Vascular Access Site Hemostasis: "An Endovascular Surgeon's Perspective"Manual Compression May Not Be Benign Part 1


1David E. Allie, MD with 2Chris J. Herbert, RT, RCIS, and 2Craig M. Walker, MD

When endovascular techniques came about almost two decades ago, I was just finishing my training, and I knew I wanted to be involved, so I went to the cath lab. I still spend a great deal of time there, and have for more than 15 years. There is a soft spot in my heart for the cath lab professional. Surgeons know how important it is to have a good surgical tech, and when I went into the cath lab, I wasn’t trained only by cardiologists, but by several very good and seasoned cath lab technologists, including Chris Hebert and Gary Chaisson, who have over 40 years of experience combined. They actually trained me on the techniques I use today. When I do cases today, I don’t necessarily want an interventionalist around, but I do want my top cath lab technologists. I also take them into the OR when we do combined endovascular procedures. Without question, cath lab techs have had an important impact on my career, and it has always had a positive bearing on our patient outcomes.

As a surgeon, I became very interested in access site issues because I was increasingly seeing problems. Over the last eight or nine years, it’s become an issue on which I’ve concentrated. I’m going to give you a brief overview of our work and hope to convince you that our gold standard, manual compression, may not be benign.

Goes with the Territory?

The field of cardiothoracic surgery has been developing less-invasive techniques over the last two decades to decrease complications (Figure 1). Less-invasive sternal and leg incisions along with radial harvests and off-pump procedures have all been possible with technological advances, with a resulting decrease in complications. Even though rapid advances were being made in the cath lab with balloons, wires and stents, I saw little focus or resources dedicated to improving femoral access site management.

As a surgeon going into the cath lab, I couldn't understand what in the world was going on there, because when I began doing procedures, I would see our patients back in the office with groin problems and hematomas. I would go to the cardiologists and say, Hey, what’s happening here? They would shrug and respond, Well, it’s part of the territory.

If you look at the groin in Figure 1, it very well could be Sellinger’s original groin, when he first described manual compression in 1952.1 It was the gold standard then and unfortunately it’s still the gold standard today, over 50 years later. We’ve put men on the moon in the past 50 years. Why in the world do we still have today, truthfully, such a gross procedure as manual compression as our gold standard for vascular access management? According to the literature, manual compression is our gold standard.2 When there are vascular closure device trials, the gold standard or control is still manual compression. You can even go to the FDA, and in their eyes, it’s still considered the gold standard by which they measure VCD trials.

Yet look at a few results of our gold standard in Figure 2. This happens more frequently than you know. Perhaps you don’t see your patients after they leave your area and they go to the floors, but these are the things we see with patients back in the office and they are not benign. Vascular access control is a very real and big problem for patients.

We went to the literature a few years ago to figure out what was going on, but found that the literature is somewhat muddled. There may be a little more data coming out now, but a few years ago there was absolutely nothing. There are no large-scale, randomized trials between the vascular closure devices. We really don’t know which device is better and which one is not. There is also no specific standardization for reporting complications. Is it minor? Is it major? Is it big? Is it small? Is it ecchymosis? There is no standard of reporting between facilities. We believe most access site complications may even go unreported, which is a much bigger problem than you might suspect. If you look for evidence and data in the literature, we just don’t have it. Incidence of access site complications are reported as anywhere from 0.5% to as high as 27%.3-5

Clinical Costs, Not Just Economic

Does this mean 27% of patients are going to the OR to get femoral artery repairs? No, but they may have moderate hematomas, with a resulting clinical and financial expense, and we still think many hematomas are unreported. Early data from Lauer et al6 came out in 1999, trying to tell us what a patient with a hematoma may cost. If somebody has one unit of blood transfusion because of a hematoma, that brings $8,000 in costs. That doesn’t mean that a unit of blood costs $8,000, but that is because the patient stays hospitalized overnight and needs to have IV tubing, CT scans, ultrasounds, and so on.

A subanalysis of the bivalirudin REPLACE-II study showed that there are not only economic but also clinical costs to access site complications.7 Bivalirudin showed an improvement in bleeding complications, but when complications did occur, investigators followed those patients out to one year. More of these bleeding complication patients had MIs, more had second stents, and more had secondary events, so the MACE rate was increased in patients post PCI who had a bleeding complication versus those who did not. We now know that a patient has a 12x greater risk of dying after a PCI within a year if they had bleeding complications, and they are 4x more likely to have an MI.8 We believe this may also be true even for a minor hematoma.

