

Other Complications

Briefly about certain other complications. So pseu aneurysms, you can see many times pseudonyms which are five millimeters smaller, usually caused by repeated cannulation. They're not treated unless they're very symptomatic. You can have large pseu aneurysms. So they are usually treated with either thrombin injection or placement of a covered stent, occasionally surgical revision. So this is a patient with an AV graft having prolonged bleeding. So in fact this was the graft portion and that's the pseudo aneurysm and it's a very wide neck pseudonym, but you can see that there is that yin yang appearance. The pulse stopper images weren't particularly helpful here, which is why I'm not showing it to you. But this was one of our, one of my colleagues' cases where, so it was a loop graft. They came in like this and injected contrast and you can see the pseudonym fill up. And the interesting thing is because pseudonyms have slow flow, the pacified blood is coming in, washing out flow from the graft, but you can see the pseudonym is still pacifying with contrast. And then this was just treated with a covered stent. So then you can also have aneurysms, which is just dilatation of the vein. And unless the patient's overtly symptomatic or there's a lot of flow, usually they don't treat it, they just let it run and they just keep using the fistula. And you can see here there's a very large aneurysm, which is around four centimeters in diameter in the outflow vein. And then there's a very large vein and you can see just like a regular aneurysm, you see this hazar type of flow within the aneurysm here. And as we go centrally into a normal portion of the vein, you have very good flows. Finally you can have collections like this. So there's just an AV fistula showing this collection with some fine internal echoes. Nothing needed to be done because the fistula was working just fine. There was another patient who had a postoperative AV graft and a collection had pain and swelling. So you can see obviously there's this collection with course internal echos and the patient developed a white count. So they got the CT scan again, which shows this collection with the graft growing through it. And a couple of days later she developed a fever as well. And at that point you can start to see foci of air and we did an aspiration and got a little bit of pus, so this was surgically excised.

Personal Opinions

I'm just gonna end up with this brief personal opinions. And this is by no means the definitive, but I just believe in this. One with regard to terminology, I think very mainstream. Everybody uses terms such as pre and post anastomotic for the artery, which is very good because it means just one thing with the veins, just like the superficial veins, there's a lot of confusion when you use terms such as proximal and distal. I highly recommend you use terms like central and peripheral, where peripheral means closer to the fingers and central means closer to the heart because then it just means one thing. Next thing is there is some recent evidence that you can do an ultrasound two weeks post fistula creation and that can help us early detect failure to mature. And I'm more of the maybe group and not a yes group as to whether this makes sense or not. And the reason for that is as a clinician, if they were to tell me that this is not maturing two weeks out, it really wouldn't change my management personally. I would say still wait for a few more weeks. If it's not maturing a three months, then we'll do something about it. And then finally, is there any role for routine ultrasound surveillance in a well-functioning AV fistula or graft? And so in the European literature you will see some people say yes and they routinely do graft surveillance. In the US literature it's not really that much out. And part of the reason is because Medicare does not reimburse this at the current time. I just want to add here that they are doing surveillance all the time at dialysis. The dialysis nurses are looking at everything clinically. So unless the graft is simply not, unless if the graft is working perfectly well, I really don't see a reason to do an ultrasound surveillance. So these are in the handouts, which you have. Alright.