Correct Femoral Access is Crucial

It’s very important that we learn, first of all, how to stick the patients. It all starts with the appropriate stick. Hopefully physicians are now starting to look at the stick as important and are really trying to hit the common femoral artery. It’s important to be somewhere under the iliac rim and above the bifurcation, near the medial head of the femoral bone (Figure 3, page 6). Figure 4 (page 6) shows what it looks like anatomically in the OR.

Figure 5 shows a significant complication after renal artery stenting. It was a late secondary bleed and we evacuated the hematoma. The patient lived and is doing fine. Two years prior, we had done an aortic arch replacement, and we replaced the aortic valve and reimplanted her coronary arteries. She was out of the hospital in five days and we gave her two units of blood. After this renal procedure, the patient was in the hospital for a month and she received 9 units of blood.

Figure 6 shows a hematoma we evacuated about two weeks after a brachial approach. When you have these catastrophic complications, they have a very large impact. We fixed the arm and the patient is doing fine, but the point is that we need good, secure ways to deal with these vessels, whether they be femoral, brachial, etc. When hematomas occur, they can be a real problem.

Remember most of our patients are now on Plavix and aspirin, therefore it shouldn’t be surprising that we’re seeing more bleeding complications and especially later bleeds. The desirable antiplatelet effects on stenting can be a negative for vascular access management and further prompts questions about our gold standard of only 20 minutes of manual compression.

What Happens to the Femoral Artery?

Over the last several years, when endovascular aneurysm stent grafting started, we were able to spend a lot of time looking at femoral arteries in the OR. We would make small cut downs on the femoral artery and fix the aneurysm. It gave us a unique opportunity to actually look at the femoral artery after various means of vascular closure.

We did approximately 200 aneurysms over approximately a 2-year period of time. All patients had to have angiography beforehand. Some of these patients had manual compression and some had vascular closure devices as femoral access control. Figure 7 shows what the normal femoral artery looks like, which is nice and plump. Figure 8 is what the femoral artery looks like two weeks after manual compression. We deal with this every day. This is not considered a complication; it is just what happens after manual compression. Have you ever stopped and thought about what happens to that artery? What happens to that 7 or 8 French hole, or maybe 9 French, if you just hold under pressure? It doesn’t just go away. The body has to create scar tissue on the vessel in order to close it, which could have a real negative impact on our vessels, an impact of which we are unaware.

Figure 9 shows the arteries we dissect for AAA access and repair. We thought this was a normal groin result 2 weeks after angioplasty, but when we go down, we see significant scar tissue encasing the vessels. You’ve probably heard your surgeons use some inflammatory words about your cardiologist if the surgeon goes in afterwards and finds a bleeding complication or a lot of scarring in the groin.

Figure 10 is a patient about two weeks after a 5 French diagnostic angiogram. We’re going to do his aneurysm repair. You can see early scar tissue, which is the tissue that forms when you’re holding for manual compression. When you bring the patient back a second or third time for an angiogram, your physician or whoever is sticking might say, Boy, that’s a scarred groin. Manual compression heals by scarring, and scarring is not benign. Manual compression, in itself, is causing more problems than we really know. In addition, since our patients all go home on Plavix, we’re not really receiving the same type of healing in these groins that we did years ago, because our patients are much more anticoagulated.

Figure 11 is a patient 3 weeks after a 6 French diagnostic, showing a tissue patch. This figure shows what we characteristically see, and it is what manual compression is doing this is our gold standard. Notice the little blister, which is actually where the sheath was inserted. We find it hard to believe that this is benign to the artery and to the patient.

What About Secondary Access?

In your labs, you probably have physicians who come back and stick groins the next day. It seems like that is acceptable, because there are no guidelines that dictate otherwise. Yet there is no consensus on when you can safely re-access patients after any of these procedures. We feel re-access early on is going to be more risky.

Figure 12 shows what these groins look like just one week post procedure. This is a very interesting case, with a completely normal 5 French diagnostic angiogram. You can see it’s in a nice area. Same thing - here’s the patch. Now we bring down the little patch - it’s what we call this. Again, this is the connective tissue patch of the vessel itself. Here you can see what’s described as the platelet plug. Platelets must accumulate, you must have pressure, you must have fibrosis, and there you have the healing process. You’re holding pressure sometimes less than 30 minutes. Are we really to expect good, secure closure with manual compression? Cath labs don’t see these patients coming back with late bleeds, but the surgeons do. Manual compression, our gold standard, may not be so benign.

Small Vessels Can Predict Complications

Access site complications also occur more often in small weight patients with small vessels, like females, diabetics, the anticoagulated, and PVD patients. Large coronary studies tell us that these are the patients who are going to have a higher risk of access site complications, so we can identify beforehand which may be more likely to have a bleeding complication. Chronic renal disease, diabetes, and PVD patients are now known to have dysfunctional platelets which may further increase the risks of bleeding complications. If we are to use the large PCI trials to identify patients at risk for bleeding, those risk factors are multiplied in PVD patients, making this a very high-risk population for bleeding complications.

Figure 13 is a patient who also had a small hematoma, but did not have an exploration; we just observed the hematoma. She went home and we brought her back for aneurysm work. To us, it looks like a little cherry red bubble that can pop off at any time. Does this look benign? Is this safe to reaccess? We think NOT!

The Ideal Vascular Closure Device (VCD)

What we really need is an ideal vascular closure device (VCD). We don’t have one at present, and let us also say that we respect all of the closure technologies. The whole field of closure technology, as well as the resources from the companies developing it, is really a very recent occurrence. By contrast, companies have spent billions of dollars on coronary stents over the last two decades. By talking about this issue, we hope to stimulate companies to put those kind of resources into access site closure so we can achieve an ideal closure device.

Ideal Device Characteristics

What would the ideal device be? Obviously it would be very simple and safe immediately secure, immediately effective, very reliable, reproducible, and it would allow you to anticoagulate the patient. Wouldn’t you like to have some of your high-risk-for-ischemia patients on IIb/IIIas, which may help on their ischemic outcomes? The device should allow you to have immediate re-access and be applicable to every patient across the board. (Currently, only between 20-25% of patients worldwide receive a vascular closure device.) The ideal device should be free of complications and also cost effective. Ideally, the device would be totally extraluminal, truly inert and biocompatible, be delivered completely sterile to the vessel wall (like a stent, to decrease infection), and allow for use in PVD, small vessels, and in high and low sticks. The device should allow use in larger sheath-based technologies (> 7-8 French), interventions, and fails safely (avoiding catastrophic complications). Now, is that a little unrealistic? Maybe so, but this list of characteristics is what we ought to aim for.

If we had the ideal closure device, it would really change the way we think. Initially, closure devices were developed for what we would call convenience. There isn’t anything wrong with convenience for the patient or the hospital, but if you had a way to close the femoral arteriotomy in the ideal fashion, it would allow the physician to formulate a whole different game plan for the patient, like keeping him on coumadin, sending him home the same day, planning to use larger sheaths, etc. The ideal VCD could change the primary goal of VCD vascular access management from one of primary convenience to the theme of primary treatment. The ideal VCD could be an important part of the treatment planning and strategy for each patient. When we have the ideal device, there would be no reason not to use it on every single patient who has a diagnostic or interventional procedure in every lab worldwide.

Current Devices Need to Evolve

There are significant limitations with the current devices. These limitations are such that 70-80% of patients still do not really receive a VCD. It varies from lab to lab. Some labs will try to close everybody and some won’t even try to close anybody because of catastrophic experiences in the past. VCDs are not complication-free. In our cath lab, many of our lab techs are the best users of our VCDs, but still the devices are complex and there can be a long learning curve.

You can’t really use the devices in small vessels because some of them have endoluminal components. A high stick, low stick, or significant PVD are all limitations. Several of the devices have components inside the lumen, which we think is a problem. Our senior cardiologist actually had a PCI. He got fixed up and had a closure device placed with an endoluminal component, but then couldn’t walk for a month.

Suture-mediated device complications. The devices we’re putting inside the patient with permanent sutures, maybe even some of the collagens, are not inert. The vessels are very prone to scarring and reaction. You can have infection, and when you have infection, it is a devastating complication. Certainly, the clinical and economic costs for treatment are high. We’re not trying to tell you that we have a high amount of catastrophic complications. Yet if any of you have seen one, you know that when your patient has an infection or a vessel thrombosis, the patient goes to the OR and it is truly catastrophic to that patient.

Figure 14 shows a suture device we used on a patient. We closed the arteriotomy site in this patient two weeks before the aneurysm and the result looks excellent. This is what we would like to happen on every patient, but the fact is, it really doesn’t. Suture-mediated devices are not benign either. Figure 15 shows a different patient, with significant scarring. Part of the problem is that we don’t really know where that knot goes every time. We assume that it goes down to the vessel wall, but in many cases it’s not as accurate as we would like to think.

Here are some potential problems you can run into. Figure 16 is what we call a ghost knot. Figure 17 is a patient who had a very high stick. The knot really never got down to the vessel on that patient, and they had a massive retroperitoneal hematoma. This is one of the catastrophic complications. When this happens, it will almost shut a whole lab down with VCD use because it’s such a devastating thing to the patient. The physicians don’t want to fool with the device anymore, and that is a shame, because we actually are a big advocate of VCDs. We’re raising the issue of limitations, but again, it’s because we want to help try and create that ideal device.

Figure 18 was an abscess. We repaired it, but it was very complex to repair, and the wound broke down. Figure 19 shows another patient who had a vascular closure device which was put in out of state. He then came into our state with a small hematoma (he traveled a little early). Another physician explored and took care of the hematoma. He didn’t even know a VCD had been used, and it got secondarily infected.

In Figure 20, we’re dissecting down and it actually shows an abscess on top of the femoral artery, where the VCD had been placed. Figure 21 shows the suture that is infected. What happens is the blood vessel wall becomes infected, so you have to remove this vessel and do a bypass. It is a major procedure with a big impact on this patient.

This is why we’re saying that when you have complications, they are potentially catastrophic. We removed the diseased and infected vessel and performed a bypass, and the patient’s doing fine. But again, we want to emphasize that it had a big impact. There are complications that are going to occur with all of the different devices. It can even occur with manual compression alone.

Collagen-based device complications. What about some of the collagen-based devices? Figure 22 shows a case of a young patient with a normal cardiac cath who went home but came back one week later with a cold leg. She had an acute thrombosis or femoral artery occlusion, which is something you want to try to avoid.

We cut down to the artery in Figure 23, and the collagen wad is exposed. You can see the suture and then the endoluminal component to this particular device. We remove the patch, but it is not the normal tissue patch; this is the collagen patch of this device. We open the artery and you can see black material, which is thrombus, which is what it looks like when it occurs acutely. You can also see the so-called endoluminal feet component of this device. We go ahead and remove it.

We believe it’s dangerous to put this kind of a device in small vessels. In certain smaller vessels, it’s going to be more of a problem. So what did we do in the case above? We resect that vessel. We brought in a vein graft, and we did a bypass. The patient is doing fine, but it had a significant impact upon the patient.

A Final Word

In this article, it was not our goal simply to come down on current vascular closure devices. The fact is, in the bigger picture, they are all first-generation devices. We need more time, we need more money, and we need more resources to develop better VCDs. Imagine if we had just hung our hat on the first coronary stents that were available. You may remember the old Gianturco-Rubin stents. We’re not coming down on those stents either, but if we believed that those first couple of stents were ideal, then none of us would be working in the cath lab right now. It’s the same thing with vascular closure devices. It’s going to take a little more time. Hopefully, when the drug-eluting stent frenzy that’s taken the mind of all of these large companies off the vascular closure devices settles down, they are going to have the resources allotted to allow us to really do what we want with these devices: develop the ideal VCD.

We believe we need to undergo the same evolutionary process with vascular closure devices as we have undergone with coronary stents. In the future, we’re going to be using a VCD on most of our patients, just like stents. It’s a lofty goal, but very achievable if we’re able to develop this ideal VCD.

1. Seldinger SI. Catheter Placement of the Needle in Percutaneous Arteriography, <i>Acta Radiol </i>1952; 39:366-376.2. Baim D, Knopf W, Hinohara T, et al. Suture-mediated closure of the femoral access site after cardiac catheterization: Results of the Suture To Ambulate aNd Discharge (STAND I and STAND II) trials. <i>Am J Cardiol</i> 2000;85:864-869. 3. Dangas G, Mehran R, Kokolis S, et al. Vascular complications after percutaneous coronary interventions following hemostasis with manual compression versus arteriotomy closure devices. <i>J Am Coll Cardiol</i> 2001 Sep;38(3):638-641. 4. Kussmaul WG 3rd, Buchbinder M, Whitlow PL, et al. Rapid arterial hemostasis and decreased access site complications after cardiac catheterization and angioplasty: results of a randomized trial of a novel hemostatic device. <i>J Am Coll Cardiol</i> 1995 Jun;25(7):1685-1692. 5. Pracyk JB, Wall TC, Longabaugh JP, et al. A randomized trial of vascular hemostasis techniques to reduce femoral vascular complications after coronary intervention. <i>Am J Cardiol</i> 1998 ;81(8):970-976. 6. Lauer M, Karweit J, Cascade E, et al. Percutaneous coronary interventions in the United States: Changing practice patterns and outcomes. <i>J Am Coll Cardiol</i> 1999;33:36A 7. Feit F. Bivalirudin Reduces Ischemic and Hemorrhagic Complications of Percutaneous Coronary Intervention: Pooled Data From 10 Prospective Studies in 6,134 Patients, an oral abstract presentation (267308) at the 2003 American College of Cardiology meeting (in press). 8. Ahmed WH, Piana RN, Bittl JA. Does bleeding cause ischemic events after coronary angioplasty? <i>Circulation</i> 1996;94:I-376.

Add new comment